effects of exogenous hyaluronic acid and serum on matrix ... · effects of exogenous hyaluronic...

9
THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 268, No. 27, Issue of September 25, pp. 20473-20481,1993 Printed in U. S. A. Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte Complex* (Received for publication, April 14, 1993) Antonella Camaioni,Vincent C. Hascall$, Masaki Yanagishita, and Antonietta SalustriQ From the Bone Research Branch, National Institute of Dental Research, National Institutes of Health, Bethesda, Maryland 20892 and the §Department of Sanita Pubblica e Biologia Celluhre, Faculty of Medicine, Second University of Rome, 001 73 Rome, Italy Compact cumulus cell-oocyte complexes (COCs) iso- lated from preovulatory mouse follicles undergo ex- pansion in vitro when high levels of hyaluronic acid (HA) are synthesized and organized into an extracel- lular matrix. We studied the effects of fetal bovine serum (FBS) and of exogenous HA and HA-oligomers on the expansion process. Maximum retention ofHA in the COC matrix, and hence complete COC expansion, occurs when 1% FBS is continuously present during the first 18 h of culture. Irrespective of the culture time, HA synthesized when serum is absent is primarily in the medium, whereas HA synthesized when serum is present is primarily in the cell matrix. These find- ings support the hypothesis that the serum factor, iden- tified as an inter-a-trypsin inhibitor by Chen et al. (Chen, L., Mao, S. J., and Larsen, W. J. (1992) J. Biol. Chem. 267, 12380-12386), is a structural component of the matrix. Addition of exogenous HA or of HA oligomers of decasaccharide size (GlcUA-GlcNAc)S or larger effectively displaces endogenously synthesized HA from the matrix into the medium, thereby prevent- ing COC expansion. Addition of exogenous chondroitin sulfate affects neither matrix organization nor COC expansion, thus indicating specificity of the binding of some structural component(s) to HA. Fully expanded COCs disassemble when cultured longer than 18 h, a process which occurs also in vivo and which correlates with loss of oocyte fertilizability both in vivo and in vitro. This process involves release of macromolecular HA from the matrix into the medium, with loss of 50% of the HA in the first 8 h of incubation after full expansion. The release is not facilitated when HA oli- gomers, long enough to prevent matrix formation, are added to the culture medium after the COCs are fully expanded. This suggests that cooperative binding to HA of either the serum factor, an endogenously syn- thesized factor(s), or both is required to stabilize the fully expandedCOC matrix. *This work was supported by grants from Minister0 dell’universita e della Ricerca Scientifica e Tecnologica (40 and 60%) and by Consiglio Nazionale delle Ricerche Grants 91.00320.CT04 and FATMA 91.00245.PF41 to (A. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accord- ance with 18 U.S.C. Section 1734 solelyto indicate this fact. $ To whom correspondence should be addressed Bone Research Branch, Bldg. 30, Rm. 106, National Institute of Dental Research, NIH, 9000 Rockville Pike, Bethesda, MD 20892. The cumulus cell-oocyte complex (COC)’ is astructural unit of the antral mammalian follicle which consists of several layers of compacted follicular cells (cumulus cells) closely surrounding the oocyte (1). As a consequence of the endoge- nous gonadotropin surge during the preovulatory period, the intercellular space between cumulus cells increases, due to the production of a mucoelastic extracellular matrix rich in hya- luronic acid (HA) (2-5). This process, called expansion or mucification, facilitatesdetachment of the COC from the folliclewall, its extrusion at ovulation, and its capture by oviductal fimbria (6, 7). Moreover, the cumulus cells and the matrix in the expanded COC improve the success of sperm penetration and fertilization (7-11). Three components are required to induce COC expansion in uitro: 1) a soluble factor produced by the oocyte (12, 13), 2) follicle stimulating hormone (FSH) or CAMPanalogues (2, 14-16), and 3) a serum factor (14, 16-18). The first two components are necessary to induce HA synthesis by cumulus cells, whereas the third is important for organizing HA within the extracellular matrix. When compact, freshly isolated mouse COCs are cultured with FSH in the absence of fetal bovine serum (FBS), the mucoelastic matrix does not form. In this culture condition, most of the cumulus cells lose contact with each other and settle individually on the culture dish. On the other hand, total HA synthesis by FSH-stimulated COCs cultured either in the presence or absence of FBS remains unaltered, but its distribution between the COC extracellular matrix and the culture medium is distinctly different (16,17). In the presence of FBS, -80% of the synthesized HA accumulates in the matrix, whereas in the absence of serum, the same proportion is released into themedium. In either case, the molecular size of the HA synthesized is >2 million daltons, suggesting that the failure in matrix formation when FBS is absent does not depend on changes in the physical properties of this polymer (16). Cultures of mouse mural granulosa cells also exhibit the same serum dependence for organizing HA within the extra- cellular matrix (19). The serum factor has been recently identified as a protein belonging to the inter-a-trypsin inhibitor family ( I d ) (18). Purified IaI at 1 mg/ml was able to replace serum in support- ing COC expansion, and itwas proposed that IaIprovides an integral structural component of the matrix, perhaps by in- The abbreviations used are: COC, cumulus cell-oocyte complex; HA, hyaluronic acid; FSH, follicle-stimulating hormone; FBS, fetal bovine serum; I d , inter-a-trypsin inhibitor; BSA, bovine serum al- bumin; hCG, human chorionic gonadotropin; MEM, Eagle’s mini- mum essential medium; CHAPS, 3-[(3-cholamidopropyl)dimethyl- ammoniol-1-propanesulfonate; CS, chondroitin sulfate; HPLC, high pressure liquid chromatography. 20473

Upload: duongdat

Post on 25-Feb-2019

218 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 268, No. 27, Issue of September 25, pp. 20473-20481,1993 Printed in U. S. A.

Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte Complex*

(Received for publication, April 14, 1993)

Antonella Camaioni, Vincent C. Hascall$, Masaki Yanagishita, and Antonietta SalustriQ From the Bone Research Branch, National Institute of Dental Research, National Institutes of Health, Bethesda, Maryland 20892 and the §Department of Sanita Pubblica e Biologia Celluhre, Faculty of Medicine, Second University of Rome, 001 73 Rome, Italy

Compact cumulus cell-oocyte complexes (COCs) iso- lated from preovulatory mouse follicles undergo ex- pansion in vitro when high levels of hyaluronic acid (HA) are synthesized and organized into an extracel- lular matrix. We studied the effects of fetal bovine serum (FBS) and of exogenous HA and HA-oligomers on the expansion process. Maximum retention of HA in the COC matrix, and hence complete COC expansion, occurs when 1% FBS is continuously present during the first 18 h of culture. Irrespective of the culture time, HA synthesized when serum is absent is primarily in the medium, whereas HA synthesized when serum is present is primarily in the cell matrix. These find- ings support the hypothesis that the serum factor, iden- tified as an inter-a-trypsin inhibitor by Chen et al. (Chen, L., Mao, S . J., and Larsen, W. J. (1992) J. Biol. Chem. 267, 12380-12386), is a structural component of the matrix. Addition of exogenous HA or of HA oligomers of decasaccharide size (GlcUA-GlcNAc)S or larger effectively displaces endogenously synthesized HA from the matrix into the medium, thereby prevent- ing COC expansion. Addition of exogenous chondroitin sulfate affects neither matrix organization nor COC expansion, thus indicating specificity of the binding of some structural component(s) to HA. Fully expanded COCs disassemble when cultured longer than 18 h, a process which occurs also in vivo and which correlates with loss of oocyte fertilizability both in vivo and in vitro. This process involves release of macromolecular HA from the matrix into the medium, with loss of 50% of the HA in the first 8 h of incubation after full expansion. The release is not facilitated when HA oli- gomers, long enough to prevent matrix formation, are added to the culture medium after the COCs are fully expanded. This suggests that cooperative binding to HA of either the serum factor, an endogenously syn- thesized factor(s), or both is required to stabilize the fully expanded COC matrix.

*This work was supported by grants from Minister0 dell’universita e della Ricerca Scientifica e Tecnologica (40 and 60%) and by Consiglio Nazionale delle Ricerche Grants 91.00320.CT04 and FATMA 91.00245.PF41 to (A. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accord- ance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed Bone Research Branch, Bldg. 30, Rm. 106, National Institute of Dental Research, NIH, 9000 Rockville Pike, Bethesda, MD 20892.

The cumulus cell-oocyte complex (COC)’ is a structural unit of the antral mammalian follicle which consists of several layers of compacted follicular cells (cumulus cells) closely surrounding the oocyte (1). As a consequence of the endoge- nous gonadotropin surge during the preovulatory period, the intercellular space between cumulus cells increases, due to the production of a mucoelastic extracellular matrix rich in hya- luronic acid (HA) (2-5). This process, called expansion or mucification, facilitates detachment of the COC from the follicle wall, its extrusion at ovulation, and its capture by oviductal fimbria (6, 7). Moreover, the cumulus cells and the matrix in the expanded COC improve the success of sperm penetration and fertilization (7-11).

Three components are required to induce COC expansion in uitro: 1) a soluble factor produced by the oocyte (12, 13), 2) follicle stimulating hormone (FSH) or CAMP analogues (2, 14-16), and 3) a serum factor (14, 16-18). The first two components are necessary to induce HA synthesis by cumulus cells, whereas the third is important for organizing HA within the extracellular matrix.

When compact, freshly isolated mouse COCs are cultured with FSH in the absence of fetal bovine serum (FBS), the mucoelastic matrix does not form. In this culture condition, most of the cumulus cells lose contact with each other and settle individually on the culture dish. On the other hand, total HA synthesis by FSH-stimulated COCs cultured either in the presence or absence of FBS remains unaltered, but its distribution between the COC extracellular matrix and the culture medium is distinctly different (16,17). In the presence of FBS, -80% of the synthesized HA accumulates in the matrix, whereas in the absence of serum, the same proportion is released into the medium. In either case, the molecular size of the HA synthesized is >2 million daltons, suggesting that the failure in matrix formation when FBS is absent does not depend on changes in the physical properties of this polymer (16). Cultures of mouse mural granulosa cells also exhibit the same serum dependence for organizing HA within the extra- cellular matrix (19).

The serum factor has been recently identified as a protein belonging to the inter-a-trypsin inhibitor family ( I d ) (18). Purified IaI at 1 mg/ml was able to replace serum in support- ing COC expansion, and it was proposed that IaI provides an integral structural component of the matrix, perhaps by in-

’ The abbreviations used are: COC, cumulus cell-oocyte complex; HA, hyaluronic acid; FSH, follicle-stimulating hormone; FBS, fetal bovine serum; I d , inter-a-trypsin inhibitor; BSA, bovine serum al- bumin; hCG, human chorionic gonadotropin; MEM, Eagle’s mini- mum essential medium; CHAPS, 3-[(3-cholamidopropyl)dimethyl- ammoniol-1-propanesulfonate; CS, chondroitin sulfate; HPLC, high pressure liquid chromatography.

20473

Page 2: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

20474 Matrix Organization of Mouse Cumulus Cell-Oocyte Complex

teracting directly with HA. However, the actual mechanism by which this factor acts remains unknown. I d could, in fact, support matrix formation through its proteinase inhibitory activity or by stimulating a specific cellular response.

The extracellular matrix of ovulated COCs isolated from hamster has been studied by light and electron microscopy using freeze substitution (20). The matrix was described as a homogeneous fibrillar structure attached to cumulus cells a t their filopodia1 extensions. These investigators subsequently showed that protease and hyaluronidase disorganized and partially degraded the microstructure of the matrix (21). These findings indicate that interactions between proteins and HA are important for stability of the extracellular net- work between the cells. Mouse and human COC matrices also contain protein(s) and HA in close association (22, 23). How- ever, the specificities of the interactions required for matrix organization have not been analyzed.

The fertilizable life of an oocyte surrounded by the cumulus cells in the expanded COC has been studied i n vivo (24) and in vitro (7, 11). In either case, the fertilizability declines with time and parallels the shedding of cumulus cells from the complex during a process of COC matrix disaggregation. How- ever, this disassembly of the HA-matrix has not been system- atically investigated.

In the present study, we analyze the effects of serum and exogenous HA on endogenous HA retention in the COC matrix and on COC expansion in vitro. Our results strongly support the hypothesis that the serum factor acts as a struc- tural component of the matrix and show that specific binding of this and/or other molecules to HA is essential for successful organization of the COC matrix. In addition, the kinetics of HA release from the cell matrix into the culture medium during the process of COC disaggregation i n vitro and the physical properties of the released macromolecules are de- scribed.

EXPERIMENTAL PROCEDURES

Materials-Female Swiss CD-1 mice were obtained from Charles River; Eagle’s minimum essential medium (MEM) with Earle’s salt (without L-glutamine), sodium pyruvate, Hepes, and dialyzed FBS were purchased from Life Technologies, Inc.; pregnant mares’ serum gonadotropin, bovine serum albumin (BSA), L-glutamine, dimethyl- polysiloxane-5X, and hyaluronic acid (from human umbilical cord) were from Sigma; gentamicin sulfate was from Whittaker Bioprod- ucts; 8 M guanidine HC1 was from Pierce Chemical Co.; papain was from Boehringer Mannheim; Triton X-100 was from Research Prod- uct Internationab [35S]sulfate (-40 Ci/mg, carrier-free) was from ICN Radiochemicals; D-[6-3H]glucosamine (35-40 Ci/mmol) was from Du Pont-New England Nuclear and from ICN Radiochemicals; Sephadex G-50 (fine), Superose 6 (prep grade), Sephacryl S-1000 (superfine), and the scintillation mixture, OptiPhase “HiSafe” 3 were from Pharmacia LKB Biotechnology Inc.; Partisphere PAC (0.4 X 25 cm) was from Whatman; chondroitinase ABC (Proteus vulgaris), and chondroitin sulfate disaccharides ( ADi-OS, ADi-4S, and ADi-6s) were from Seikagaku America; hyaluronidase (Streptomyces hyaluronlyti- cus nov. sp.), CHAPS, and chondroitin sulfate C (CS) (from shark cartilage) were from Calbiochem. Highly purified rat follicle stimu- lating hormone (NIDDK-rat-FSH-1-8) was kindly provided by the National Institute of Diabetes and Digestive and Kidney Diseases and the National Hormone and Pituitary Program of the University of Maryland School of Medicine. Oligomers of HA: octasaccharide (HA,), decasaccharide (HA,,), tetradecasaccharide (HA,,), and HA-35, obtained from high molecular weight HA after testicular hyaluroni- dase digestion, were kindly provided by Dr. James H. Kimura.

Isolation and Culture of COG-Immature 18/21-day-old female Swiss CD-1 mice were injected with 5 IU of pregnant mares’ serum gonadotropin in 0.1 ml of physiological saline and sacrificed by cervical dislocation 44-48 h later. Ovaries were removed and placed in MEM buffered with 25 mM Hepes, containing 1 mg/ml BSA and 50 pg/ml gentamicin. Cultures were prepared as described previously (19). Briefly, COCs were released into the culture medium by punc-

turing large follicles. They were then collected, washed twice using a micropipette, and transferred into droplets of incubation medium covered with dimethylpolysiloxane-5X to prevent evaporation. All cultures were carried out in MEM supplemented with 3 mM gluta- mine, 0.3 mM sodium pyruvate, 50 pg/ml gentamicin, 1 pg/ml BSA, and 100 ng/ml FSH. Unless otherwise indicated, FBS was used at a concentration of 1% (v/v). For each culture, 20 COCs were transferred into a 20-4 droplet of medium and incubated at 37 “C with 5% CO, in humidified air for the times specified in the text.

Morphological Evaluation of the Expansion and Dissociation Process of COCs-The extents of expansion and dissociation of the COCs were evaluated at the end of each culture by morphological and physical criteria under stereomicroscopic observation. The elasticity of the COCs (that is the capacity to recover the original shape without any disruption after stretching inside a narrow micropipette) and an increase in volume compared with compact COCs, were indicators for expansion. Conversely, the decrease or lack of elasticity of the com- plexes and the presence of single cumulus cells settled on the culture dish without contact with each other, were indicators for dissociation. COCs were assessed by subjective evaluation from 0 (absence of dissociation) to 4 (maximum dissociation). In each experiment, the reference culture for normal expansion was incubated for 18 h in complete medium with 1% FBS (dissociation index = 0), whereas the reference culture for maximum dissociation was incubated for 18 h in medium without serum (dissociation index = 4).

Metabolic Radiolabeling of Cultures and Extraction of Radiolabeled Molecules-Cultures were incubated in the presence of [35S]sulfate (-60 pCi/ml) and [3H]glucosamine (100 pCi/ml) for the times indi- cated in the text. At the end of each culture, the incubation medium (20 pl) was aspirated with a micropipette. The COCs were then washed with 20 pl of fresh medium, and the wash added to the medium fraction. Medium and cell matrix were then subjected to the same extraction procedure. Each sample was treated with 20 pl of a papain solution (50 pg/ml final concentration) in 0.1 M Tris, 5 mM EDTA, pH 7.3, at 65 “C for 1 h. The 20-4 aliquot added to the medium (40 ml) was 3 X concentrated, whereas the aliquot added to the cell layer was 1 X concentrated. After digestion, the extraction was completed by adding 1 volume of 8 M guanidine HC1 containing 4% (v/v) Triton X-100.

Quantitation of HA-One-hundred micrograms each of BSA and HA were added to each extract to improve the recovery of radiolabeled macromolecules during subsequent gel filtration procedures. Each extract was then heated at 100 “C for 3 min to inactivate the papain, diluted to 500 pl by adding 0.1 M Tris, 0.1 M sodium acetate, 0.5% Triton X-100, pH 7.3, and eluted on a column of Sephadex G-50 (2- ml bed volume) equilibrated with the same buffer. This step removes the unincorporated radioisotopes and exchanges the guanidine HC1 for a buffer suitable for chondroitinase treatment. The excluded macromolecules were then digested for 40 min by adding 20 p1 (0.2 unit) of chondroitinase ABC dissolved in 0.1 M Tris, 0.1 M sodium acetate, pH 7.3. Approximately one-third of each digested sample was chromatographed on a column of Sephadex G-50 (4-ml bed volume) equilibrated and eluted with the same buffer used for the first elution. The excluded and included fractions were counted for radioactivity to determine the proportion of the radiolabeled macromolecules di- gested by the enzyme. The remaining portion of each sample was analyzed by high pressure liquid chromatography (HPLC) on a Par- tisphere PAC to determine the relative proportion of HA and CS disaccharides. For this purpose, each sample was ultrafiltered through a M, 5,000 cutoff membrane (Ultrafree-MC filter, Millipore) by centrifugation at 4,000 X g to separate digested products from en- zymes and undigested macromolecules. The filtered portion, contain- ing the disaccharides, was then concentrated to a final volume of 20- 50 p1 under vacuum on a Speed Vac concentrator (Savant), and applied to the HPLC column after addition of ADi-OS, ADi-4S, and ADi-6S disaccharides standards (5 pg each). The ADi-HA standard in each sample was derived from the enzymatic digestion of the carrier HA. The column was eluted with acetonitri1e:methanol: aqueous buffer in a ratio 52:12:36 at a flow rate of 0.7 ml/min. The aqueous buffer contained 0.5 M Tris, 0.1 M boric acid, pH 8.0. Fractions of 0.7 ml were collected for analysis. The mass of HA synthesized during the labeling period (in hexosamine equivalents) was determined as described previously (25) by calculating the specific activity of the UDP-N-acetylhexosamine pools from the ratio of 3H to 35S in the ADi-4s.

Gel Filtration-A column of Sephacryl S-1000 (0.8 X 30 cm) was eluted with 4 M guanidine HCl, 50 mM sodium acetate, 0.5% CHAPS, 0.1% Triton X-100, pH 6.0, at a flow rate of 0.4 ml/min. A column of

Page 3: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

Matrix Organization of Mouse Cumulus Cell-Oocyte Complex 20475

Superose 6 (0.8 X 30 cm) was eluted with the same solvent at a flow rate of 0.4 ml/min. Fractions of 0.4 ml were collected for analysis.

Treatment with Streptomyces Hyaluronidase-The. enzyme, dis- solved in 0.1 M sodium acetate, 100 pg/ml BSA (10 enzyme units in 20 pl), was added to the samples (100-300 pl of 0.1 M Tris, 0.1 M sodium acetate, 0.5% Triton X-100, pH 7.3) followed by incubation at 37 “C, for 1 h.

Measurement of Radioactiuity-Standards with 35S and 3H activity were prepared in the same solvent conditions as for the samples in each experiment to determine quench and spillover corrections.

Statistical Analysis-All data shown in Figs. 1-8 were means of duplicate determinations, and standard deviations were less than 10% of the mean values. The standard error of the mean for total HA synthesis within each experiment was always less than 10% of the mean value (Figs. 1, 3, 5, 7, 8, and 10). The differences in the mean values were evaluated with the Student’s t test; p values of less than 0.05 were considered statistically significant (Figs. 2, 7, 8, and 10). Data were fit with four-parameter logistic functions (Figs. 1, 7, and 8) and with exponential functions (Fig. lo), using least squares methods.

RESULTS

Dose Response of FBS on HA Retention in the COC Matrix- Mouse cumulus COCs are stimulated to synthesize HA i n uitro by FSH treatment. In the presence of serum, most of the HA is retained in the extracellular matrix and the COC expands. In its absence most of the HA is released into the culture medium, and a large proportion of the cumulus cells dissociate from the complex before settling on the culture dish as individual cells (see Fig. 9, A and B ) .

We tested different concentrations of FBS for their ability to retain HA in the matrix and support COC expansion. Freshly isolated mouse COCs were incubated for 18 h with FSH and different serum concentrations in the presence of radiolabeled precursors for glycosaminoglycan synthesis. Cell matrix and culture medium compartments were then sepa- rated and the amounts of HA in each fraction determined. Total HA synthesis did not change significantly with increas- ing concentrations of serum compared with the control value for cultures without serum (Fig. 1, upper panel). In the ab- sence of serum, or at concentrations up to 0.1%, only 10-20% of the total HA produced by the cells was retained in the matrix (Fig. 1, lower panel), and a large number of cumulus cells were dissociated from the complex (see dissociation index between the two panels). As the serum concentration in- creased, the percentage of HA in the matrix increased (mid- point, -0.25% FBS). Maximum retention of HA (85-90%) and complete COC expansion were obtained when a minimum of -1% FBS was present. Thus, this concentration of a single lot of FBS was used for all the subsequent experiments. These findings agree with previous reports (17, 18).

Time Course of HA Synthesis and Distribution with and without FBS-Total synthesis of HA and its distribution between cell matrix and culture medium compartments were determined after different incubation times in the presence or absence of serum for up to 25 h. There were no differences in the temporal pattern of total HA synthesis in the two experimental conditions (Fig. 2, triangles). In the presence of serum (Fig. 2, upper panel), most (70-90%) of the HA syn- thesized up to 20 h of culture was retained within the matrix (filled circles). After that time, a higher proportion of HA was released into the culture medium (unfilled circles), and the COC began to disaggregate. This process of matrix disassem- bly was analyzed in more detail in subsequent experiments (see below). In the absence of serum (Fig. 2, lower panel), after 6 h, only 10-40% of the synthesized HA was retained in the matrix.

From the data shown in Fig. 2, the average rate of HA synthesis per hour was determined for the indicated time

0 0.01 0.1 1 10

FBS (%)

FIG. 1. Effect of FBS concentration on HA synthesis, its retention in the COC matrix, and COC expansion. Cultures were incubated for 18 h with radioisotopes in the presence of different concentrations of FBS. For each culture, the total amount of radio- labeled HA synthesized (upper panel) and its proportion in the cell extract (lower panel) were determined. The values for total HA synthesis (upper panel) were not significantly different by statistical analysis. The midpoint of the curve is -0.25% FBS (arrow). The extent of dissociation of COCs assessed at the end of each culture is shown between the two panels (range: 0-4, see “Experimental Pro- cedures”).

0.7 % BSA

3 6 9 12 15 18 21 24

Culture time (h)

FIG. 2. Time course of HA synthesis and its distribution between cell extract and medium in the presence and absence of FBS. Cultures were incubated with radioisotopes for 6, 9, 12, 20, and 25 h in the presence (upper panel) or absence (lower panel) of FBS. At the end of each culture, the total amount of HA (V) and its distribution between medium (0) and cell extract (0) were deter- mined. The total values at any time point (V) in the two different conditions were not significantly different by statistical analysis. The inset shows the rate of HA synthesis during the indicated time periods, either in the presence (filled bars) or absence (unfilled bars) of FBS.

periods in the presence and absence of FBS (Fig. 2, inset). In both cases synthesis was transient, reaching maximal rates between 6 and 12 h of culture. No detectable HA was synthe- sized after 20 h of culture even though proteoglycan synthesis

Page 4: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

20476 Matrix Organization of Mouse Cumulus Cell-Oocyte Complex

(data not shown) continued at a nearly constant rate during both the 12-20-h and the 20-25-h time periods.

Temporal Effect of Serum on HA Retention in the Matrix- Serum was added or removed at different times during the culture, and the effect on HA distribution was analyzed at the end of an 18-h incubation period. In one series of experiments, FBS was present in the incubation medium from the begin- ning of the culture and then removed at different times (3, 6, 9, and 12 h) by washing the COCs with serum-free medium. In the first of these experiments the radiolabeled precursors were added at time 0 and were present throughout the 18-h incubation (Fig. 3), whereas in the second experiment the radioisotopes were added immediately after serum removal (Fig. 4). Diagrams of the experimental designs are presented at the top of each figure.

The experiment in which the isotopes were added from the beginning showed that total HA synthesis was unaffected by serum removal (Fig. 3, lowerpanel), whereas its accumulation in the cell matrix increased with the increase of exposure time to FBS (Fig. 3, upper panel). When serum was removed at 9 h of culture, the amount of HA present in the matrix was about 50% of the total, a sufficient amount to give nearly complete expansion by morphological criteria. However, only when serum was continuously present did the COCs show maximal HA retention (-80%) and full expansion. The ad- dition of isotopes after removing the serum (Fig. 4) showed that almost all the HA synthesized afterwards (70-90%) was released into the culture medium, independent of the time of serum removal.

The effect of serum addition after different incubation times (3, 6, 9, and 12 h) was investigated in another series of experiments. In the first of them, radioisotopes were added at

- 1% FBS ..... 0.1% BSA

4 Isotopes addition t ........................... L"" ..................

L .................. .I ............. .I..... .... '

- a Hours with serum

FIG. 3. Effect of FBS removal during the culture on HA synthesis and its accumulation in the COC matrix. The sche-

the results are shown in the lower panels. The lengths of the solid matic diagram at the top illustrates the experimental protocol, and

lines indicate the culture times with FBS and correspond to the data points present in the lowerpanels directly below the end of each solid line. All cultures were incubated for 18 h with radioisotopes. The first and last cultures represent controls incubated in the absence or presence of 1% FBS, respectively. Serum was removed from the remaining cultures at 3, 6, 9, and 12 h. The upper panel shows the proportion of the total HA present in the cell extract in each condi- tion, and the lower panel shows the total HA synthesis.

- 1% FBS 0.1% BSA

4 Isotopes addition + ........................... A ...................... J. .................. 2. ............

A"""." i.

x I " " " ' I

I 0 3 6 9 12 15 18

Hours with serum FIG. 4. Effect of FBS removal during the culture on HA

synthesis and its distribution between cell matrix and me- dium. The schematic diagram at the top illustrates the experimental protocol, and the results are shown in the lowerpanel. The lengths of the solid lines indicate the culture times with FBS and correspond to the data points present in the lower panel directly below the end of each solid line. The first and the last cultures represent controls incubated in the absence or presence of 1% FBS, respectively, and with radioisotopes for the entire 18 h. In the remaining cultures, radioisotopes were added at 3, 6,9, and 12 h, respectively, when FBS was removed. The panel shows the total amount of HA synthesized in each culture (V) and its distribution between medium (0) and cell extract (0).

1 """" - - - - -

I.. ............... .___ + ..........................

- 0 " - E, 0 3 6 9 12 15 18

Hours without serum - FIG. 5. Effect of FBS addition during the culture on HA

synthesis and its accumulation in the COC matrix. The sche- matic diagram at the top illustrates the experimental protocol, and the results are shown in the lower panels. The lengths of the dashed lines indicate the culture times without FBS and correspond to the experimental points in the lower panels directly below the end of each dashed line. All cultures were incubated for 18 h with radioisotopes. The first and last cultures represent controls incubated in the pres- ence and absence of 1% FBS, respectively. In the remaining cultures, FBS was added at 3, 6, 9, and 12 h. The upper panel shows the proportion of the total HA present in the cell extract, and the lower panel shows the total HA synthesis.

Page 5: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

Matrix Organization of Mouse Cumulus Cell-Oocyte Complex 20477

time 0 (Fig. 5) and in the second they were added at the time of serum addition (Fig. 6).

The experiment in which the radiolabeling precursors were present from the beginning showed that total HA synthesis was unaffected by FBS addition (Fig. 5, lowerpanel), whereas the percentage of HA present in the matrix (Fig. 5, upper panel) correlated with the time of serum addition (the earlier the serum was added, the higher the percentage). Morpholog- ical analyses performed at the end of each culture indicated that the absence of serum for the first 6 h did not significantly affect COC expansion. This indicates that if only -60% of the total HA is organized within the matrix, the COCs still appear fully expanded.

In the parallel experiment, the radioisotopes were added at the time of serum addition to determine the distribution of HA synthesized afterwards (Fig. 6). In the first four time points tested, irrespective of the time of serum addition (0, 3, 6, or 9 h of culture), the proportion of matrix-associated HA remained unchanged (70-80%). However, when serum was added at 12 h of culture, only -50% of the HA synthesized afterwards was retained in the matrix. The amount of HA synthesized in this period (12-18 h) represents -25% of the total HA synthesized during the entire 18 h of culture (Fig. 2, inset). Therefore, -12% of the total HA retained in the matrix is derived from molecules synthesized after 12 h of culture. In contrast, the results in Fig. 5 indicate that 50% of the total HA synthesized can be recovered in the COC matrix when serum is added at 12 h. Thus, a significant proportion of HA synthesized before 12 h can be organized into the matrix when serum is added at 12 h, probably because of the high local concentration of HA around the cumulus cells at this time.

Effect of Exogenous Glycosaminoglycans and HA Oligomers on Endogenous HA Distribution and COC Expansion-We analyzed total HA synthesis and its distribution between matrix and medium in cultures with varying concentrations

- 1% FBS ..~-. 0.1 % BSA

4 Isotopes addition

............. -L

.................. 4 4""" ....................

L 0 3 6 9 12 15 18 Hours without serum

FIG. 6. Effect of FBS addition during the culture on HA synthesis and its distribution between cell extract and me- dium. The schematic diagram at the top illustrates the experimental protocol, and the results are shown in the lower panel. The lengths of the dashed lines indicate the cdture times without FBS and corre- spond to the experimental points in the lower panel directly below the end of each dashed line. The first and the last cultures represent controls incubated in the presence or absence of 1% FBS, respectively, and with radioisotopes for the entire 18 h. In the remaining cultures, FBS and radioisotopes were added at 3, 6, 9, and 12 h. The panel shows the total amount of HA synthesized in each culture (V) and its distribution between cell extract (0) and medium (0).

of exogenous HA or CS (Fig. 7) or HA oligosaccharides (Fig. 8) added from the beginning of culture to medium containing 1% FBS.

Exogenous HA at a concentration of 1 mg/ml induced the release of about 80% of the newly synthesized HA into the culture medium (midpoint, -0.6 mg/ml HA; Fig. 7, lower panel). In this condition, expansion does not occur. Further- more, some of the cumulus cells dissociated from the com- plexes, and most of them remained compact around the oocyte (Fig. 9C). The addition of CS up to 5 mg/ml affected neither HA distribution nor COC expansion. Neither HA nor CS addition altered total HA synthesis significantly (Fig. 7, upper panel).

Similar experiments were done with HA oligomers to iden-

80

40

20

0 0.01 0.1 1 10 Exogenous HA or CS (mg/ml)

FIG. 7. Effect of increasing concentrations of exogenous HA and CS on endogenous HA synthesis and its accumulation in the medium. Cultures were incubated for 18 h with radioisotopes in the presence of 1% FBS and increasing concentrations of HA (0) or CS (0). The upper panel shows the total HA synthesis in the two different conditions and the lower panel the proportion of the total HA that accumulates in the medium. The midpoint of the curve for HA is -0.6 mg/ml (arrow).

P

0 0.01 0.1 1 10 HA Oligomers (mg/ml)

FIG. 8. Effect of increasing concentrations of exogenous HA oligomers on endogenous HA synthesis and its accumulation in the medium. Cultures were incubated for 18 h with radioisotopes in the presence of 1% FBS and increasing concentrations of HA decasaccharides (GlcUA-GlcNAc), (0) or HA octasaccharides (GlcUA-GlcNAc)d (0). The upperpanel shows the total HA synthesis in the two different conditions and the lower panel the proportion of the total HA that accumulates in the medium. The midpoint of the curve for the HA decasaccharide is -0.2 mg/ml (arrow).

Page 6: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

20478 Matrix Organization of Mouse Cumulus Cell-Oocyte Complex

FIG. 9. Different ia l in terference contrast photomicrographs of COCs a f t e r 18 h of incubation in different cu l tu re conditions i n t h e presence of FSH. A shows a COC cultured with 1% FBS. The cumulus cells are embedded in the matrix and form a spherical struc- ture around the oocyte. The plane of the image is at the level of the oocyte and above the culture dish. B shows a COC cultured with 0.1% BSA. Some of the cumulus cells are still compacted around the oocyte, but most have settled on the culture dish after losing contact with each other. The plane of the image in this panel and in C and D is at thelevel of the culture dish. C shows a COC cul- tured with 1% FBS and l mg/ml HA. ., .. ,

Most of the cumulus cells are still com- pacted around the oocyte, whereas some of them have settled on the culture dish and started to spread because of the pres- ence of FBS which facilitates adhesion. D shows a COC cultured with 1% FBS and 1 mg/ml HA decasaccharide. Some cells are still compacted around the oo- cyte, but most of them have detached from the complex. Many are flattened and attached to the culture dish because ' (i of the presence of FBS. The bur repre- 6. ~~ 2 sents 50 pm.

r' +: c

k i

tify the minimum length that was able to displace endogenous HA from the cell matrix (Fig. 8). Most (-80%) of the endog- enous HA was displaced into the culture medium with accom- panying complete COC dissociation when HA decasaccharides were present a t 1 mg/ml (midpoint, -0.2 mg/ml). In this case, most of the cumulus cells dissociated from the complexes and settled onto the culture plate in a manner similar to that observed when FBS is absent (Fig. 9, B and D). Longer HA fragments (HAl4 and HA+) were also tested and showed the same effects as the decasaccharides (data not shown). Con- versely, the addition of up to 2 mg/ml of the HA octasacchar- ide did not significantly displace the endogenous HA, and COC expansion occurred normally.

Disassembly of the COC Matrix with Time-Maximum COC expansion and retention of HA occurs at approximately 18 h in culture. After this time the COC undergoes a process of disaggregation, characterized by a progressive disassembly of the matrix, while the released cumulus cells individually settle on the dish. This process was studied by analyzing the time course of HA release from the matrix after 18 h of culture, either in the presence or absence of HA oligosaccharides. Cultures were incubated for 18 h with radioisotopes and 1% serum to allow full COC expansion. Some cultures were then

analyzed for HA content and distribution (18-h control point). Identical aliquots (1/4 of the final volume) of the radiolabeling medium with FBS and either with or without HA oligomers (14 monosaccharides in length) were added to the remaining cultures, followed by incubation for different times up to 72 h total (ie. 54 h after maximum COC expansion). HA distri- bution for all cultures from three separate experiments are shown in Fig. 10 as the percentage of the total HA present in the culture medium a t each time. The total amount of HA synthesized was the same within experimental error for all cultures (data not shown), confirming the evidence that HA synthesis stops by about 20 h of culture (Fig. 2, inset).

HA release from the expanded COCs was rapid during the first 12 h (30 h of total culture) with the loss of -50% of the HA originally in the matrix occurring after -8 h of incubation (25 h of total culture) (Fig. 10, upper punel and inset). The plateau value of -85% of total HA present in the medium was reached at the end of the subsequent 18 h (48 h of total culture). The addition of HA oligomers a t 18 h of culture had no effect on the kinetics of HA release nor on the plateau value reached. Thus, HA oligosaccharides which can effec- tively prevent the assembly of the endogenous HA in the matrix if present during COC expansion (Figs. 7 and 8) are

Page 7: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

Matrix Organization of Mouse Cumulus Cell-Oocyte Complex 20479

1 % FBS v

1% FBS t HA,, after 18 h

“A””

20 40 60 Culture time (h)

FIG. 10. Disassembly of the COC matrix and time course of HA release from the matrix after normal expansion in vitro. Cultures were incubated for 18 h with 1% FBS and radioisotopes. After this time, a solution of HA tetradecasaccharide (HAla) a t 1-2 mg/ml in culture medium with radioisotopes was added to some cultures that were then incubated for additional times up to a total of 72 h (lower panel). Control cultures were treated identically, but without the presence of HA oligomers in the culture medium added at 18 h (upper panel). The two curves show results from three different experiments (0,0, and V) which did not differ significantly by statistical analysis. The inset shows the proportion of HA left in the matrix plotted on a semilogarithmic scale for the conditions shown in the two panels. The unfilled squares are the means of the values obtained in the control condition and the filled squares the means of the values obtained after addition of HA oligomers at 18 h of culture.

unable to facilitate COC disassembly after the matrix is formed.

The physical properties of the radiolabeled macromolecules present in medium and cell matrix fractions derived from cultures incubated for 18, 47, and 72 h (duplicates of cultures shown in Fig. 10, upper panel) were analyzed by chromatog- raphy on Sephacryl S-1000 in dissociative conditions, after the standard papain digestion and extraction procedure. In all cases 3H-labeled HA was present as a major peak eluting in the excluded column volume with a trailing shoulder (Fig. 11, unfilled circles), as indicated by the sensitivity of this peak to digestion with Streptomyces hyaluronidase (dashed lines). At 47 and 72 h, however, the chromatographic patterns of the 3H-labeled HA in the medium fractions showed some degra- dation as indicated by a more extensive trailing into the column (Fig. 11, c and e compared with a). In a separate experiment, S-1000 analyses were done on medium and cell matrix fractions derived from a culture incubated for a total of 30 h, by which time -70% of total HA was already released into the medium (see Fig. 10, upper panel). In this case the elution patterns did not show any significant differences from those of the control 18 h culture (data not shown). Thus, the limited HA degradation observed at 47 and 72 h of culture is unrelated to the redistribution of HA from the matrix into the medium. In all cases, the 35S-labeled material eluted as a single peak (shown only in Fig. l l a , offset) which was not affected by the Streptomyces hyaluronidase treatment.

The Sephacryl S-1000 profiles for the medium fractions

s- 1000 MEDIUM

4

2

0 0.0 1 .o

Kd

2 -

1 -

0 -

4 -

2 -

0 -

3 -

2 -

1 -

0 -

CELL MATRIX

20% d.

, ’ / ’ / I

0.0 1 .o Kd

FIG. 11. Sephacryl 5-1000 elution profiles of medium and cell matrix fractions. Duplicates of samples from some of the cultures shown in Fig. 10 were analyzed by S-1000 chromatography in dissociative conditions after the standard extraction procedure. The panels show analyses of cultures incubated with 1% FBS for 18 h ( a and b ) , for 47 h ( c and d ) , and for 72 h ( e and f ) . In all panels, the dashed lines are the elution profiles of separate samples after digestion with Streptomyces hyaluronidase. A profile for the 35S- labeled material is shown offset in a and showed no significant difference either between the different samples or after treatment with the enzyme. The percentage in the left corner of each panel indicates the proportion of the total HA present in the sample.

contained significant amounts of 3H in or near the column total volume. Since this could result from the presence of much smaller HA fragments, aliquots of the same media were subjected to Superose 6 chromatography (Fig. 12). All samples showed a 3H-labeled peak eluting in or near the excluded column volume (Fig. 12, unfilled circles) that was susceptible to Streptomyces hyaluronidase digestion (dashed lines). Two additional major 3H-labeled peaks were observed, the first eluting with the 35S-labeled peak, representing glycosamino- glycan chains attached to peptidic fragments, and the second eluting just ahead of the column total volume. Neither of these peaks were susceptible to the enzyme treatment. Thus all of the HA molecules present in the culture medium at 47 and 72 h were still excluded on a column of Superose 6 (exclusion limit of 300 kDa).

DISCUSSION

The presence of FBS in the culture medium is necessary for organizing HA in the extracellular matrix during COC expansion in uitro (16, 17). The serum factor responsible for this effect has been identified by Chen et al. (18) as a member of the I d family. Among several possibilities of action, the structural function of I d is supported by its immunolocali- zation in vivo in the extracellular matrix of preovulatory follicles (18) at sites where HA has been localized by histo- chemical analyses (5). The results described in this paper provide strong support for this hypothesis and also show the importance of specific interactions between HA and other component(s) for organizing the matrix around the cells. When serum is added to the incubation medium after different

Page 8: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

20480 Matrix Organization of Mouse Cumulus Cell-Oocyte Complex

SUPEROSE 6 MEDIUM

I I

FIG. 12. Superose 6 elution profiles of medium fractions shown in Fig. 11. The panels represent media from cultures incu- bated with 1% FBS for 18 h (a), for 47 h ( b ) , and for 72 h (c). In all panels, the doshed lines are the elution profiles of separate samples after digestion with Streptomyces hyaluronidase. The profile for the 36S-labeled material, shown offset only in a, showed no significant difference from all the remaining profiles with or without enzyme treatment.

times of culture in the presence of FSH, HA synthesized prior to adding the serum is preferentially recovered in the medium compartment, whereas HA synthesized afterwards is prefer- entially retained in the matrix. Conversely, when serum is removed after different times in culture, HA synthesized before removal is preferentially retained in the matrix and that synthesized afterwards preferentially recovered in the medium. These results indicate that maximum HA retention within the matrix is always obtained when FBS is continu- ously present in the culture, thus supporting the structural function of IaI. This finding, however, does not rule out the possibility that IaI can also act as an anti-protease within the matrix, protecting it from premature degradation.

When macromolecular HA (but not CS) or HA oligosac- charides with at least five repeating disaccharide units, (GlcUA-GlcNAc)B, are added to the culture medium contain- ing FBS, the endogenously synthesized HA is displaced from the matrix (Figs. 7 and 8). Moreover, the cumulus cells, especially when cultured in the presence of the HA decasac- charide, dissociate from the complex in the same way that they dissociate if serum is absent (Fig. 9, b, c, and d) . The effective concentration of endogenous HA in the expanded COC matrix is 0.25-0.5 mg/ml (16), the approximate range required for exogenous HA and HA oligosaccharides to exert their displacement effect. This indicates that some structural component(s) that binds to HA with specificity for the HA decasaccharide is required to organize and stabilize the newly synthesized HA within the matrix. Whether this structural component is the serum factor, an endogenously synthesized molecule, or both remains to be determined.

It has been shown that proteoglycans synthesized by cu- mulus cells are not directly involved in organizing the COC extracellular matrix. For example, inhibition of their synthe- sis in the presence of xylosides does not alter COC expansion

(16). Furthermore, the proteoglycans produced by cumulus cells have molecular properties similar to those synthesized by mural granulosa cells in culture which do not bind to HA (26). On the other hand, evidence has been provided that suggests that IaI binds to HA in sinovial fluid during an inflammatory response (27). Moreover, another protein (85 kDa) has been isolated from serum which appears to bind to HA with a specificity for HA decasaccharides (28). This suggests that the 85-kDa serum protein could be part of the I d complex recently purified from FBS (29) and that repre- sents the extracellular matrix stabilizing factor described by Chen et al. (18). This component contains a light chain (-30 kDa) with anti-protease activity and a heavy chain (86 kDa) of unidentified function linked by chondroitin sulfate (29). A similar structure has been also described for the human IaI (30).

Once formed in uitro, the HA matrix dissociates over a period of 1-2 days. Initially the process is rapid, with about 50% of the net loss of HA occurring over 7 h, between 18 and 25 h of culture (Fig. 10, upperpanel and inset). Thus, the loss of HA from the expanded COC matrix begins at or near the time when the cumulus cells stop synthesizing HA (Fig. 2, inset). Cell viability remains high, since dissociated cells attach to the plate during the following days in culture. During the disaggregation process, the HA molecules show limited degradation, primarily at later times, after most of the HA has already been released into the culture medium. This suggests that dissociation is not the direct result of extensive cleavage of HA by enzymatic activity. It appears more likely that protease(s), produced by cumulus cells and/or the oocyte (31,32), can destabilize other proteins important in the matrix structure. On the other hand, the limited HA degradation observed at longer incubation times may involve such proc- esses as oxidative/reductive depolymerization by oxygen- derived free radicals (33, 34).

The fertilizable life of mouse oocytes has been studied in uiuo after hCG-induced ovulation and subsequent insemina- tion at different times (24). Compared with rates of fertiliza- tion obtained just after ovulation (-14 and -12 h after hCG injection in immature and mature mice, respectively), the rates remained high until 24-25 h (immature mice) and 27- 28 h (mature mice) after hCG injection. In parallel, the authors analyzed the morphological appearence of the cumu- lus mass in COCs from mature mice at different times after ovulation and observed a progressive reduction in size of the cellular “clot” around the oocyte, which nearly disappeared by 30 h after hCG injection. Thus, the disappearance of the microenvironment around the oocyte seems significantly re- lated to successful fertilization. The fertilizability of COCs matured in vitro for different times has also been studied (7). Maximum fertilization rates were reached at 20 h of culture, correlating closely with the time of maximum expansion and retention of HA in the matrix. After 20 h, the fertilizability slowly declined and was completely abolished by 40 h of total culture, correlating closely with the kinetics of release of HA from the matrix (Fig. 10). Thus, in uitro as well as in uiuo, optimal fertilization appears to correlate with the presence of an organized HA-matrix which retains the cumulus cells around the oocyte.

Although exogenous HA oligomers (decasaccharides or larger) can prevent COC matrix assembly, they do not accel- erate the disaggregation process. This result distinguishes this matrix from other pericellular matrices. Chondrocytes, for example, exhibit large “coats” associated with their plasma membrane which are retained by interactions between HA and a cell surface HA receptor. As in other cell types, these

Page 9: Effects of Exogenous Hyaluronic Acid and Serum on Matrix ... · Effects of Exogenous Hyaluronic Acid and Serum on Matrix Organization and Stability in the Mouse Cumulus Cell-Oocyte

Matrix Organization of Mouse Cumulus Cell-Oocyte Complex 20481

HA receptors (probably belonging to the CD-44 family) show HA-hexasaccharide specificity for binding (35). Thus, this size of HA oligomer not only can prevent the assembly of the pericellular coat around the chondrocytes but can also dis- place a fully organized pericellular matrix within 2 h (36). On the other hand, the effects of HA oligomers on the COC extracellular matrix have some similarities with those on cartilage proteoglycan aggregates. Decasaccharides, but not octasaccharides, in this case, effectively compete with the ability of the monomer proteoglycan (or the link protein) to bind to HA. However, when the monomer and link protein cooperatively bind to HA, the ternary aggregate is not disso- ciated in the presence of the HA decasaccharides (37). Once the COC matrix is formed, then, cooperativity of several structural elements may be responsible for the stability of the structure to the presence of such HA oligomers. Future studies will focus on the identification of these elements and their characteristics of binding to HA in order to compare them with other known HA-binding proteins.

REFERENCES

2. Dekel, N., and Kraicer, P. F. (1978) Endocrinology 102, 1797-1802 1. Pedersen, T., and Peters, H. (1968) J. Reprod. Fertil. 17, 555-557

3. Dekel, N., Hillensjo, T., and Kraicer, P. F. (1979) Biol. Reprod. 20, 191-

4. Dekel, N., and Phillips, D. M. (1979) BioL Reprod. 21, 9-18 5. Salustri, A., Yanagiahita, M., Underhill, C. B., Laurent, T. C., and Hascall,

6. Mahi-Brown, C. A., and Yanagimachi, R. (1983) Gamete Res. 8.1-10 7. Chen, L., Russell, P. T., and Larsen, W. J. (1993) Mol. Reprod. Deu. 34,

8. De Felici, M., and Siracusa, G. (1982) Gamete Res. 6, 107-113 9. Meizel, S. (1985) Am. J. Amt . 174,285-302

197

V. C. (1992) Deu. Biol. 151,541-551

87-93

10.

11. 12.

13.

14. 15. 16.

17. 18.

19.

20.

21.

22. 23. 24. 25.

26.

27.

28.

29. 30.

31. 32.

33.

34.

35.

36. 37.

Vanderhyden, B. C., and Armstrong, D. T. (1989) Biol. Reprod. 40, 720-

Fukui, Y. (1990) Mol. Reprod. Deu. 26,40-46 Salustri, A., Yanagishita, M., and Hascall, V. C., (1990) Deu. Biol. 138,

728

Buccione, R., Vanderhyden, B. C., Caron, P. J., and Eppig, J. J. (1990) Deu.

Ep ig J. J. (1979) Nature 281,483-484 Eppig, J. J. (1979) J. Exp. Zool. 208,111-120

Sagstk, A., Yanagishita, M., and Hascall, V. C. (1989) J. Biol. Chem. 264,

E pig, J. J. (1980) Biol. Re rod 22 629 633 &en, L., Mao, S. J., and tarsen, d. J. (1992) J. Biol. Chem. 267, 12380-

26-32

Biol. 138,16-25

13840-13847

12.186 Saiu&i, A., Ulisse, S., Yanagishita, M., and Hascall, V. C. (1990) J. Biol.

Yudin, A. I., Cherr, G. N., and Katz, D. F. (1988) Cell Tissue Res. 261, Chem. 266,19517-19523

5.5.5-564 Cherr, G. N., Yudin, A. I., and Katz, D. F. (1990) Deu. Growth & Difier. 4, ." ."

36.1-3fi.5 Talbot, P., and DiCarlantonio, G. (1984) Gamete Res. 10, 127-142 Dandekar, P and Talbot P. (1992) Mol. Re rod Deu 31,135-143

Yana 'shita, M Salustri, A., and Hascall, V. (1989) Methods Enzymol. Marston, J. H., and Chan'g, M. C. (1964) J. gy:.Zool.' 156, 237-252

Yanagmhlta, M., Rodbard, D., and Hascall, V. C. (1979) J. Biol. Chem.

Hutadilok, N., Ghosh, P., and Brooks, P. M. (1988) Ann. Rheum. Dis. 47,

." ."

1 7 9 435-44s

264,911-920 277-9115

Talbot, P., and DiCarlantonio, G. (1984) Gamete Res. 10, 127-142 Dandekar, P and Talbot P. (1992) Mol. Re rod Deu 31,135-143

Yana 'shita, M., Salustri, A., and Hascall, V. (1989) Methods Enzymol. Marston, J. H., and Chan'g, M. C. (1964) J. gy:.Zool.' 156, 237-252

." ."

1 8 A9K-AAC.

Yanagishita, M., Rodbard, D., and Hascall, V. C. (1979) J. Biol. Chem.

Hutadilok, N., Ghosh, P., and Brooks, P. M. (1988) Ann. Rheum. Dis. 47,

A . Y , "" 77"

264,911-920 277-9115

Yoneda, M., Suzuki, S., and Kimata, K. (1990) J. Biol. Chem. 265,5247-

Enghild: J. J., 'Salvesen, G., Heka, S. A., Th@gersen, I. B., Rutherfmd, S., Castillo G. M. and Templeton D. M. (1993) FEBS Lett. 318,292-296

". . ".," 5257

and Pizzo. S. V. 11991) J. Rinl. ChPm 266. 747-7.51 ~ ." ~ ~~~~

Huarte, J., Be&, D., and Vassalli, J. D. (I986).Cell.43, 551-558 Canipari, R., O'Connell, M. L., Meyer, G., and Strickland, S. (1987) J. Cell

Herp, A. (1980) in The Carbohydrates: Chemistry and Biochemistry (Pig- Biol. 105,977-981

man, W., and Horton. D., eds) 2nd Ed.. Vol. lB, DD. 1276-1297. Academlc

~ -. . -, . . - . . .. - . . . . . . .

Press, New York

Biol. &em. '265,7758-ff59

Underhill 2 (1992) J. Cell B e l . 111,2765-2774

. "

Uchiyama H. Dobashi Y Ohkouchi, K., and Nagasawa, K., (1990) J.

Culty, M., Mi ake, K., Kincade, P. W., Silorski, E., Butcher, E. C., and

Knudson, C: B. (1993) J. cell Btol. 120,825-834 Hascall, V. C., and Heinegard, D. (1974) J. Biol. Chem. 249,4242-4249