Rat Spongiotrophoblast-specific Protein Is Predominantly a Unique Low Sulfated Chondroitin Sulfate Proteoglycan*

We have previously demonstrated that the human placenta contains a uniquely low sulfated extracellular aggrecan family chondroitin sulfate proteoglycan (CSPG). This CSPG is a major receptor for the adherence of Plasmodium falciparum-infected red blood cells (IRBCs) in placentas, causing pregnancy-specific malaria. However, it is not known whether such low sulfated CSPGs occur in placentas of other animals and, if so, whether IRBCs bind to those CSPGs. In this study, we show that rat placenta contains a uniquely low sulfated extracellular CSPG bearing chondroitin sulfate (CS) chains, which comprise only ∼2% 4-sulfated and the remainder nonsulfated disaccharides. Surprisingly, the core protein of the rat placental CSPG, unlike that of the human placental CSPG, is a spongiotrophoblast-specific protein (SSP), which is expressed in a pregnancy stage-dependent manner. The majority of rat placental SSP is present in the CSPG form, and only ∼10% occurs without CS chain substitution. Of the total SSP-CSPG in rat placenta, ∼57% is modified with a single CS chain, and ∼43% carries two CS chains. These data together with the previous finding on human placental CSPG suggest that the expression of low sulfated CSPG is a common feature of animal placentas. Our data also show that the unique species-specific difference in the biology of the rat and human placentas is reflected in the occurrence of completely different CSPG core protein types. Furthermore, the rat SSP-CSPG binds P. falciparum IRBCs in a CS chain-dependent manner. Since IRBCs have been reported to accumulate in the placentas of malaria parasite-infected rodents, our results have important implications for exploiting pregnant rats as a model for studying chondroitin 4-sulfate-based therapeutics for human placental malaria.

tendon, and in cornea, bone, umbilical cord, and blood vessels (1)(2)(3). CSPGs are also ubiquitous in tissues and cell types of mammalian and other vertebrate animals as components of extracellular matrices as well as cell surface molecules. The CSPGs of cartilage and those from a wide variety of tissues and cell types of different animals have been extensively characterized with respect to their core proteins and the structural characteristics of the constituent CS chains (1)(2)(3)(4)(5)(6). However, very little is known about the CSPGs in the maternal blood space of the placentas of various animals. Previously, we have shown that human placenta contains a major extracellular, unusually low sulfated aggrecan family CSPG, localized predominantly in the intervillous space (7,8). The CS chains of the placental CSPG consist of, on average, ϳ8% 4-sulfated and ϳ92% nonsulfated disaccharide residues (7). The human placental CSPG is a mixture of two major distinctively sulfated aggrecan species, one with 2-3% and the other with 9 -14% sulfated disaccharide moieties (9). Although the biological function of the low sulfated aggrecan CSPG in human placenta is not known, it is possible that the placental low sulfated CS chains are involved in mobilizing nutrients, cytokines, hormones, and growth factors required for placental function and for the development of the fetus. The low sulfated CS chains are ideally suited for this function, since they are likely to reversibly and readily release the bound factors for their function.
We have previously shown that the low sulfated CSPG is the receptor for the adherence of Plasmodium falciparum IRBCs in the placentas of pregnant women and that the sulfate groupclustered regions of the C4S chains are the IRBC-binding sites (7)(8)(9). Previous studies have also shown that P. falciparum IRBC binding to C4S involves the participation of both sulfated and nonsulfated disaccharide residues, and a dodecasaccharide motif with two 4-sulfated disaccharides is the minimal structural motif for optimal binding of IRBCs (9 -11). Recently, we have defined the role of the key functional groups, including the acetamido and carboxyl groups of C4S for IRBC binding. 3 Although the data from these studies offer strategies for developing C4S oligosaccharide-based therapy for placental malaria, animal models for studying the therapeutic potentials of such compounds have not been defined.
In pregnant rats infected with P. berghei, as in the case of human placental malaria, IRBCs and monocytes accumulate in high density in the placenta, causing severe placental pathology (12)(13)(14). However, the molecular interactions involved in the observed IRBC sequestration in the rat placenta are not known. As a part of our effort to determine whether pregnant rats can be useful as a model for studying pregnancy-specific malaria, we studied, in detail, the extracellular CSPG of the rat placenta. Interestingly, we found that although the rat placenta synthesizes very low sulfated CSPG, this proteoglycan is totally unrelated to aggrecan, and the sulfate content of the CS chains of rat placental CSPG is much more lower than that of the low sulfated human placental CSPG. The core protein of the rat placental CSPG is SSP, which is expressed in a pregnancy stagespecific manner (15). The expression of similarly low sulfated CS chains both in humans and rats, although the core proteins of CSPGs from these species are entirely different, points to important biological functions for the unusually low sulfated CS chains of placentas. Here, we report the structural characterization and IRBC-binding properties of the rat placental SSP-CSPG.
Isolation of Extracellular Proteoglycans from Rat Placenta-The procedures used were similar to those previously described for the isolation of CSPGs from the human term placentas (7,9). All procedures were performed on ice or in a cold room at 4°C. Placentas (each weighing 0.6 -0.7 g) from 20-or 21-day pregnant Wistar rats were collected by surgical dissection after anesthetizing the animals. Placentas from several rats were cut into small pieces, and the combined tissue (28 g) was extracted with 50 ml of PBS, pH 7.2, containing 10 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride, 0.1 mM TLCK, 0.25 mM TPCK, 1 mM benzamidine, and 0.1 mM N-ethylmaleimide by stirring with a magnetic stirrer for 4 h. The suspension was centrifuged at 10,000 rpm for 30 min in a RC2-B Sorvall centrifuge using SS-34 rotor. The clear supernatant was collected, and the tissue pellet was extracted two more times with 40 ml each time. The combined extract (ϳ120 ml) was passed through a DEAE-Sephacel column (1.5 ϫ 25 cm) equilibrated with PBS, pH 7.2, and the column was washed successively with 25 mM Tris-HCl, 150 mM NaCl, 10 mM EDTA, pH 8.0, and 20 mM NaOAc, 150 mM NaCl, pH 5.5, until absorption of the effluent at 280 nm was at background level. The bound proteoglycan was eluted with a linear gradient of 0.15-0.75 M NaCl in 50 mM NaOAc, 4 M urea, pH 5.5. Fractions (2 ml) were collected, absorption at 260 and 280 nm was measured, and aliquots analyzed for uronic acid content (16). The uronic acid-containing fractions were combined, dialyzed, and lyophilized. The yield of crude proteoglycan was 16 mg.
Purification of the CSPG by Cesium Bromide Density Gradient Centrifugation-The proteoglycan preparation from the DEAE-Sephacel chromatography of the placental extract was dissolved (2 mg/ml) in 25 mM sodium phosphate, pH 7.2, containing 50 mM NaCl, 0.02% NaN 3 , 4 M GdnHCl, and 42% (w/w) CsBr. The solutions were centrifuged (8 ml/tube) in a Beckman 50 TI rotor at 44,000 rpm for 65 h at 14°C (17). Gradients were fractionated from the bottom of the tubes into 15 equal portions, absorption at 260 and 280 nm was measured, and uronic acid contents were determined (16). The uronic acid-positive fractions were pooled, dialyzed, and lyophilized; yield was 7.5 mg.
Purification of the CSPG by Size Exclusion Chromatography-The proteoglycan partially purified by CsBr density centrifugation was chromatographed on a column of Sepharose CL-6B (2 ϫ 65 cm) in 50 mM NaOAc, 150 mM NaCl, pH 6.0, containing 4 M GdnHCl. Fractions (2 ml) were collected and monitored for proteins by measuring absorption at 280 nm, and aliquots were analyzed for uronic acid contents (16); yield was 4.9 mg.
Isolation of the CS Chains of Rat Placental CSPG-The purified CSPG (0.6 mg) was dissolved in 0.5 ml of 0.1 M NaOH, 1 M NaBH 4 and incubated at 45°C for 24 h under nitrogen (18). The solution was cooled in an ice bath, neutralized with 1 M HOAc, and dried in a rotary evaporator. Boric acid was removed by repeated evaporation with 0.1% acetic acid in methanol. The residue was dissolved in 0.2 M NaCl and chromatographed on a Sepharose CL-6B column (1 ϫ 49 cm) in 0.2 M NaCl. Fractions (0.67 ml) were collected, and aliquots were assayed for uronic acid content (16). Fractions containing the CS chains were combined, dialyzed (molecular weight cut-off 3,500) against distilled water, and lyophilized.
Carbohydrate Composition Analysis-The purified proteoglycan or the CS chains (5-10 g each) released by alkaline ␤-elimination of the CSPG were hydrolyzed with 400 l of 4 M HCl at 100°C for 4 h, and the hydrolysates were dried in a SpeedVac and analyzed for hexosamines. For analysis of neutral sugar in the core proteins, the purified CSPG was treated with protease-free chondroitinase ABC (see below) and analyzed by SDS-PAGE using 15% gels, and the protein bands in the gels were transferred onto PVDF membranes, stained for 30 s with 0.1% Coomassie Blue in 40% MeOH and 1% AcOH, and destained with 50% MeOH. The membrane portions corresponding to the protein bands were cut into pieces, suspended in 400 l of 2.5 M trifluoroacetic acid and hydrolyzed at 100°C for 5 h, and the hydrolysates were dried as above. The hydrolysates from both procedures were analyzed on a CarboPac PA1 high pH anion exchange column (4 ϫ 250 mm) using a Dionex BioLC HPLC system with a pulsed amperometric detector (19).
The elution was performed with 20 mM sodium hydroxide, and the response factors for the monosaccharides were determined using standard sugar solutions.
Disaccharide Compositional Analysis of GAGs-The GAGs (10 -15 g), obtained by the treatment of the purified proteoglycan with 0.1 M NaOH, 1 M NaBH 4 , followed by Sepharose CL-6B chromatography, were digested with chondroitinase ABC (5 milliunits) as above. The released unsaturated disaccharides were analyzed on a 4.6 ϫ 250-mm amine-bonded silica PA03 column (YMC Inc., Milford, MA), using a Waters 600E HPLC system (Milford, MA), and eluted with a linear gradient of 16 -530 mM NaH 2 PO 4 over 70 min at room temperature at a flow rate of 1 ml/min (22). The elution of disaccharides was measured by recording the absorption at 232 nm with a Waters 484 variable wavelength UV detector. The data were processed with the Millennium 2010 chromatography manager using NEC PowerMate 433 data processing system.
Analysis of the CSPG Core Protein by SDS-PAGE and Western Blotting-The purified CSPG (100 g) was treated with protease-free chondroitinase ABC (20 milliunits; protease inhibitors were also added to the incubation mixture as a precaution) in 100 l of 100 mM Tris-HCl, 30 mM NaOAc, pH 8.0, at 37°C for 6 h (20). The enzyme digest corresponding to 20 g of CSPG and 30 g of untreated CSPG was electrophoresed on 4 -15% polyacrylamide gradient gels under reducing conditions using 2-mercaptoethanol. The gels were stained sequentially with Coomassie Blue followed by Alcian Blue. For Western blot analysis, the protein bands (corresponding to 6 g of CSPG) on gels were transferred onto PVDF membranes, and the membranes blocked with 1% BSA in 50 mM Tris-HCl, 150 mM NaCl, pH 8.0, containing 0.1% Tween 20. The membranes were incubated with 1:1000 diluted rabbit anti-SSP antiserum, and the bound antibodies were detected using alkaline phosphatase-conjugated goat anti-rabbit IgG secondary antibodies and nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate color-developing reagent.
NH 2 -terminal Sequencing of the CSPG Core Proteins-The core proteins released from the purified CSPG (100 g) by protease-free chondroitinase ABC (20 milliunits) were electrophoresed (20 g CSPG/well) on 15% gels under reducing conditions. The protein bands on gels were transferred onto PVDF membranes using 10 mM 3-(cyclohexylamino)propane sulfonic acid buffer, pH 11.0, containing 10% methanol. The membranes were stained with Ponceau S, and the protein bands were cut out and sequenced by the Edman degradation method at the protein sequencing facility of the Penn State University College of Medicine using Applied Biosystems Procise 491 peptide sequencer.

Analysis of Nonglycosylated and CSPG Form of SSP in Rat
Placenta-The isotonic buffer extracts of rat placentas were analyzed on 15% polyacrylamide minigels before and after treatment with protease-free chondroitinase ABC, and the protein bands in the gels were transferred onto PVDF membranes. The membranes were treated with 1:1000-diluted rabbit anti-SSP antiserum followed by 1:2000 diluted horseradish peroxidase-conjugated goat anti-rabbit IgG. The bound secondary antibody was detected using chemiluminescence substrate.
Other Analytical Procedures-The uronic acid contents were determined by the Dische method (16). Protein contents were estimated by using the micro-BCA protein estimation kit from Pierce (23).
IRBC Adherence Assay-The assay was performed as described previously (24). Briefly, the purified rat placental CSPGs at various concentrations were coated onto plastic Petri dishes as 0.4-cm circular spots, blocked with 2% BSA in PBS, pH 7.2, and then overlaid with a 2% suspension of parasite culture in PBS, pH 7.2. The unbound IRBCs and RBCs were washed with PBS, and the bound IRBCs were fixed with 2% glutaraldehyde. The bound IRBCs were stained with Giemsa and examined under the light microscope.
IRBC Adherence Inhibition Assay-Solutions (1 g/ml) of the purified rat placental CSPG were coated onto plastic plates as 4-mm circular spots, and the spots were blocked with 2% BSA. The IRBC suspensions in PBS, pH 7.2, were preincubated with the indicated amounts of partially sulfated C4S (40% 4-sulfate, 59% nonsulfated, and 1% 6-sulfate), C6S, chondroitin, or hyaluronic acid at room temperature for 30 min with occasional mixing. The cell suspensions were overlaid onto the CSPGcoated spots and allowed to stand at room temperature for 30 min. After washing the unbound cells, the bound cells were fixed, stained, and counted by light microscopy.

RESULTS
Isolation and Purification of the Extracellular Proteoglycan of Rat Placenta-DEAE-Sephacel chromatography of the isotonic buffer extracts of the placentas from 20-or 21-day-old pregnant rats using a 0.15-0.75 M NaCl gradient eluted the uronic acid-containing material (proteoglycan) as a single homogeneous peak at 0.35 M NaCl (Fig. 1). Final elution of the column with 1.2 M NaCl did not elute any uronic acid-containing material (not shown). Upon CsBr gradient centrifugation, the proteoglycan was fractionated as a broad peak to the middle of the density gradient, separating from most of the associated proteins, which remained at the top of the gradient (Fig. 2). Upon further purification by size exclusion chromatography on Sepharose CL-6B column, the proteoglycan was eluted as a single but somewhat asymmetrical peak (K av ϭ 0.67) with an estimated average molecular weight of ϳ120,000 (Fig. 3). The yield and composition of the proteoglycan is given in Table 1. The presence of protein, uronic acid, and hexosamine in high amounts is consistent with the proteoglycan nature of the purified material from rat placentas.
Characterization of the GAG Chains of Rat Placental Blood Space Proteoglycan-The carbohydrate compositional analysis revealed that the purified proteoglycan contains Ͼ99% GalN and a trace amount of GlcN (Table 1). Upon treatment with chondroitinase ABC or chondroitinase AC II, the GAG chains were quantitatively converted into disaccharides (not shown). These results indicated that the proteoglycan is a CSPG. The CSPG was subjected to alkaline ␤-elimination using NaOH/NaBH 4 , and the released CS chains were purified by gel filtration on a Sepharose CL-6B column calibrated with chondroitin sulfates of known molecular weight (Fig. 4). The CS chains of the CSPG were eluted as a single symmetrical peak with an estimated molecular weight of ϳ40,000. Digestion of the CSPG with chondroitinase ABC and HPLC analysis of the disaccharides formed showed that the CS chains consist of 2% 4-sulfated and 98% nonsulfated disaccharides, demonstrating  Purification of rat placental CSPG by CsBr density gradient centrifugation. The rat placental CSPG, isolated by DEAE-Sephacel chromatography (Fig. 1), was dissolved in 25 mM sodium phosphate, pH 7.2, containing 4 M GdnHCl, 0.02% sodium azide, and 42% CsBr. The solutions were centrifuged in a Beckman 50 TI rotor at 44,000 rpm for 65 h at 14°C. The gradients were collected into 15 equal fractions from the bottom of the tube, and the aliquots were monitored for uronic acid (F) and protein (E). The CSPG-containing fractions were pooled as indicated by the horizontal bar.   that the CS chains of the CSPG are extremely low sulfated, resembling nonsulfated chondroitin chains ( Table 1). Characterization of Core Proteins of the Rat Placental CSPG-Upon SDS-PAGE analysis, the purified CSPG was electrophoresed as two partially separated broad bands with mobility corresponding to molecular weights of 60,000 -80,000 (average ϳ70,000) and 80,000 -180,000 (ϳ130,000) (Fig. 5A). The core proteins released by the treatment of the purified CSPG with chondroitinase ABC were electrophoresed as a doublet corresponding to molecular mass of ϳ17 and ϳ18 kDa (Fig.  5A). The relative abundance of the 17-and 18-kDa core protein bands in gels, as assessed by the densitometric scan, was ϳ57 and ϳ43%, respectively (Fig. 5B). The NH 2 -terminal sequence analysis of the two core protein bands on PVDF membranes revealed, in each case, an AILPDTTLYAELQEQ sequence for the first 15 amino acids. These results indicated that the 17-and 18-kDa doublets of the rat placental CSPG represent a single protein but differ either in posttranslational modifications or differentially processed at the COOH terminus.
Blast search analysis of the NCBI protein data base indicated that the NH 2 -terminal amino acid sequence of the core protein of the rat placental CSPG is identical to the NH 2 -terminal sequences of the rat placental SSP (GenBank TM accession number AB009890), which has been shown to be a 107-amino acid secretory protein with the theoretically calculated molecular weight of 12,210 (15) (Fig. 6). Iwatsuki et al. (15) have previously shown that SSP is expressed specifically during the late stage of the placental development. To confirm whether the core protein of the rat placental CSPG is SSP, the CSPG was treated with chondroitinase ABC, and the released core proteins were analyzed by Western blotting using anti-rabbit antiserum raised against the recombinant SSP expressed in E. coli. The 17-and 18-kDa core protein bands were specifically reactive to the anti-SSP polyclonal antibodies (see Fig. 5C), demonstrating that both core proteins of the CSPG represent rat placental SSP.
SSP contains two potential GAG attachment sites, one at Ser 74 and the other at Ser 96 (see Fig. 6). Therefore, we reasoned that the two distinct SSP-CSPG species, 60,000 -80,000 and 80,000 -180,000 (see Fig. 5A), observed upon SDS-PAGE represent SSP bearing one and two CS chains, respectively, and that the 17-and 18-kDa core proteins of SSP-CSPG correspond to SSP carrying, respectively, one and two core carbohydrate moieties. The ϳ1-kDa difference in molecular mass between 17-and 18-kDa core proteins of SSP-CSPG probably corresponds to the difference in the amount of sugars that remain at the CS attachment sites on the core proteins after treatment of the CSPG with chondroitinase ABC. To determine whether this is the case, the core proteins released by chondroitinase ABC were separated by SDS-PAGE and blotted onto PVDF membranes, and the carbohydrate composition of protein bands was determined ( Table 2). The results indicated that the 18-kDa core protein band has about twice the amount of galactose and xylose per mol of protein compared with those present in 1 mol of the 17-kDa protein band. Since the rat CSPG has no N-glycans, the difference in the amounts of galactose to xylose in the 18-and 17-kDa core proteins must correspond to the difference in the number of CS chains attached to the core proteins. It is known that the common core glycan moiety at the GAG attachment regions of the proteoglycans is a tetrasaccharide consisting of GlcA-Gal-Gal-Xyl and that chondroitinase ABC treatment of the CSPGs removes all of the GAG chain disaccharide repeat moieties except the one that is attached directly to the tetrasaccharide glycan core. Since the residual disaccharide moiety remaining on the core glycan moiety is present as an unsaturated stub, and because the GAG chains of rat SSP-CSPG are predominantly nonsulfated, the structure of the residual glycan left on the core protein after chondroitinase ABC treatment probably corresponds to ⌬GlcA-GalNAc-GlcA-Gal-Gal-Xyl. The calculated mass of the GAG chain core glycan with one attached unsaturated nonsulfated disaccharide stub is 1017 Da. This value is comparable with the observed difference in the molecular mass of the 17-and 18-kDa core proteins released from SSP-CSPG. From these results, it is  likely that the 60,000 -80,000 and 80,000 -180,000 CSPG species observed on SDS-PAGE (see Fig. 5A) represent SSP-CSPG with one and two CS chains, respectively. Analysis of the Relative Levels of Free and CSPG Form of SSP-Analysis of the PBS extract of rat placentas by Western blotting indicated the presence of a protein band at a mobility slightly faster than that of the 17-kDa core protein band and a diffused high molecular mass, chondroitinase ABCsusceptible proteoglycan band (Fig. 7). These bands correspond, respectively, to SSP lacking the CS chains and SSP-CSPG. Western blotting of the placental extract after treatment with chondroitinase ABC showed the presence of 17-and 18-kDa SSP bands (not well resolved, probably because high levels of other components in the total placental extract interfering with the electrophoretic mobility) with concomitant abolition of high molecular weight CSPG band (Fig. 7). These data taken together indicate that SSP is present predominantly in the CSPG form. Based on the intensity of the SSP bands in Western blots, before and after treatment with chondroitinase ABC, SSP without the GAG chains and the CSPG form of SSP are present in the approximate proportions of 10 and 90%, respectively.
Adherence of P. falciparum-IRBCs to the Purified Rat Placental CSPG-To determine whether the purified rat placental CSPG binds IRBCs, the CSPG was coated onto plastic plates at different concentrations and tested with IRBCs selected for binding to human placental CSPGs (24). Despite the presence of very low level of sulfate in the chondroitin chains, the rat placental CSPG could efficiently bind IRBCs in a concentrationdependent manner similar to IRBC binding to human placental aggrecan CSPG (Fig. 8A) (7). The CSPG exhibited a significant level of IRBC adherence at coating concentrations of 50 -500 ng/ml and a saturated level of binding at the coating concentration of 1 g/ml. The IRBC binding was totally abolished when the CSPG-coated spots were treated with chondroitinase ABC (not shown). The binding of IRBCs to the CSPG-coated plates was inhibited efficiently by partially sulfated C4S but not by chondroitin, C6S, hyaluronic acid, or heparin, indicating that the adhesion is C4S-specific and that the IRBCs bind to the chondroitin sulfate chains of the SSP-CSPG (Fig. 8B) (data not shown).
The IRBC-binding strength of the rat placental CSPG was assessed by measuring the inhibition of IRBC binding by a C4S containing 40% 4-sulfated disaccharide moieties. The C4S could inhibit the binding of IRBCs to rat CSPG in a dose-dependent manner with an IC 50 of 1.25 g/ml C4S (Fig. 8B).

DISCUSSION
This paper describes the biochemical characterization of a novel uniquely low-sulfated secretory CSPG of rat placenta. The CSPG can be extracted readily by isotonic buffer from rat placentas, and, as indicated by the SDS-PAGE analysis, it consists of two distinct species with electrophoretic mobility corresponding to the molecular masses of ϳ70,000 and ϳ130,000 Da. The CS chains of the rat CSPGs are extremely low sulfated; on an average, two disaccharide moieties of the CS chains per 100 disaccharides are sulfated, and the sulfation is exclusively at C-4. The occurrence of very low sulfated CSPGs in animal tissues is very rare. Previously, only the aggrecan type CSPG of the human placental intervillous space and a subset of bovine corneal decorin CSPG have been identified as very low sulfated proteoglycans with ϳ8 and ϳ15%, respectively, of the disaccharide moieties of the CS chains being 4-sulfated (9,26). Compared with the CS chains of human placental and bovine corneal low sulfated CSPGs, the degree of sulfation in the CS chains of rat placental CSPG is even lower. Thus, the rat placental CSPG is the most undersulfated among the CSPGs of various animal tissues that have been characterized to date. The purified rat placental CSPG was treated with chondroitinase ABC, and the released core proteins were electrophoresed on polyacrylamide gels. Either the gels were stained with Coomassie Blue and the intensity of bands was measured, or the core proteins were transferred onto PVDF membranes and analyzed for the sugar composition.

Core protein band a
Relative abundance of core protein bands a By densitometric scanning of core protein bands on gels (see Fig. 5B). b By HPLC of 2.5 M trifluoroacetic acid hydrolysates of the core protein bands. c The amount of Gal in 18-and 17-kDa core proteins was normalized to 2 and 1 mol of Xyl/mol of the respective core proteins.

FIGURE 7. Assessment of relative level of free and CSPG form of SSP by
Western blotting. The total PBS extract of the rat placentas at 14, 17, and 21 days of pregnancy was electrophoresed using 15% polyacrylamide gel before and after treatment with chondroitinase ABC, and the protein bands in the gels were transferred onto PVDF membranes, blocked with 1% BSA, and detected with rabbit antiserum against SSP using HRP-conjugated goat antirabbit IgG secondary antibody. Shown is the analysis of the PBS extract from placentas at the 17th day of pregnancy. Lane 1, untreated rat placental PBS extract; lane 2, chondroitinase ABC-treated rat placental PBS extract. Equal amounts of placental extracts were used for Western blotting before and after chondroitiase ABC treatment. The PBS extracts from placentas at 14 and 21 days of pregnancy were also analyzed, and the results (not shown) were similar to those of the placenta at the 17th day of pregnancy.
A previous study from our laboratory has demonstrated that the extracellular low sulfated CSPG of the human placental intervillous space is a high molecular weight (0.57-1 ϫ 10 6 ) aggrecan family proteoglycan with 6 -8 low sulfated CS chains of ϳ60 kDa per molecule (7,9). In this study, interestingly, we found that the low sulfated rat placental CSPG is completely different, and actually it is SSP bearing one or two chondroitin sulfate chains of ϳ40 kDa. The following evidence supports this finding. (i) The rat placental CSPG consists of two distinct core proteins with a molecular mass of ϳ17 and ϳ18 kDa as revealed by SDS-PAGE analysis before and after treatment with chondroitinase ABC. The NH 2 -terminal amino acid sequence of each core protein is identical to that of rat SSP. Based on the 40-kDa size of the CS chains, it appears likely that the 17-and 18-kDa core proteins are derived from ϳ70and ϳ130-kDa CSPG species bearing, respectively, one and two CS chains. (ii) The exclusive presence of galactose and xylose as neutral sugars, in a molar ratio of ϳ2:1 in both 17-and 18-kDa core proteins strongly suggests that these sugars represent the core sugar residues at the linkage region of the GAG chains. Further, the overall stoichiometry of these two sugars in the core pro-teins (i.e. ϳ1.8 mol of galactose and ϳ1 mol of xylose/mol of ϳ17-kDa core protein and ϳ3.7 mol of galactose and ϳ2 mol of xylose/mol of 18-kDa core protein) confirms that the 17-and 18-kDa core proteins are substituted with one and two CS chains, respectively. This conclusion also agrees with the observed ϳ1-kDa difference in the molecular mass of the two core proteins released by the chondroitinase ABC-treatment of the rat placental CSPG. As described under "Results," the observed ϳ1-kDa mass difference between 17-and 18-kDa core proteins is due to the presence of one and two ⌬GlcA-GalNAc-GlcA-Gal-Gal-Xyl-carbohydrate moieties, respectively (where ⌬GlcA is 4,5-unsaturated glucuronyl residue), in the former and the latter core proteins released by chondroitinase ABC. (iii) Based on the amino acid sequence of rat SSP (see Fig. 6), it is logical to conclude that the ϳ70-kDa CSPG species represent SSP with one CS chain at either Ser 74 or Ser 96 , and the ϳ130-kDa CSPG species represent SSP carrying CS chains on both Ser 74 and Ser 96 . (iv) Finally, the deduced amino acid sequence of rat SSP lacks potential N-glycosylation sites. Consistent with this information, the carbohydrate compositional analysis showed only a trace amount of GlcNAc in the intact CSPG as well as in the 17-and 18-kDa core proteins obtained by chondroitinase ABC treatment of the CSPG. Thus, the low sulfated rat placental CSPG is the proteoglycan form of SSP carrying one and two chondroitin sulfate chains.
Previously, SSP has been reported as nonglycosylated secretory molecule. Although this is true with regard to N-glycosylation, the results of this study clearly show that SSP occurs predominantly in the CSPG form and that only a minor portion of SSP is present without GAG chain substitution. In rat placentas, the expression of SSP has been reported to start on the 14th day of the pregnancy with the maximum level of expression at day 16 and remains at this level until the term. Our results show that throughout its expression from day 14 until the term, SSP is present predominantly in the CSPG form. The expression of SSP appears to correspond to the period of full establishment of spongiotrophoblast layer at the maternal and fetal interface between the trophoblast giant cell layer of the junctional zones and the labyrinth (15). The strategic location of SSP-CSPG expression in the placenta and the pregnancy stage-and placental developmental stage-specific expression of SSP-CSPG are likely to have important biological relevance.
Placenta produces pregnancy-specific hormones and a multitude of chemokines, growth factors, cytokines, and other factors that are essential for the maintenance and progression of pregnancy, modulation of immune and endrocrine functions, development of the fetus, and delivery (27)(28)(29)(30)(31). For example, prostaglandins and endothelins produced in response to certain cytokines have been reported to function as the key myometrial contratile agents essential for delivery (27,32). Placental hormones and growth factors are likely to enter fetal compartments, where they promote fetal development. Thus, based on the ability of GAG chains to interact with a variety of cytokines and other factors (33), we speculate that the low sulfated SSP-CSPG, secreted by the spongiotrophoblast cells to the maternal blood that baths the labyrinth, is involved in reversibly capturing and mobilizing cytokines, hormones and growth factors produced by placenta, creating concentrated FIGURE 8. Analysis of P. falciparum IRBC binding to rat placental SSP-CSPG. A, concentration-dependent binding of IRBCs to rat placental CSPG. Plastic Petri dishes were coated as circular spots using the indicated concentrations of the CSPG in PBS, blocked with 2% BSA, and overlaid with parasite cultures. After washing the unbound cells, the bound IRBCs were fixed with 2% glutaraldehyde, stained with 1% Giemsa, and counted under light microscope. B, inhibition of IRBC adhesion to the rat placental CSPG by GAGs. IRBCs preincubated with the indicated concentrations of different GAGs were overlaid onto CSPG-coated spots, and the bound cells were measured as above. F, C4S with 40% 4-sulfate; E, C6S; OE, chondroitin; f, hyaluronic acid. Both adherence and inhibition assays were carried out three times each in duplicate, and the results are presented as mean Ϯ S.E. (n ϭ 6). OCTOBER 27, 2006 • VOLUME 281 • NUMBER 43 pools of these molecules in the maternal and fetal interface and modulating the functions of placenta. The low sulfated CSPGs are ideally suited for this purpose, because the low net negative charge allows capturing and mobilizing nutrients, cytokines, hormones, and growth factors by low affinity interactions and nonetheless readily release the bound factors for their functions. It is anticipated that the results reported here will facilitate efforts toward understanding the functional role of SSP-CSPG in the placenta.

Rat Spongiotrophoblast-specific CSPG
Recently, it has been found that P. chabaudi AS also sequesters in the parasite-infected mouse placenta, and this system has been suggested as a suitable mouse model for studying maternal malaria (34). P. chabaudi AS has been reported to have binding characteristics similar to those of P. falciparum (34). If the low sulfated SSP-CSPG is found to have a role in the sequestration of IRBCs in the placentas of infected rodents, then pregnant rats would be useful as a model to study the therapeutic benefits of C4S oligosaccharides and/or their mimetics. Further in vivo studies are required to validate the model.