Evaluating specific adhesion of Plasmodium falciparum-infected erythrocytes to immobilised hyaluronic acid with comparison to binding of mammalian cells

International Journal for Parasitology 32 (2002) 1245–1252 www.parasitology-online.com

Evaluating specific adhesion of Plasmodium falciparum-infected erythrocytes to immobilised hyaluronic acid with comparison to binding of mammalian cells James G. Beeson*, Stephen J. Rogerson, Graham V. Brown Department of Medicine, University of Melbourne, Royal Melbourne Hospital, Melbourne, Vic. 3050, Australia Received 17 April 2002; received in revised form 16 May 2002; accepted 16 May 2002

Abstract A feature of infection with Plasmodium falciparum is the ability of parasite-infected erythrocytes to adhere to vascular endothelial cells and accumulate in vital organs, associated with severe clinical disease. Hyaluronic acid was recently identified as a receptor for adhesion and has been implicated in mediating the accumulation of parasites in the placenta. Here, we report in vitro assays to measure specific adhesion of infected erythrocytes to hyaluronic acid that is distinct from binding to chondroitin sulphate A, another glycosaminoglycan implicated as a receptor in placental malaria. In this study, specific adhesion of mature stage infected erythrocytes to hyaluronic acid of high purity immobilised on plastic surfaces was abolished by pre-treating hyaluronic acid with a specific hyaluronate lyase from Streptomyces, whereas the same treatment of chondroitin sulphate A had little effect. Adhesion to hyaluronic acid could not be explained by the presence of chondroitin sulphate A or other glycosaminoglycans in the hyaluronic acid preparations. Chinese hamster ovary cells bound in a similar manner in the assays and confirmed that hyaluronic acid was appropriately immobilised for cell adhesion. In contrast to parasites, these cells did not adhere to chondroitin sulphate A. The adsorption of hyaluronic acid onto plastic surfaces was also confirmed by the use of a specific hyaluronic acid-binding protein. Fixing cells with glutaraldehyde at the completion of adhesion assays reduced the number of parasites remaining adherent to hyaluronic acid, but not to chondroitin sulphate A or CD36. These findings have important implications for understanding and evaluating interactions between P. falciparum and hyaluronic acid that may be involved in disease pathogenesis. q 2002 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. Keywords: Plasmodium falciparum; Cell adhesion; Hyaluronic acid; Chondroitin sulphate; Placenta; Ovarian cell

1. Introduction The pathogenesis of malaria caused by Plasmodium falciparum involves the adhesion of parasite-infected erythrocytes to the vascular endothelium (reviewed in Beeson and Brown, 2002). Adhesion may contribute to parasite survival through avoidance of immune-mediated destruction and splenic clearance from the circulation. However, the accumulation or sequestration of large numbers of parasites in the vascular beds of organs such as the brain, lung, and placenta can lead to severe disease in the host, with substantial mortality. Infected erythrocytes can adhere to a range of receptors on the endothelial surface, such as CD36, intercellular adhesion molecule 1, and chondroitin sulphate A (Barnwell et al., 1989; Berendt et al., 1989; Ockenhouse et al., 1989; Rogerson et al., 1995). We have recently identified hyaluronic acid (or hyaluronan) as an additional recep-

tor for parasite adhesion, in the context of placental malaria (Beeson et al., 2000). Hyaluronic acid is a non-sulphated high molecular weight glycosaminoglycan composed of repeating disaccharide units of N-acetylglucosamine linked to glucuronic acid. It is abundant in the extracellular matrix of many tissues, but is also expressed on the surface of endothelial cells and syncytiotrophoblasts, which line the placental blood spaces (Matejevic et al., 2001; Mohamadzadeh et al., 1998; Sunderland et al., 1985). Infected erythrocytes can bind to immobilised purified hyaluronic acid, in static assays and under conditions of physiologically-relevant flow, and the minimal length of hyaluronic acid for parasite interaction appears to be 12 monosaccharide units (Beeson et al., 2000; Chai et al., 2001). Plasmodium falciparum-infected erythrocytes are phenotypically diverse and can undergo clonal antigenic variation (Beeson and Brown, 2002). As for adhesion to other receptors, parasite adhesion to hyaluronic acid

* Corresponding author. Tel.: 161-3-8344-6252; fax: 161-3-9347-1863. E-mail address: [email protected] (J.G. Beeson). 0020-7519/02/$20.00 q 2002 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved. PII: S 0020-751 9(02)00097-8

1246

J.G. Beeson et al. / International Journal for Parasitology 32 (2002) 1245–1252

is variant-specific, being a property of particular isolates (Beeson et al., 2000). We found that most isolates from infected placentas contained parasites that could bind to immobilised hyaluronic acid in vitro, suggesting that hyaluronic acid can act as a receptor for parasite adhesion in the placenta (Beeson et al., 2000). Isolates from the peripheral blood of pregnant women or children bound hyaluronic acid less frequently and at lower levels. Chondroitin sulphate A has also been implicated as a receptor for parasite adhesion in the placenta (Beeson et al., 1999; Fried and Duffy, 1996; Maubert et al., 1997), and there is considerable overlap between adhesion to chondroitin sulphate A and hyaluronic acid by different isolates (Beeson et al., 2000). Many isolates were found to bind both receptors, however, some bound either hyaluronic acid or chondroitin sulphate A (Beeson et al., 2000). Furthermore, trypsin cleavage of proteins on the infected erythrocyte surface had a differential effect on adhesion to hyaluronic acid and chondroitin sulphate A. Together, these results suggest separate, but overlapping, parasite specificities for adhesion to hyaluronic acid and chondroitin sulphate A. Here, we describe in vitro adhesion of P. falciparuminfected erythrocytes to hyaluronic acid immobilised on plastic surfaces that is distinct from binding to chondroitin sulphate A. The specificity of parasite adhesion is established by using hyaluronic acid of high purity and enzymatic degradation of receptors that could form the basis of a simple protocol for use in in vitro assays. Furthermore, we have performed comparative studies with hyaluronic acidbinding ovarian cells to further evaluate the nature of parasite adhesion. 2. Methods 2.1. Plasmodium falciparum isolates and culture Isolate CS2 was derived from a clone of ItG selected for adhesion to Chinese hamster ovary (CHO) cells naturally expressing chondroitin sulphate A, followed by selection for adhesion to immobilised chondroitin sulphate A (Cooke et al., 1996; Rogerson et al., 1995). FCR3selCSA was generated from isolate FCR3 by four cycles of selection for adhesion to immobilised chondroitin sulphate A. Isolates were cultured in vitro, as previously described (Beeson et al., 1999), in RPMI-HEPES medium, pH 7.4, supplemented with hypoxanthine 50 mg/ml, NaHCO3 25 mM, gentamicin 25 mg/ml and 10% v/v pooled human serum from donors resident in Australia, and maintained an atmosphere of 1% O2, 4% CO2, 95% N2, 37 8C. Parasite cultures were free of Mycoplasma, as determined by PCR. 2.2. Cell adhesion assays The following chemicals were used in cytoadherence assays (all glycosaminoglycans and polysulphated sugars

were obtained from Sigma, unless otherwise stated): hyaluronic acid from bovine vitreous humour or human umbilical cords (potassium salt); hyaluronic acid from Streptococcus (Calbiochem, SanDiego); chondroitin sulphate A from bovine trachea or porcine rib cartilage; chondroitin sulphate C from shark cartilage; heparan sulphate from bovine kidney; colominic acid (a sialic acid polymer); chondroitin sulphate B from intestinal mucosa; dextran sulphate with an average Mr of 500 kDa; and CD36 purified from human platelets (a gift of M. Berndt, Baker Medical Research Institute, Melbourne). In all static adhesion assays, hyaluronic acid from bovine vitreous and chondroitin sulphate A from bovine trachea were used as the receptors unless stated otherwise. Saturating coating concentrations were used in assays, as determined by titration experiments. These were 50-100 mg/ml for hyaluronic acid from bovine vitreous, and 5–10 mg/ml for hyaluronic acid from human umbilical cords and for chondroitin sulphate A (from bovine trachea). All other carbohydrates were tested at 100 mg/ml. Adhesion of P. falciparum was generally tested using trophozoite-infected erythrocytes at 3–8% parasitaemia, 2% haematocrit, as previously described (Beeson et al., 1998, 1999). Briefly, purified receptors diluted in PBS were adsorbed onto the surface of plastic petri dishes (Falcon 1058; Becton Dickinson) for 18–24 h at 4 8C and spots subsequently blocked with 1% BSA in PBS. Receptor spots were overlaid with parasites suspended in RPMIHEPES pH6.8 with 10% pooled human serum, and adhesion allowed to occur for 30 min at 37 8C. Unbound cells were removed by gentle washing with RPMI-HEPES (pH 6.8) before bound cells were fixed with 2% glutaraldehyde (Unilab APS) in PBS, stained with Giemsa (BDH Merck) 10% in water, and counted by light microscopy. In some experiments, bound cells were counted prior to fixing with glutaraldehyde, on an inverted microscope using phase contrast and/or fluorescence. Glycosaminoglycans were tested at varying concentrations for inhibition of parasite adhesion to purified receptors by prior incubation with cell suspensions for 5–20 min before performing adhesion assays. Chinese hamster ovary cells (ATCC) were cultured in RPMI, pH 7.4, with 10% foetal calf serum, in an atmosphere of 5% CO2, 37 8C, prior to use in assays. Adhesion of CHO cells was tested in a similar manner to parasites, except cells were suspended in 0.5% BSA in PBS or RPMI and unbound cells were removed by gentle washing with PBS pH 7.2. 2.3. Enzymatic digestion of glycosaminoglycans Hyaluronic acid from bovine vitreous 500 mg/ml and chondroitin sulphate A 100 mg/ml were incubated with hyaluronate lyase from Streptomyces hyalurolyticus (Calbiochem) 5–25 turbidity reducing units/ml in PBS (pH 7.2) at 60 8C for 4 h or overnight. Additional experiments in which the enzyme and glycosaminoglycans were

J.G. Beeson et al. / International Journal for Parasitology 32 (2002) 1245–1252

incubated in acetate buffer 0.02 M (pH 5.0) were performed for comparison (Ohya and Kaneko, 1970). Control enzyme treatment, performed in parallel with hyaluronate lyase treatment, involved incubating hyaluronic acid or chondroitin sulphate A alone or with an equivalent amount of inactive protein (BSA) in place of the lyase. Digestion was stopped by heating to 100 8C for 15 min. Reaction mixtures were diluted in PBS (to 100 mg/ml for hyaluronic acid, and to 10 mg/ml for chondroitin sulphate A), spotted onto plastic petri dishes and allowed to adsorb for 24 h at 4 8C. Adhesion was then tested as described above. As an additional control, hyaluronic acid and chondroitin sulphate A were also mixed with lyase previously inactivated by repeated boiling prior to coating receptors. Lyase digests were also tested as inhibitors of parasite adhesion, as above.

1247

A from bovine trachea or porcine rib 20–100 mg/ml, or BSA 100 mg/ml in PBS. Wells were subsequently blocked with 1% BSA or 5% skimmed milk in PBS with 0.05% Tween for 1 h. Biotinylated hyaluronic acid binding protein (Calbiochem, CA, USA) was incubated at 10 mg/ml in PBS containing BSA for 1 h at 37 8C, washed with PBS– Tween 0.05%, and incubated with streptavidin–horseradish peroxidase diluted 1:1000 in PBS for 30 min. After washing with PBS, colour was developed with 2,2 0 -azino-bis(3ethylbenzathiazoline-6-sulphonic acid), and read by spectrophotometry. All incubations were performed at room temperature. BSA-coated and blank wells were used to determine levels of background absorbance.

3. Results 2.4. Binding of hyaluronic acid-binding protein to immobilised hyaluronic acid Polystyrene plastic 96-well plates (Falcon 3077, Becton Dickinson) were coated overnight at 4 8C with hyaluronic acid from bovine vitreous 100 mg/ml, chondroitin sulphate

3.1. Adhesion of mature pigmented P. falciparum-infected erythrocytes to immobilised hyaluronic acid of high purity Parasitised erythrocytes adhered to immobilised hyaluronic acid isolated from bovine vitreous humour at high levels

Fig. 1. (A) Adhesion of Plasmodium falciparum-infected erythrocytes to various carbohydrates coated onto plastic surfaces: chondroitin sulphate A, B, and C (CSA, CSB, CSC), hyaluronic acid (HA; isolated from bovine vitreous humour), heparan sulphate (HS), dextran sulphate (DexS), and colominic acid (ColA). (B) Lack of adhesion of P. falciparum to chondroitin sulphate A (from bovine trachea, CSA-BT, and porcine rib, CSA-PR) coated at a concentration (0.5 mg/ ml) reflecting the potential level of contamination in hyaluronic acid preparations that support high levels of adhesion. Hyaluronic acid from bovine vitreous humour (HA-BVH) was coated at 50 mg/ml, and hyaluronic acid from human umbilical cords (HA-HUC) at 5 mg/ml. (C) Binding of biotinylated hyaluronic acid binding protein to hyaluronic acid from bovine vitreous and chondroitin sulphate A coated on plastic surfaces. (D) Adhesion of P. falciparum to hyaluronic acid, chondroitin sulphate A, and CD36 before (grey bars) and after (black bars) fixation of bound cells with 2% glutaraldehyde, using isolate FCR3selCSA. All values represent mean ^ S.E.M., multiple experiments.

1248

J.G. Beeson et al. / International Journal for Parasitology 32 (2002) 1245–1252

ð.1400 infected erythrocytes/mm 2 at 7% parasitaemia using CS2) and adhesion was dependent on the coating concentration used, being maximum at 50 mg/ml. Adhesion was specific to mature pigmented parasites with little adhesion of ring forms and no adhesion of uninfected erythrocytes. CS2-infected erythrocytes also bound chondroitin sulphate A (purified from bovine trachea; adhesion maximal at a coating concentration of 10 mg/ml), whereas there was no specific adhesion to other polysaccharides such as chondroitin sulphate B or C, heparan sulphate, dextran sulphate, or colominic acid used at 100 mg/ml (Fig. 1a). Some preparations of hyaluronic acid contain co-purified chondroitin sulphate. In these experiments, and elsewhere (Beeson et al., 2000; Chai et al., 2001), we used hyaluronic acid isolated from bovine vitreous humour because it is reportedly free of chondroitin sulphates and other glycosaminoglycans (Chai et al., 2001; Grimshaw et al., 1994; Rowen et al., 1956). The batch of hyaluronic acid we used had , 0:2% chondroitin sulphate, analysed by NMR spectroscopy (Chai W., Kogelberg, H., personal communication). Parasites also bound at similar high levels to hyaluronic acid purified from human umbilical cords, which is known to contain some chondroitin sulphates (amount not quantified). When chondroitin sulphate A was coated onto plastic at a concentration reflecting potential levels of contamination present in hyaluronic acid, there was no adhesion. Hyaluronic acid from bovine vitreous and human umbilical cords support high level adhesion at coating concentrations of 50 and 5 mg/ml, respectively. Two types of chondroitin sulphate A were thus coated at a concentration of 0.5 mg/ml (representing the presence of up to 1% chondroitin sulphate A in hyaluronic acid from bovine vitreous or 10% in hyaluronic acid from umbilical cord) and tested together with the hyaluronic acid preparations for parasite adhesion. Parasites bound at high levels to both types of hyaluronic acid with little or no adhesion to chondroitin sulphate A, suggesting adhesion to hyaluronic acid was not due to the presence of chondroitin sulphate A (Fig. 1b). Similarly, chondroitin sulphates B and C contain significant amounts of chondroitin sulphate A (up to 30%; supplier’s information); however, neither preparation supported adhesion in our assays when coated at 100 mg/ ml (Fig. 1a). We next established that hyaluronic acid was immobilised on plastic surfaces at sufficient levels and with an appropriate conformation for biological interactions with the coating procedure we employed. A specific biotinylated hyaluronic acid binding protein (Mohamadzadeh et al., 1998; Tammi et al., 1998; Wang and Underhill, 1992) was used to probe hyaluronic acid and chondroitin sulphate A immobilised onto the surface of microtitre wells made from the same plastic as the petri dishes used in cell adhesion assays. The biotinylated probe bound to immobilised hyaluronic acid whereas there was little apparent binding to chondroitin sulphate A above background levels (Fig. 1c). Unexpectedly, we found that cell fixation with glutaral-

dehyde, prior to staining and counting, a standard procedure used in adhesion assays, substantially and variably reduced the numbers of parasites remaining bound to hyaluronic acid, but there was no apparent effect on adhesion to chondroitin sulphate A or CD36 (Fig. 1d). Adhesion to bovine vitreous hyaluronic acid measured after fixation was 42 ^ 3.7% lower (mean ^ S.E.M., two experiments) than

Fig. 2. (A) Adhesion of Plasmodium falciparum-infected erythrocytes to hyaluronic acid from bovine vitreous (HA; black bars), but not chondroitin sulphate A from bovine trachea (CSA; grey bars), was abolished by preincubation of receptors with hyaluronate lyase from Streptomyces at 5 or 10 units/ml. (B) Hyaluronic acid incubated with hyaluronate lyase lost its ability to inhibit the adhesion of P. falciparum to immobilised hyaluronic acid, compared to mock enzyme-treated hyaluronic acid. (C) Chinese hamster ovary cells (grey bars) and P. falciparum (black bars) bound to hyaluronic acid (from bovine vitreous humour) in a manner dependent on the coating concentration (measured at 10 and 100 mg/ml), and to hyaluronic acid from human umbilical cords (HA-HUC, 10 mg/ml), but not to lyase-treated hyaluronic acid. Only P. falciparum bound to immobilised chondroitin sulphate A in the same assays. Neither cell type bound to hyaluronic acid from Streptococcus (HA-Strep). All values represent mean ^ S.E.M., multiple experiments.

J.G. Beeson et al. / International Journal for Parasitology 32 (2002) 1245–1252

that observed prior to fixation using isolate FCR3selCSA and 24 ^ 6% (mean ^ S.E.M., two experiments) lower with CS2-infected erythrocytes. We confirmed that the majority ð.99%Þ of cells bound to hyaluronic acid, prior to fixation, were parasitised by labelling with ethidium bromide and counting adhesion using fluorescence.

3.2. Streptomyces hyaluronate lyase treatment inhibits adhesion of infected erythrocytes to hyaluronic acid but not chondroitin sulphate A Purified hyaluronic acid from bovine vitreous and chondroitin sulphate A preparations were incubated with hyaluronate lyase from Streptomyces, an enzyme that specifically degrades hyaluronic acid and does not cleave chondroitin sulphates (Ohya and Kaneko, 1970). Lyase treatment abolished parasite adhesion to hyaluronic acid (Fig. 2a); enzyme previously inactivated by repeated boiling or an inactive protein had no effect on adhesion. Furthermore, lyase-treated hyaluronic acid was unable to inhibit adhesion to immobilised receptor (Fig. 2b). By contrast, the same treatment of chondroitin sulphate A had little effect on its ability to support or inhibit adhesion (Fig. 2a and data not shown). Similarly, hyaluronate lyase treatment of hyaluronic acid abolished adhesion of CHO cells, described below (Fig. 2C).

3.3. Chinese hamster ovary cells adhere to immobilised hyaluronic acid in a similar manner to P. falciparuminfected erythrocytes Ovarian cells are known to adhere to hyaluronic acid immobilised on plastic surfaces (Catterall et al., 1997). CHO cells used in our assays bound immobilised hyaluronic acid from bovine vitreous or human umbilical cords in a manner dependent on the coating concentration, similar to P. falciparum (Fig. 2c). The relative binding of CHO cells to bovine vitreous hyaluronic acid coated at 10 and 100 mg/ml and umbilical cord hyaluronic acid coated at 10 mg/ml was 16, 57 and 100%, respectively. The relative binding of mature pigmented-infected erythrocytes showed the same pattern: 11, 73 and 100% for adhesion to bovine vitreous (10 and 100 mg/ml) and umbilical cord (10 mg/ml) hyaluronic acid, respectively. Hyaluronic acid from Streptococcus did not support adhesion of parasites or CHO cells when used as the receptor in adhesion assays. As was observed with P. falciparum, adhesion of CHO cells to bovine vitreous hyaluronic acid was abolished by incubation of the receptor with hyaluronate lyase from Streptomyces, consistent with specific degradation of hyaluronic acid by the enzyme. In contrast to infected erythrocytes, CHO cells did not adhere to immobilised chondroitin sulphate A. Cos cells (derived from monkey fibroblasts) also bound hyaluronic acid but not chondroitin sulphate A (data not shown).

1249

4. Discussion Inhibition of parasite adhesion to hyaluronic acid by treatment with hyaluronate lyase from Streptomyces provided clear evidence of a specific interaction between hyaluronic acid and P. falciparum, confirming that parasite adhesion did not represent binding to possible contaminants. This enzyme is specific to hyaluronic acid, and does not degrade chondroitin sulphates (Ohya and Kaneko, 1970; Valiyaveettil et al., 2001). Using the CS2 isolate, pre-treatment of hyaluronic acid completely inhibited parasite adhesion, whereas there was no effect of the same treatment on adhesion to chondroitin sulphate A. Consistent with this finding, incubation with hyaluronate lyase, which degrades hyaluronic acid to tetra- and hexasaccharide fragments (Ohya and Kaneko, 1970), also abolished the ability of hyaluronic acid to inhibit adhesion of infected erythrocytes to immobilised receptor. Enzyme-treated chondroitin sulphate A retained its inhibitory activity. Streptomyces hyaluronate lyase-treated hyaluronic acid (from bovine vitreous) could serve as a practically useful control in future studies of parasite adhesion to hyaluronic acid and chondroitin sulphate A. This can be prepared and tested in advance and used in conjunction with untreated hyaluronic acid in adhesion assays; parasite adhesion to immobilised untreated hyaluronic acid but not to lyase-treated hyaluronic acid would indicate specific adhesion. Results can be standardised if all isolates are tested against the same preparation of hyaluronic acid and its lyase-treated control. This approach would be particularly useful when testing large numbers of clinical isolates. High levels of parasite adhesion were observed with hyaluronic acid purified from bovine vitreous humour. Analysis of the preparation used here, and in other studies (Beeson et al., 2000; Chai et al., 2001), indicated the presence of , 0:2% chondroitin sulphate A by NMR spectroscopy, and no other sulphated glycosaminoglycan components were detected (Chai, W., Kogelberg, H., personal communication). This agrees with analyses of this hyaluronic acid by other investigators using different methods (Grimshaw et al., 1994; Rowen et al., 1956). One report (Valiyaveettil et al., 2001) suggested higher levels of chondroitin sulphates or other glycosaminoglycans are present in bovine vitreous hyaluronic acid and may reflect variability in purity of this preparation. Although some preparations of hyaluronic acid contain low levels of chondroitin sulphates, when chondroitin sulphate A was coated at concentrations representing the maximum level of potential contamination of hyaluronic acid it did not support parasite adhesion (Fig. 1b), and immobilised chondroitin sulphates B and C did not support adhesion (Fig. 1a) despite containing substantial amounts of chondroitin sulphate A. Specific adhesion of parasites to hyaluronic acid is supported by other observations, including inhibition of adhesion by structurallydefined oligosaccharide fragments, a differential effect of trypsin treatment on adhesion to hyaluronic acid and chon-

1250

J.G. Beeson et al. / International Journal for Parasitology 32 (2002) 1245–1252

droitin sulphate A, and differences in parasite adhesive characteristics (Beeson et al., 2000; Chai et al., 2001). We used CHO cells to validate our assays and draw comparisons with parasite adhesion as ovarian cells are known to bind hyaluronic acid (Catterall et al., 1997). The binding of CHO cells in the assays confirmed that hyaluronic acid was appropriately immobilised for cell adhesion. In contrast to parasitised erythrocytes, CHO cells did not adhere to chondroitin sulphate A. The adsorption of hyaluronic acid to plastic surfaces was further confirmed by the use of a specific probe, and is consistent with previously published studies using similar methods to immobilize hyaluronic acid (Catterall et al., 1997; DeGrendele et al., 1996; Legras et al., 1997; Parkar et al., 1998; Stamenkovic and Aruffo, 1994). These findings directly address recent concerns regarding the efficiency with which hyaluronic acid polysaccharides coat plastic surfaces (Valiyaveettil et al., 2001). Catterall et al. (1997) found significant differences in levels of ovarian cell adhesion to hyaluronic acid using different types of plastic plates, and different ovarian cell lines varied in their ability to bind hyaluronic acid. Fixing bound cells with glutaraldehyde prior to staining and counting, an established procedure in the field, had the unexpected effect of substantially reducing adhesion to hyaluronic acid, to varying degrees with different isolates, but not to other receptors, such as chondroitin sulphate A and CD36. Therefore, to reliably measure parasite adhesion relative to other receptors, bound cells may need to be counted prior to fixation, or alternative approaches found. This observation has implications for interpreting previously published results on adhesion of clinical isolates to hyaluronic acid (Beeson et al., 2000; Fried et al., 2000). The reported values may represent underestimates of actual adhesion that occurred to hyaluronic acid in vitro. Some clinical and laboratory-propagated isolates, such as CS2, contain parasites that can adhere to hyaluronic acid and chondroitin sulphate A. Other isolates appear to bind hyaluronic acid or chondroitin sulphate A rather than both receptors, consistent with separate parasite specificity for adhesion to each of hyaluronic acid and chondroitin sulphate A (Beeson et al., 2000). Other investigators tested a 3D7 isolate that could bind to chondroitin sulphate A and concluded there was no specific binding to hyaluronic acid, but that the isolate may bind to chondroitin sulphates present in hyaluronic acid isolated from human umbilical cords (Valiyaveettil et al., 2001). Of note, the 3D7 isolate did not bind in a specific manner to hyaluronic acid from bovine vitreous (uninfected erythrocytes, ring forms and pigmented trophozoites reportedly bound with equal propensity) (Valiyaveettil et al., 2001), in contrast to the apparently specific high-level adhesion to this source of hyaluronic acid we have observed in the present and previous studies (Beeson et al., 2000; Chai et al., 2001). If specific parasite binding is defined as adhesion of mature stage infected erythrocytes to bovine vitreous hyaluronic

acid, that is inhibited by Streptomyces hyaluronate lyase treatment, the 3D7 isolate of Valiyaveetil et al. would be regarded as binding to chondroitin sulphate A, but not hyaluronic acid, as reported for certain clinical isolates (Beeson et al., 2000), whereas CS2 and other isolates would be considered chondroitin sulphate A and hyaluronic acid binding isolates. By comparison, many isolates bind both CD36 and intracellular adhesion molecule-1 (ICAM1), whereas others bind CD36 and not ICAM-1 (Beeson et al., 1999; Chaiyaroj et al., 1996; Gardner et al., 1996; Udomsangpetch et al., 1997). Findings on parasite interactions are influenced by the source of the hyaluronic acid used in studies. We demonstrate here that hyaluronic acid from Streptococcus is unable to support adhesion of P. falciparum or CHO cells in standard assays suggesting it is not immobilised at sufficient levels or in an appropriate form for cellular interactions. It was reported that hyaluronic acid polysaccharides from Streptococcus are unable to inhibit adhesion to immobilised hyaluronic acid (Fried et al., 2000), whereas hyaluronic acid from bovine vitreous humour is inhibitory (Beeson et al., 2000; Chai et al., 2001). Dodecasaccharide fragments generated by digestion of hyaluronic acid with Streptomyces hylauronate lyase or testicular hyaluronidase, and rigorously purified by HPLC, inhibited adhesion to immobilised hyaluronic acid but not chondroitin sulphate A (Chai et al., 2001). These differences in inhibitory activity may be due to differences in mol. wt., or other structural features. Hyaluronic acid from Streptococcus is approximately 800 kDa (supplier’s information, Calbiochem), hyaluronic acid from bovine vitreous has a Mr range of 200–400 kDa (Kujawa et al., 1986) (J.G. Beeson, unpublished observations), and dodecasaccharides are approximately 2.3 kDa (Chai et al., 2001). Various studies have indicated that chain length and structural features can have substantial effects on the biologic activity of hyaluronic acid. Fragments (,250 kDa), but not high molecular weight hyaluronic acid, induced intracellular signalling (reviewed in Turley et al., 2001) and stimulated cellular differentiation (Kujawa et al., 1986), and oligosaccharides, but not high molecular weight hyaluronic acid, stimulated angiogenesis (West et al., 1985). Heat-treated hyaluronic acid was found to have potent anti-complement activity, whereas native hyaluronic acid was relatively inactive, suggesting that structural features, such as intra-molecular coupling of hyaluronic acid chains, can influence activity (Chang et al., 1985). The precise nature of these structural and size differences and how they influence biological activity is currently unclear. There may be technical or other biological factors that account for a lack of inhibitory activity of hyaluronic acid from Streptococcus on parasite adhesion. By way of comparison, soluble ICAM-1 was unable to inhibit parasite adhesion to immobilised ICAM-1 in vitro (Craig et al., 1997), but this did not exclude a specific interaction with ICAM-1. Consistent with a possible role of adhesion to hyaluronic

J.G. Beeson et al. / International Journal for Parasitology 32 (2002) 1245–1252

acid in placental parasite sequestration, hyaluronic acid has been identified on the surface of syncytiotrophoblasts by cytochemistry, immunohistochemistry, and biochemical analysis (Martin et al., 1974; Matejevic et al., 2001; Sunderland et al., 1985). Labelling of hyaluronic acid was found to be particularly prominent in and around fibrinoid deposits and in necrotic areas (Matejevic et al., 2001), which are common in malaria-exposed placentas (Walter et al., 1982). A biochemical analysis of glycosaminoglycans in the placenta reported that hyaluronic acid constituted 0% and ,15% of two different glycosaminoglycan fractions from placental blood, and hyaluronic acid formed an estimated 20% of a glycosaminoglycan fraction obtained from detergent extracts (presumably yielding cell-associated glycosaminoglycans) of placental tissue (Achur et al., 2000). The expression of hyaluronic acid in malariaexposed placentas has not yet been reported. In this study, we have validated parasite assays used to study the adhesion to hyaluronic acid of laboratory-propagated and clinical isolates in the context of placental malaria. Comparative studies of CHO cell adhesion enabled further evaluation of P. falciparum adhesion and provide informative positive controls for hyaluronic acid adhesion assays. We also identified a substantial negative effect of glutaraldehyde fixation of bound cells on measuring hyaluronic acid adhesion in vitro, which may have important practical implications. Additional evidence of the specificity of parasite adhesion to immobilised hyaluronic acid has been provided through the use of hyaluronate lyase from Streptomyces and this may form the basis of a simple protocol using lyase-treated controls when testing isolates for adhesion to hyaluronic acid. These findings are significant for understanding and evaluating interactions between P. falciparum and hyaluronic acid and, more broadly, are relevant to interactions that may be involved in disease pathogenesis. Acknowledgements We thank Wengang Chai for advice and reviewing the manuscript, Marjorie Dunlop and Bob Fraser for helpful discussions and comments, Tim Byrne for help with the culture and preparation of CHO cells, and Sonia Caruana for P. falciparum culture. This work was supported by the National Health and Medical Research Council of Australia. S.J.R. is supported by a Wellcome Trust Fellowship (ref. 063215), and J.G.B. was supported by a Cottrell Fellowship of the Royal Australasian College of Physicians. Blood products were kindly provided by the Red Cross Blood Bank, Victoria. References Achur, R.N., Valiyaveettil, M., Alkhalil, A., Ockenhouse, C.F., Gowda, D.C., 2000. Characterization of proteoglycans of human placenta and

1251

identification of unique chondroitin sulfate proteoglycans of the intervillous spaces that mediate the adherence of Plasmodium falciparuminfected erythrocytes to the placenta. J. Biol. Chem. 275, 40344–56. Barnwell, J.W., Asch, A.S., Nachman, R.L., Yamaya, M., Aikawa, M., 1989. A human 88-kD membrane glycoprotein (CD36) functions in vitro as a receptor for a cytoadherence ligand on Plasmodium falciparum-infected erythrocytes. J. Clin. Invest. 84, 765–72. Beeson, J.G., Brown, G.V., 2002. Pathogenesis of Plasmodium falciparum malaria: the roles of parasite adhesion and antigenic variation. Cell. Mol. Life Sci. 59, 258–71. Beeson, J.G., Chai, W., Rogerson, S.J., Lawson, A.M., Brown, G.V., 1998. Inhibition of binding of malaria-infected erythrocytes by a tetradecasaccharide fraction from chondroitin sulfate A. Infect. Immun. 66, 3397–402. Beeson, J.G., Brown, G.V., Molyneux, M.E., Mhango, C., Dzinjalamala, F., Rogerson, S.J., 1999. Plasmodium falciparum isolates from infected pregnant women and children are associated with distinct adhesive and antigenic properties. J. Infect. Dis. 180, 464–72. Beeson, J.G., Rogerson, S.J., Cooke, B.M., Reeder, J.C., Chai, W., Lawson, A.M., Molyneux, M.E., Brown, G.V., 2000. Adhesion of Plasmodium falciparum-infected erythrocytes to hyaluronic acid in placental malaria. Nat. Med. 6, 86–90. Berendt, A.R., Simmons, D.L., Tansey, J., Newbold, C.I., Marsh, K., 1989. Intercellular adhesion molecule-1 is an endothelial cell adhesion molecule for Plasmodium falciparum. Nature 341, 57–59. Catterall, J.B., Gardner, M.J., Jones, L.M., Turner, G.A., 1997. Binding of ovarian cancer cells to immobilized hyaluronic acid. Glycoconjug. J. 14, 867–9. Chai, W., Beeson, J.G., Kogelberg, H., Brown, G.V., Lawson, A.M., 2001. Inhibition of adhesion of Plasmodium falciparum-infected erythrocytes by structurally defined hyaluronic acid dodecasaccharides. Infect. Immun. 69, 420–5. Chaiyaroj, S.C., Angkasekwinai, P., Buranakiti, A., Looareesuwan, S., Rogerson, S.J., Brown, G.V., 1996. Cytoadherence characteristics of Plasmodium falciparum isolates from Thailand: Evidence for chondroitin sulfate A as a cytoadherence receptor. Am. J. Trop. Med. Hyg. 55, 76–80. Chang, N.S., Boackle, R.J., Armand, G., 1985. Hyaluronic acid-complement interactions – I. Reversible heat-induced anticomplementary activity. Mol. Immunol. 22, 391–7. Cooke, B.M., Rogerson, S.J., Brown, G.V., Coppel, R.L., 1996. Adhesion of malaria-infected red blood cells to chondroitin sulfate A under flow conditions. Blood 88, 4040–4. Craig, A.G., Pinches, R., Khan, S., Roberts, D.J., Turner, G.D., Newbold, C.I., Berendt, A.R., 1997. Failure to block adhesion of Plasmodium falciparum-infected erythrocytes to ICAM-1 with soluble ICAM-1. Infect. Immun. 65, 4580–5. DeGrendele, H.C., Estess, P., Picker, L.J., Siegelman, M.H., 1996. CD44 and its ligand hyaluronate mediate rolling under physiologic flow: a novel lymphocyte-endothelial cell primary adhesion pathway. J. Exp. Med 183, 1119–30. Fried, M., Duffy, P.E., 1996. Adherence of Plasmodium falciparum to chondroitin sulfate A in the human placenta. Science 272, 1502–4. Fried, M., Lauder, R.M., Duffy, P.E., 2000. Plasmodium falciparum: adhesion of placental isolates modulated by the sulfation characteristics of the glycosaminoglycan receptor. Exp. Parasitol. 95, 75–78. Gardner, J.P., Pinches, R.A., Roberts, D.J., Newbold, C.I., 1996. Variant antigens and endothelial receptor adhesion in Plasmodium falciparum. Proc. Natl. Acad. Sci. USA 93, 3503–8. Grimshaw, J., Kane, A., Trocha-Grimshaw, J., Douglas, A., Chakravarthy, U., Archer, D., 1994. Quantitative analysis of hyaluronan in vitreous humor using capillary electrophoresis. Electrophoresis 15, 936–40. Kujawa, M.J., Carrino, D.A., Caplan, A.I., 1986. Substrate-bonded hyaluronic acid exhibits a size-dependent stimulation of chondrogenic differentiation of stage 24 limb mesenchymal cells in culture. Dev. Biol. 114, 519–28. Legras, S., Levesque, J.P., Charrad, R., Morimoto, K., Le Bousse, C., Clay,

1252

J.G. Beeson et al. / International Journal for Parasitology 32 (2002) 1245–1252

D., Jasmin, C., Smadja-Joffe, F., 1997. CD44-mediated adhesiveness of human hematopoietic progenitors to hyaluronan is modulated by cytokines. Blood 89, 1905–14. Martin, B.J., Spicer, S.S., Smythe, N.M., 1974. Cytochemical studies of the maternal surface of the syncytiotrophoblast of human early and term placenta. Anat. Rec. 178, 769–86. Matejevic, D., Neudeck, H., Graf, R., Muller, T., Dietl, J., 2001. Localization of hyaluronan with a hyaluronan-specific hyaluronic acid binding protein in the placenta in pre-eclampsia. Gynecol. Obstet. Invest. 52, 257–9. Maubert, B., Guilbert, L.J., Deloron, P., 1997. Cytoadherence of Plasmodium falciparum to intercellular adhesion molecule 1 and chondroitin4-sulfate expressed by the syncytiotrophoblast in the human placenta. Infect. Immun. 65, 1251–7. Mohamadzadeh, M., DeGrendele, H., Arizpe, H., Estess, P., Siegelman, M., 1998. Proinflammatory stimuli regulate endothelial hyaluronan expression and CD44/HA-dependent primary adhesion. J. Clin. Invest. 101, 97–108. Ockenhouse, C.F., Tandon, N.N., Magowan, C., Jamieson, G.A., Chulay, J.D., 1989. Identification of a platelet membrane glycoprotein as a falciparum malaria sequestration receptor. Science 243, 1469–71. Ohya, T., Kaneko, Y., 1970. Novel hyaluronidase from Streptomyces. Biochim. Biophys. Acta 198, 607–9. Parkar, A.A., Kahmann, J.D., Howat, S.L., Bayliss, M.T., Day, A.J., 1998. TSG-6 interacts with hyaluronan and aggrecan in a pH-dependent manner via a common functional element: implications for its regulation in inflamed cartilage. FEBS Lett. 428, 171–6. Rogerson, S.J., Chaiyaroj, S.C., Ng, K., Reeder, J.C., Brown, G.V., 1995. Chondroitin sulfate A is a cell surface receptor for Plasmodium falciparum-infected erythrocytes. J. Exp. Med. 182, 15–20.

Rowen, J.W., Brunish, R., Bishop, F.W., 1956. Form and dimensions of isolated hyaluronic acid. Biochim. Biophys. Acta 19, 480–9. Stamenkovic, I., Aruffo, A., 1994. Hyaluronic acid receptors. Methods Enzymol. 245, 195–216. Sunderland, C.A., Bulmer, J.N., Luscombe, M., Redman, C.W.G., Stirrat, G.M., 1985. Immunohistological and biochemical evidence for a role for hyaluronic acid in the growth and development of the placenta. J. Reprod. Immunol. 8, 197–212. Tammi, R., MacCallum, D., Hascall, V.C., Pienimaki, J.-P., Hyttinen, M., Tammi, M., 1998. Hyaluronan bound to CD44 on keratinocytes is displaced by hyaluronan decasaccharides and not hexasaccharides. J. Biol. Chem. 273, 28878–88. Turley, E.A., Noble, P.W., Bourguignon, L.Y., 2001. Signalling properties of hyaluronan receptors. J. Biol. Chem. 20, 20. Udomsangpetch, R., Reinhardt, P.H., Schollaardt, T., Elliott, J.F., Kubes, P., Ho, M., 1997. Promiscuity of clinical Plasmodium falciparum isolates for multiple adhesion molecules under flow conditions. J. Immunol. 158, 4358–64. Valiyaveettil, M., Achur, R.N., Alkhalil, A., Ockenhouse, C.F., Gowda, D.C., 2001. Plasmodium falciparum cytoadherence to human placenta: evaluation of hyaluronic acid and chondroitin 4-sulfate for binding of infected erythrocytes. Exp. Parasitol. 99, 57–65. Walter, P.R., Garin, Y., Blot, P., 1982. Placental pathologic changes in malaria. A histologic and ultrastructural study. Am. J. Pathol. 109, 330–42. Wang, H.S., Underhill, C.B., 1992. Hyaluronan can be non-enzymatically linked to protein through an alkali sensitive bond. Connect. Tissue Res. 28, 29–48. West, D.C., Hampson, I.N., Arnold, F., Kumar, S., 1985. Angiogenesis induced by degradation products of hyaluronic acid. Science 228, 1324–6.