Parasite adhesion and immune evasion in placental malaria

Review

TRENDS in Parasitology Vol.17 No.7 July 2001

331

Parasite adhesion and immune evasion in placental malaria James G. Beeson, John C. Reeder, Stephen J. Rogerson and Graham V. Brown Parasite sequestration in the placenta is a key feature of infection by Plasmodium falciparum during pregnancy and is associated with severe adverse outcomes for both mother and baby. Here, James Beeson and colleagues draw together the findings of recent studies on parasite mechanisms that mediate this process. They review evidence for novel parasite variants that appear able to evade pre-existing immunity, for the adhesion of P. falciparum-infected erythrocytes to placental glycosaminoglycans (and the molecular basis of these parasite properties) and for the expression of var genes encoding the variant antigen and adhesive ligand P. falciparumerythrocyte membrane protein 1 (PfEMP1).

James G. Beeson* John C. Reeder Stephen J. Rogerson Graham V. Brown Dept of Medicine, University of Melbourne, Royal Melbourne Hospital, Parkville, VIC 3050, Australia. *e-mail: [email protected]

Pregnant women are particularly susceptible to infection with Plasmodium falciparum, which has serious consequences for both mother and baby. With around 40% of the world’s population estimated to reside in malaria-endemic areas, tens of millions of pregnancies are at risk each year. Infection can lead to low birthweight and prematurity and to infant anemia, which increases infant morbidity and mortality, and contributes to maternal anaemia and deaths1–3. The risk of infection is highest for women in first pregnancies (primigravidae) and tends to be reduced, but not absent, in subsequent pregnancies; this pattern might result from the development of effective immune responses to these infections. Several host factors could account for the increased susceptibility of pregnant women to malaria4,5, including changes in immunological responses that might reduce parasite clearance (e.g. impaired or altered T-cell responses to infection6,7) and changes in specific antimalarial antibodies8,9. The characteristic feature of infection during pregnancy is the accumulation of P. falciparuminfected erythrocytes in the intervillous blood spaces of the placenta10–12. It is uncertain how placental infection contributes to fetal morbidity, but it could lead to impaired exchange of nutrients and gases and to the production by maternal immune cells of proinflammatory cytokines, such as tumor necrosis factor-α and interferon-γ, which are detrimental to fetal development6,13. Infections in the placenta often reach very high densities and can be observed in the absence of a detectable peripheral parasitemia. Typically, placental parasites consist of mature forms not seen in the periphery10,12. Together, these observations suggest that placental parasites accumulate as a consequence of specific processes, such as cell adhesion. This review focuses on three key parasite determinants of P. falciparum infection of the

placenta: (1) the emergence of novel parasite variants or serotypes in pregnancy, which are able to evade pre-existing immunity; (2) adhesion of P. falciparuminfected erythrocytes to glycosaminoglycans (GAGs) lining placental blood spaces [e.g. chondroitin sulfate A (CSA) and hyaluronic acid (HA)]; and (3) the expression of var genes encoding the parasite protein P. falciparum erythrocyte membrane protein 1 (PfEMP1), which is important in determining the antigenic and adhesive phenotypes of infected erythrocytes. Emergence of new variants of P. falciparum-infected erythrocytes during pregnancy

Men, non-pregnant women and older children (≥ 12 years) of both sexes in endemic areas typically demonstrate substantial immunity to malaria. They also show a large repertoire of variant-specific agglutinating antibodies against different P. falciparum isolates taken from children, which reflects previous exposure14,15. These antibodies correlate with protection from infection and clinical disease16. This aspect of naturally acquired immunity appears to be intact in pregnancy; sera from both primigravidae and multigravidae (women in their third or subsequent pregnancy) agglutinated infected erythrocytes that had been isolated from men, nonpregnant women and children to a similar extent as did sera from men and non-pregnant women17,18. By contrast, variant-specific antibodies to isolates from the placenta or the peripheral blood of pregnant women were rarely detected in the serum of men or primigravidae17,18. This suggests that pregnancy selects for a subpopulation of parasites to which individuals are not exposed before pregnancy. The emergence of new variants or serotypes in pregnancy would enable parasite-infected erythrocytes to evade an important aspect of the acquired immune response to malaria and this could be a key event in enabling infections to take hold. The humoral antibody repertoire appears to diversify with each pregnancy: sera from multigravidae frequently agglutinated placental isolates17 or peripheral blood isolates from pregnant women18, suggesting that following exposure in first or second pregnancies, women can develop antibodies that could protect against infections in subsequent pregnancies. Similarly, variant-specific antibodies against laboratorypropagated isolates selected for high levels of adhesion to CSA (a feature of placental isolates) were

http://parasites.trends.com 1471-4922/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S1471-4922(01)01917-1

332

Review

TRENDS in Parasitology Vol.17 No.7 July 2001

Malawi17 and Cameroon30,31. In most cases, placental isolates bound to purified CSA and could bind to placental tissue sections21 or cultured syncytiotrophoblasts31 in a CSA-dependent manner. By contrast, isolates from men, non-pregnant women and children bind to CSA infrequently and usually do so only at low levels17,21,32.

Fig. 1. Histopathology of infected placental tissue. Some Plasmodium falciparum-infected erythrocytes (arrow) in the intervillous space (IVS) appear to be directly adherent to the surface of the syncytiotrophoblast cell layer (arrowhead) of a placental villus (PV). Scale bar = ~20 µm.

Adhesion to HA

also associated with gravidity, and were more common among sera from pregnant multigravidae18,19. Further evidence that complications in pregnancy are related to phenotypic change in parasite populations comes from the demonstration that placental isolates show different adhesive characteristics to isolates from men, non-pregnant women and children. Typically, isolates bind to CSA and HA (Ref. 20), but do not adhere to intercellular adhesion molecule (ICAM)-1 or CD36 (Refs 17,21) or form erythrocyte rosettes (infected–uninfected erythrocyte complexes)22,23, properties exhibited by most isolates from men, non-pregnant women and children.

HA has recently been shown to support adhesion of P. falciparum-infected erythrocytes in vitro and appears to be an additional receptor for the sequestration of infected erythrocytes in the placenta20. In most cases, parasitized erythrocytes isolated from infected placentas were found to adhere to HA, whereas isolates from the peripheral blood of either children or pregnant women bound HA less commonly and generally at lower levels20. The majority of placental isolates examined demonstrated some adhesion to both CSA and HA to varying degrees. In establishing the specificity of adhesion to HA, it was shown that parasitized cells bound to a purified form of HA containing no detectable chondroitin sulfates (CS) or other GAGs, and adhesion could be inhibited by highly purified and structurally defined oligosaccharide fragments20,33. Treatment of receptors with a specific hyaluronidase from Streptomyces (J.G. Beeson, unpublished) or trypsin treatment of parasite surface proteins prevented adhesion to HA but not to CSA (Ref. 20). Furthermore, some isolates showed binding to only CSA or HA, rather than to both, suggesting separate parasite specificity for each receptor. Relative roles of CSA and HA

Adhesion to GAGs in the placenta

The syncytiotrophoblast cell layer, which lines the placental blood spaces, constitutes an extensive area of fetal tissue that is in contact with the maternal circulation. Histopathological examination (Fig. 1) of infected placental tissue shows that some infected erythrocytes appear to adhere to syncytiotrophoblasts10,12; this suggests that cellular adhesion to receptors on this cell layer might be important for the sequestration of parasite-infected erythrocytes in the placenta. Studies in Africa suggest that circulating parasites can sequester in the placenta through adhesion to CSA and HA, which are constituents of a prominent coat of GAGs on the syncytiotrophoblast layer24–26. Adhesion to CSA

CSA was first identified as a parasite adhesion receptor on the surface of Chinese hamster ovary cells27 and endothelial cells28, and infected erythrocytes can adhere to immobilized purified CSA or to CSA chains on the proteoglycan thrombomodulin, under static and flow conditions29. Adhesion to CSA in vitro was found to be a common feature of placental parasite isolates in Kenya21, http://parasites.trends.com

Understanding the relative importance or roles of CSA, HA and possibly other molecules as adhesion receptors in the placenta will be important for the development of novel therapies. Detailed studies are needed to determine the relative expression of CSA and HA on the syncytiotrophoblast layer, particularly during malaria infections. Immunohistochemistry has demonstrated that there are substantial levels of CS on the syncytiotrophoblast surface in both infected and uninfected placentas21,31. Studies using cytochemical, biochemical and immunohistochemical analysis24,26 have suggested that HA is also prominent on the placental lining. A recent study34 suggested that CS is the major GAG present in uninfected placental tissue and blood, but the expression of CS and other GAGs on syncytiotrophoblasts where adhesion occurs was not specifically examined. Pro-inflammatory cytokines, which are increased in malaria infection35, increase expression of HA on microvasular endothelial cells36, but do not alter the expression of CS on syncytiotrophoblasts37. An ability to express multiple adhesive ligands on the infected erythrocyte surface for adhesion to different receptors might convey a survival advantage

Review

TRENDS in Parasitology Vol.17 No.7 July 2001

for P. falciparum; this would increase the likelihood of sequestering and thus avoiding splenic clearance from the circulation, even in the face of possible host responses targeting one of the adhesive interactions. In laboratory-propagated clonal isolates, infected erythrocytes can coexpress ligands or binding sites for both HA and CSA (Ref. 20), a property that could augment parasite sequestration through synergy between adhesion to the two receptors, as shown with coexpression of CD36 and ICAM-1 on endothelial cells38. Cause for caution

Some caution must be exerted in the interpretation of adhesion studies because the conditions and receptors used in these in vitro experiments might not be representative of those in vivo. Assays of adhesion of parasite isolates in vitro have important limitations. Most assays simply determine that a population of parasite-infected erythrocytes contains some cells that can bind to the receptors tested and, in standard assays, this usually represents less than 1% of all cells added (J.G. Beeson, unpublished). Not all placental isolates tested have demonstrated binding to CSA or HA (Refs17,20) or to cultured syncytiotrophoblasts31, and some isolates bind only at low levels. This suggests that other factors, possibly other receptors, are involved in sequestration of parasite-infected erythrocytes in the placenta. Recently, it was reported that early ring forms of P. falciparum-infected erythrocytes can adhere to placental tissue sections via an unidentified receptor39. Previously, it was believed that only mature parasite forms were sequestered, and these observations require further investigation. Histological examination of infected placental tissue has shown that many infected erythrocytes are not adherent to syncytiotrophoblasts (Fig. 1) and might be retained in the placental blood spaces by other means10,11. Examining tissue collected by caesarean section rather than following vaginal delivery could help to clarify the significance of these observations. Other mechanisms for parasite sequestration might include reduced cell deformability of mature-stage infected erythrocytes40, which could favor trapping in the intervillous spaces or the formation of large parasite aggregates by unknown mechanisms. Two studies in Africa22,23 have found that rosette formation, which might be involved in the sequestration of parasite-infected erythrocytes is rare among placental isolates, suggesting that it is not an important mechanism for sequestration in this organ. Adhesion-blocking antibodies

Multigravidae from Kenya, Malawi and Thailand were found to have serum antibodies that inhibited adhesion to CSA in vitro; primigravidae, who are at a greater risk of contracting malaria, however, did not have these antibodies41. Furthermore, the presence of http://parasites.trends.com

333

such antibodies was associated with reduced placental malaria41. Inhibition of adhesion to CSA was also reported using plasma from pregnant women resident in Cameroon18,30. No association between gravidity and the ability of plasma to inhibit adhesion was found30, in spite of associations between gravidity and agglutinating antibodies, and between agglutination and the ability of plasma to inhibit adhesion to cultured syncytiotrophoblasts18. Among Kenyan women, no association was found between agglutination and inhibition of CSA-adhesion by sera41. The possible protective effect of adhesion-blocking antibodies, however, does not appear to be complete, because serum from some women with active placental infection was found to inhibit parasite adhesion18. In addition, one study reported that, although primigravidae face a high risk of infection, all plasma tested from primigravidae could inhibit adhesion30. Further studies with plasma from women in Ghana found no association between gravidity and inhibition of adhesion, despite an association between gravidity and antibodies to variant antigens on parasitized erythrocytes19. In these studies, plasma with the highest antibody levels had the greatest inhibitory effect in adhesion assays. Important differences between the various studies could include: the use of brain endothelial cells expressing CSA as the substrate30, rather than purified CSA (Ref. 41) or syncytiotrophoblasts18; the use of in vitro propagated parasites18,19,30 rather than placental parasite isolates41; and the sample populations examined. Factors influencing adhesion in the placenta

CS and HA are widely distributed in peripheral vascular beds29,36 and it is unclear whether P. falciparum uses these molecules as receptors for adhesion outside the placenta or to what extent sequestration occurs in other organs during pregnancy. Postmortem studies of deaths during pregnancy might help to resolve this question and give important insights into the pathogenesis of severe maternal malaria. Several studies have shown that parasite-infected cell adhesion to CSA is not restricted to pregnancy. However, it is not as common a phenotype among men, non-pregnant women and children as it is among pregnant women, and of those isolates that do show some adhesion, it is not usually the predominant phenotype17,20,21,32,42. Furthermore, it is not known whether CSA and HA in tissues other than the placenta are expressed at sufficient levels or in an appropriate configuration to support substantial parasite adhesion. Although a study of postmortem tissue collected from adults who died from malaria demonstrated the presence of CS in the cerebral microvasculature, CS expression was generally low43. Endothelial cells cultured from squirrel monkey (Saimiri) brain and human lung support CSA-dependent adhesion in vitro28,30 and in

Review

334

TRENDS in Parasitology Vol.17 No.7 July 2001

Extracellular domain

Internal domain

Exon 1 6.7 kb

Exon 2 1.4 kb

DBL1 CIDR DBL3 DBL4

SVL

var CS2 8.8 kb PfEMP1~300 kDa

Exon 1 9.9 kb

Exon 2 0.7 kb

DBL1 CIDR DBL2 DBL3 DBL4 DBL5 DBL6 DBL7

FCR3var -CSA 10.8 kb PfEMP1~400 kDa

Erythrocyte membrane

placental surface is low or absent, and/or that during pregnancy, a selective immune pressure exists against common variants that typically bind to those receptors. Previous studies52,53 have demonstrated that serum from immune adults can effectively inhibit adhesion of parasites to cells expressing CD36 and ICAM-1. CD36 has not been detected on the surface of syncytiotrophoblasts, whereas ICAM-1 has been found on the surface of syncytiotrophoblast cells in culture but at low levels on the syncytiotrophoblast layer in placental tissue sections31,37,54. It would be interesting to examine adhesion of placental isolates to other receptors in women with no previous exposure to malaria.

TRENDS in Parasitology

var genes and PfEMP1: adhesion and antigenic variation Fig. 2. Predicted domain organization of two var genes that encode CSA-binding PfEMP1 proteins, varCS2 (Ref. 59) and FCR3var-CSA (Ref. 60). The DBL3 domain mediates adhesion to CSA in both genes. With varCS2, the CIDR domain also appears to be involved. The proposed61 active binding sites of varCS2 DBL3 and CIDR regions are indicated by black bars. Abbreviations: CIDR, Cys-rich interdomain region; CSA, chondroitin sulfate A; DBL, Duffy-binding like; PfEMP1, Plasmodium falciparum erythrocyte membrane protein 1; SVL, segment of variable length.

Thailand, an association was found between CSAmediated adhesion to human lung endothelial cells and severe malaria44. However, this association was not found in Malawi32 or in a previous Thai study42. The sulfation and chain length of CS polysaccharides can differ widely from tissue to tissue45, which could influence the pattern of sequestration of infected erythrocytes. Adhesion is dependent on 4-O-sulfated disaccharide repeats in CSA and might be increased by the presence of undersulfated elements46,47 (J.G. Beeson and W. Chai, unpublished), which are common among placental CS (Ref. 34). Interactions of mammalian cells with HA can be highly dependent on the chain length of the GAG (Ref. 48) and this might also be an important factor for parasite adhesive interactions. Sequestration of infected erythrocytes might be influenced by the blood flow rate in the placenta, which is slow compared with that in other vascular beds49. Using an in vitro model, high levels of adhesion to HA were observed only at low wall shear stress, conditions that are thought to prevail in the placenta, but adhesion was greatly reduced or absent under higher stresses predicted to exist in capillaries and postcapillary venules20. By contrast, CSA and its proteoglycan, thrombomodulin, can support substantial adhesion at high wall shear stresses50,51; therefore, the effect of vascular flow rate might not be an important factor influencing the adhesion of infected erythrocytes to CSA in the placenta, rather than in other organs. Flow studies need to be performed with placental isolates to complement or validate the findings using laboratory-propagated isolates. In pregnant women exposed to malaria, parasites generally do not adhere to other known receptors such as CD36 and ICAM-1 (Refs 17,21,31), which suggests that the expression of these molecules on the http://parasites.trends.com

Both antigenic variation and parasite adhesion are properties that have been attributed to the protein PfEMP1 on the surface of infected erythrocytes. However, other factors might be involved55. PfEMP1 is encoded by a large multigene family termed var (up to 50 genes per genome) and expression of different var genes results in different PfEMP1 species being presented on the erythrocyte surface, conferring different adhesive and antigenic properties to the cell56–58. PfEMP1 has been identified as the parasite protein mediating adhesion to CSA59,60, which fits comfortably with the model of immune evasion being driven by phenotypic advantage. Using isolates selected for high levels of adhesion to CSA, substantial evidence has implicated both the Duffy-binding-like (DBL) 3 domain and the Cys-rich interdomain region (CIDR) in adhesion to CSA (Refs 59–62). In the CS2 isolate [derived from ItGF2 (Ref. 27)], a major var gene transcript was identified and sequenced (Fig. 2), and antibodies raised against recombinant proteins, corresponding to DBL3 and CIDR, inhibited parasite adhesion59. DBL3 and CIDR recombinant proteins also have GAG-binding activity61. Different domains from the major var transcript identified in the FCR3-CSA isolate (Fig. 2) were expressed on the surface of mammalian cells and only the DBL3 domain was found to bind CSA in a specific manner60. PfEMP1 extracted from FCR3-CSA parasite cultures also bound to CSA (Ref. 60). In separate studies, recombinant proteins, corresponding to CIDR domains that were identified in an ItGF2 isolate, bound to cells naturally expressing CSA62. These genes were not fully sequenced and DBL3 domains were not examined. Although the genes identified were different, it is probable that the ItGF2derived and FCR3-CSA isolates are derived from the same parent and might be genetically identical or very similar63. Interestingly, var genes expressed in CSAbinding parasite-infected cells derived from isolate 3D7 (used in the Malaria Genome Project) do not appear to have DBL3-like domains64. Recently, a new system based on sequence analysis for classifying and naming PfEMP1 domains has been proposed65, in

Review

TRENDS in Parasitology Vol.17 No.7 July 2001

Non-pregnant

Pregnant

Vascular endothelium

Placenta

Primigravidae

Y

Y

Multigravidae

Y

Y

Y Y

Y

Y

antibodies

Y

Y

Immune

Y

Y

antibodies

Y

Non-immune

Y

Y

Y

Y

Y

Y

Vascular endothelium Y

Y

?

TRENDS in Parasitology

Fig. 3. The possible roles of antigenic variation and host-cell adhesion in infection with Plasmodium falciparum. Non-immune, non-pregnant individuals are thought to be infected with parasite populations (blue) that adhere to the vascular endothelium of various organs via receptors such as CD36 and ICAM-1 (green). Immune individuals possess a broad range of variant-specific antibodies that appear to be protective, and show antibodies able to inhibit parasite adhesion to endothelial cells. Primigravidae (likened to a ‘non-immune’ state) develop infections with antigenically distinct subpopulations of parasites (yellow) that can adhere to specific receptors, such as chondroitin sulfate A (CSA) and hyaluronic acid (HA) in the placenta (purple). Multigravidae (likened to an ‘immune’ state) are less susceptible to malaria, partly because of specific immune responses to these subpopulations developed from exposure in previous pregnancies. Such responses might include antibodies that inhibit the adhesion of parasites to placental syncytiotrophoblasts and facilitate parasite clearance. During pregnancy, it is not known whether separate parasite populations also sequester in organs other than the placenta through adhesion to endothelial receptors such as CD36 and ICAM-1 (green), or CSA and HA (purple).

Acknowledgements We thank our colleagues and collaborators for many constructive discussions that helped shape ideas expressed in this article. Financial support was provided by the National Health and Medical Research Council, Australia and the Wellcome Trust, UK.

which the specific DBL3 and CIDR domains referred to in this review are classified as DBLγ and CIDRα, respectively. Examination of the var gene sequences suggests that identification of target sites for possible therapy might not be trivial; the two DBL3 domains identified have little similarity in sequence other than that shared by all DBL3-type domains60, most of which probably do not bind CSA. In vitro studies are currently limited, however, by a lack of knowledge about the tertiary structure of the domains. Subsequent studies on varCS2 (Ref. 61) have further defined the location of binding sites in the DBL3 domain and CIDR (Fig. 2). Both the N- and C-terminal ends of the DBL3 domain, where Cys residues are found, demonstrated GAG-binding activity, whereas a polypeptide from the center of DBL3 did not bind. The C-terminal half of CIDR, adjacent to the DBL3 region, bound CSA, whereas the N-terminal half of the domain did not. A ten amino acid peptide sequence from the CIDR region was identified that bound to CSA and shared homology with a mammalian CS-binding protein61. The finding that pregnant women develop antibodies that inhibit adhesion to CSA suggests http://parasites.trends.com

335

that functional immunity against such a highly variable protein as PfEMP1 does develop. Intriguingly, sera were found to inhibit adhesion of different CSA-binding isolates, including those from geographically different areas41. This suggests that the binding site for CSA might be relatively well conserved, or antigenically restricted and would, therefore, represent a feasible therapeutic target. Alternatively, single or persistent placental infections could result in the expression of a range of different PfEMP1 types by parasites, thus stimulating production of antibodies that convey subsequent protection. If this were the case, it would pose a greater challenge to developing an effective vaccine for use in pregnancy. Agglutination of placental isolates by sera from Malawian women of different parities was isolate specific and panagglutinating sera were not seen. This suggests that placental isolates are not antigenically restricted and that agglutinating antibodies are not targeting conserved motifs17. It is possible that a conserved binding site for CSA exists that is dependent on conformation. This will only become apparent after closer investigation of the binding regions in a wider range of parasites, or examination of the tertiary structure of the ligand(s). Studies with varCS2 suggest that tertiary structure and conformation do influence binding61. Even if a conserved CSA ligand can be uncovered, will targeting this site be effective in blocking placental infection if adhesion to HA (and/or possibly other receptors) remains unaffected? Although the ligand mediating adhesion to HA has not yet been identified, ligand(s) for the adhesion of parasites to each of HA and CSA can be coexpressed20 and might be mediated by a single molecule, namely PfEMP1. Dissecting the molecular detail of the adhesive interactions will help evaluate the probable benefit of therapeutically targeting parasite adhesion. However, it remains possible that there are other receptors or mechanisms contributing to the sequestration of parasites in the placenta. These need further investigation before approaches to novel adhesion-blocking therapies can proceed. Conclusion

We believe that a combination of immune selection and selection by adhesion of parasite-infected erythrocytes to placental GAGs – properties that appear to be governed by the expression of key var genes – drives the emergence of novel or rare phenotypes in the development of placental malaria (Fig. 3). In primigravidae, new variants that can evade existing antimalarial immunity, and also adhere to CSA, HA and possibly other receptors, enable an infection to take hold and expand in the relatively protected environment of the placenta. Immune responses against placental infections might then prevent or limit infections in subsequent pregnancies. These immune responses could include

Review

336

TRENDS in Parasitology Vol.17 No.7 July 2001

2

1

4 5 3

CSA

HA

Unknown receptors

Parasite-infected erythrocytes TRENDS in Parasitology

Fig. 4. Possible mechanisms accounting for the sequestration of Plasmodium falciparum-infected erythrocytes in the placenta. Blood enters the placenta and slows as it moves through the intervillous space (arrows). Mature-stage parasitized erythrocytes expressing appropriate ligands can adhere to receptors on the syncytiotrophoblast cell layer. Parasites might initially roll on receptors (1), until they become tethered by adhering to both HA and CSA (2). Alternatively, parasites might adhere to HA or CSA only, or to other receptors yet to be identified (3). Other infected erythrocytes appear to be retained, as parasites aggregate by unknown means (4). Infected erythrocytes without adhesive ligands on the cell surface, or those with ligands for adhesion to other receptors, such as CD36, that are not present on the placental lining are thought to pass through the placental blood spaces and do not sequester (5). Abbreviations: CSA, chondroitin sulfate A; HA, hyaluronic acid.

References 1 Brabin, B.J. (1983) An analysis of malaria in pregnancy in Africa. Bull. WHO 61, 1005–1016 2 McGregor, I.A. et al. (1983) Malaria infection of the placenta in The Gambia, West Africa; its incidence and relationship to stillbirth, birthweight and placental weight. Trans. R. Soc. Trop. Med. Hyg. 77, 232–244 3 Granja, A.C. et al. (1998) Malaria-related maternal mortality in urban Mozambique. Ann. Trop. Med. Parasitol. 92, 257–263 4 Rogerson, S.J. and Beeson, J.G. (1999) The placenta in malaria: mechanisms of infection, disease and fetal morbidity. Ann. Trop. Med. Parasitol. 93, S35–S42 5 Menendez, C. (1995) Malaria during pregnancy: a priority area of malaria research and control. Parasitol. Today 11, 178–183 6 Lea, R.G. and Calder, A.A. (1997) The immunology of pregnancy. Curr. Opin. Infect. Dis. 10, 171–176 7 Riley, E.M. et al. (1989) Suppression of cellmediated immune responses to malaria antigens in pregnant Gambian women. Am. J. Trop. Med. Hyg. 40, 141–144 8 Nambei, W.S. et al. (1998) Imbalanced distribution of IgM and IgG antibodies against Plasmodium falciparum antigens and merozoite surface protein-1 (MSP1) in pregnancy. Immunol. Lett. 61, 197–199 9 Deloron, P. et al. (1989) Serological reactivity to the ring-infected erythrocyte surface antigen and circumsporozoite protein in gravid and nulligravid women infected with Plasmodium http://parasites.trends.com

10 11

12

13

14

15

16

17

18

variant-specific and adhesion-inhibition antibodies, and other changes in immune function, such as in T-cell responses. However, immunity that develops is not complete, as multigravidae are still at some risk of infection. For parasite adhesion within the placenta (Fig. 4), sequestration is associated with the presence of GAGs in an appropriate form and at sufficient levels, and the unique low-pressure environment of the placenta might influence its susceptibility to infection by parasitized erythrocytes. Strategies for therapy or prevention of placental malaria include targeting the adhesive interaction between infected erythrocytes and the placental GAG receptors, or augmenting variant-specific immunity to combat the emergence of novel antigenic variants in pregnancy. Future priorities for research should include the identification and thorough analysis of parasite ligands that mediate sequestration, particularly looking for conserved motifs that could be targets for therapy. Careful evaluation of the role of specific immune responses in infection might yield insights into how these can be augmented to protect pregnancies, and how pregnancies that are most at risk can be identified. Prospective studies could be particularly valuable in this regard. Understanding the mechanisms of placental malaria and the host responses to infection are clearly important areas for future research. These areas might also yield insights into the pathogenesis of malaria in other organs, and are relevant to a broader understanding of adverse events at the maternal–fetal interface.

falciparum. Trans. R. Soc. Trop. Med. Hyg. 83, 58–62 Walter, P.R. et al. (1982) Placental pathologic changes in malaria. Am. J. Pathol. 109, 330–342 Bray, R.S. and Sinden, R.E. (1979) The sequestration of Plasmodium falciparum-infected erythrocytes in the placenta. Trans. R. Soc. Trop. Med. Hyg. 73, 716–719 Yamada, M. et al. (1989) Plasmodium falciparum associated placental pathology: a light and electron microscopic and immunohistologic study. Am. J. Trop. Med. Hyg. 41, 161–168 Fried, M. et al. (1998) Malaria elicits type 1 cytokines in the human placenta: IFN-γ and TNF-α associated with pregnancy outcomes. J. Immunol. 160, 2523–2530 Marsh, K. and Howard, R.J. (1986) Antigens induced on erythrocytes by P. falciparum: expression of diverse and conserved determinants. Science 231, 150–153 Reeder, J.C. et al. (1994) Diversity of agglutinating phenotype, cytoadherence, and rosette-forming characteristics of Plasmodium falciparum isolates from Papua New Guinean children. Am. J. Trop. Med. Hyg. 51, 45–55 Bull, P.C. et al. (1998) Parasite antigens on the infected red cell surface are targets for naturally acquired immunity to malaria. Nat. Med. 4, 358–360 Beeson, J.G. et al. (1999) Plasmodium falciparum isolates from infected pregnant women and children are associated with distinct adhesive and antigenic properties. J. Infect. Dis. 180, 464–472 Maubert, B. et al. (1999) Development of

19

20

21

22

23

24

25

26

antibodies against chondroitin sulfate A-adherent Plasmodium falciparum in pregnant women. Infect. Immun. 67, 5367–5371 Ricke, C.H. et al. (2000) Plasma antibodies from malaria-exposed pregnant women recognize variant surface antigens on Plasmodium falciparum-infected erythrocytes in a paritydependent manner and block parasite adhesion to chondroitin sulfate A. J. Immunol. 165, 3309–3316 Beeson, J.G. et al. (2000) Adhesion of Plasmodium falciparum-infected erythrocytes to hyaluronic acid in placental malaria. Nat. Med. 6, 86–90 Fried, M. and Duffy, P.E. (1996) Adherence of Plasmodium falciparum to chondroitin sulfate A in the human placenta. Science 272, 1502–1504 Rogerson, S.J. et al. (2000) Plasmodium falciparum rosette formation is uncommon in isolates from pregnant women. Infect. Immun. 68, 391–393 Maubert, B. et al. (1998) Plasmodium falciparumisolates from Cameroonian pregnant women do not rosette. Parasite 5, 281–283 Martin, B.J. et al. (1974) Cytochemical studies of the maternal surface of the syncytiotrophoblast of human early and term placenta. Anat. Rec. 178, 769–786 Parmley, R.T. et al. (1984) Ultrastructural localization of glycosaminoglycans in human term placenta. Anat. Rec. 210, 477–484 Sunderland, C.A. et al. (1985) Immunohistological and biochemical evidence for a role for hyaluronic

Review

27

28

29

30

31

32

33

34

35

36

37

38

TRENDS in Parasitology Vol.17 No.7 July 2001

acid in the growth and development of the placenta. J. Reprod. Immunol. 8, 197–212 Rogerson, S.J. et al. (1995) Chondroitin sulfate A is a cell surface receptor for Plasmodium falciparum-infected erythrocytes. J. Exp. Med. 182, 15–20 Robert, C. et al. (1995) Chondroitin-4-sulphate (proteoglycan), a receptor for Plasmodium falciparum-infected erythrocyte adherence on brain microvascular endothelial cells. Res. Immunol. 146, 383–393 Rogerson, S.J. and Brown, G.V. (1997) Chondroitin sulfate A as an adherence receptor for Plasmodium falciparum-infected erythrocytes. Parasitol. Today 13, 70–75 Gysin, J. et al. (1999) Ex vivo desequestration of Plasmodium falciparum-infected erythrocytes from human placenta by chondroitin sulfate A. Infect. Immun. 67, 6596–6602 Maubert, B. et al. (2000) Cytoadherence of Plasmodium falciparum-infected erythrocytes in the human placenta. Parasite Immunol. 22, 191–199 Rogerson, S.J. et al. (1999) Cytoadherence characteristics of Plasmodium falciparuminfected erythrocytes from Malawian children with severe and uncomplicated malaria. Am. J. Trop. Med. Hyg. 61, 467–472 Chai, W. et al. (2001) Inhibition of adhesion of Plasmodium falciparum-infected erythrocytes by structurally defined hyaluronic acid dodecasaccharides. Infect. Immun. 69, 420–425 Achur, R.N. et al. (2000) Characterization of proteoglycans of human placenta and identification of unique chondroitin sulfate proteoglycans of the intervillous spaces that mediate the adherence of Plasmodium falciparum-infected erythrocytes to the placenta. J. Biol. Chem. 275, 40344–40356 Grau, G.E. et al. (1989) Tumor necrosis factor and disease severity in children with falciparum malaria. New Engl. J. Med. 320, 1586–1591 Mohamadzadeh, M. et al. (1998) Proinflammatory stimuli regulate endothelial hyaluronan expression and CD44/HA-dependent primary adhesion. J. Clin. Invest. 101, 97–108 Maubert, B. et al. (1997) Cytoadherence of Plasmodium falciparum to intercellular adhesion molecule 1 and chondroitin-4-sulfate expressed by the syncytiotrophoblast in the human placenta. Infect. Immun. 65, 1251–1257 McCormick, C.J. et al. (1997) Intercellular adhesion molecule-1 and CD36 synergize to mediate adherence of Plasmodium falciparuminfected erythrocytes to cultured human microvascular endothelial cells. J. Clin. Invest. 100, 2521–2529

39 Pouvelle, B. et al. (2000) Cytoadhesion of Plasmodium falciparum ring-stage-infected erythrocytes. Nat. Med. 6, 1264–1268 40 Miller, L.H. et al. (1971) Alteration in the rheologic properties of Plasmodium knowlesi-infected red cells. A possible mechanism for capillary obstruction. J. Clin. Invest. 50, 1451–1455 41 Fried, M. et al. (1998) Maternal antibodies block malaria. Nature 395, 851–852 42 Chaiyaroj, S.C. et al. (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 43 Silamut, K. et al. (1999) A quantitative analysis of the microvascular sequestration of malaria parasites in the human brain. Am. J. Pathol. 155, 395–410 44 Traore, B. et al. (2000) Cytoadherence characteristics of Plasmodium falciparum isolates in Thailand using an in vitro human lung endothelial cells model. Am. J. Trop. Med. Hyg. 62, 38–44 45 Karamanos, N.K. et al. (1994) Determination of 24 variously sulfated galactosaminoglycan- and hyaluronan-derived disaccharides by highperformance liquid chromatography. Anal. Biochem. 221, 189–199 46 Beeson, J.G. et al. (1998) Inhibition of binding of malaria-infected erythrocytes by a tetradecasaccharide fraction from chondroitin sulfate A. Infect. Immun. 66, 3397–3402 47 Alkhalil, A. et al. (2000) Structural requirements for the adherence of Plasmodium falciparuminfected erythrocytes to chondroitin sulfate proteoglycans of human placenta. J. Biol. Chem. 275, 40357–40364 48 McKee, C.M. et al. (1996) Hyaluronan (HA) fragments induce chemokine gene expression in alveolar macrophages. The role of HA size and CD44. J. Clin. Invest. 98, 2403–2413 49 Ramsey, E.M. and Donner, M.W. (1980) Placental Vasculature and Circulation. Anatomy, Physiology, Radiology, Clinical Aspects: Atlas and Textbook, Georg Thieme Publishers 50 Cooke, B.M. et al. (1996) Adhesion of malariainfected red blood cells to chondroitin sulfate A under flow conditions. Blood 88, 4040–4044 51 Rogerson, S.J. et al. (1997) Plasmodium falciparum-infected erythrocytes adhere to the proteoglycan thrombomodulin in static and flowbased systems. Exp. Parasitol. 86, 8–18 52 Southwell, B. et al. (1989) Field applications of agglutination and cytoadherence assays with Plasmodium falciparum from Papua New Guinea. Trans. R. Soc. Trop. Med. Hyg. 83, 464–469

337

53 Udeinya, I.J. et al. (1983) Plasmodium falciparum strain-specific antibody blocks binding of infected erythrocytes to amelanotic melanoma cells. Nature 303, 429–431 54 Sartelet, H. et al. (2000) Hyperexpression of ICAM-1 and CD36 in placentas infected with Plasmodium falciparum: a possible role of these molecules in sequestration of infected red blood cells in placentas. Histology 36, 62–68 55 Newbold, C.I. et al. (1997) PfEMP1, polymorphism and pathogenesis. Ann. Trop. Med. Parasitol. 91, 551–557 56 Baruch, D.I. et al. (1995) Cloning the P. falciparum gene encoding PfEMP1, a malarial variant antigen and adherence receptor on the surface of parasitized human erythrocytes. Cell 82, 77–87 57 Smith, J.D. et al. (1995) Switches in expression of Plasmodium falciparum var genes correlate with changes in antigenic and cytoadherent phenotypes of infected erythrocytes. Cell 82, 101–110 58 Su, X.Z. et al. (1995) The large diverse gene family var encodes proteins involved in cytoadherence and antigenic variation of Plasmodium falciparum-infected erythrocytes. Cell 82, 89–100 59 Reeder, J.C. et al. (1999) The adhesion of Plasmodium falciparum-infected erythrocytes to chondroitin sulfate A is mediated by PfEMP1. Proc. Natl. Acad. Sci. U. S. A. 96, 5198–5202 60 Buffet, P.A. et al. (1999) Plasmodium falciparum domain mediating adhesion to chondroitin sulfate A: a receptor for human placental infection. Proc. Natl. Acad. Sci. U. S. A. 96, 12743–12748 61 Reeder, J.C. et al. (2000) Identification of glycosaminoglycan binding domains in Plasmodium falciparum erythrocyte membrane protein 1 of a chondroitin sulfate A-adherent parasite. Infect. Immun. 68, 3923–3926 62 Degen, R. et al. (2000) Plasmodium falciparum: cloned and expressed CIDR domains of PfEMP1 bind to chondroitin sulfate A. Exp. Parasitol. 95, 113–121 63 Triglia, T. et al. (1991) Amplification of the multidrug resistance gene pfmdr1 in Plasmodium falciparum has arisen as multiple independent events. Mol. Cell. Biol. 11, 5244–5250 64 Noviyanti, R.N. et al. Multiple var gene transcripts are expressed in Plasmodium falciparum-infected erythrocytes selected for adhesion. Mol. Biochem. Parasitol. (in press) 65 Smith, J.D. et al. (2000) Classification of adhesive domains in the Plasmodium falciparum erythrocyte membrane protein 1 family. Mol. Biochem. Parasitol. 110, 293–310

Articles of interest in other Trends journals • Interpreting cell wall 'virulence factors' of Mycobacterium tuberculosis, by C.E. Barry III (2001) Trends in Microbiology 9, 237–241 • How should pathogen transmission be modelled, by H. McCallum, N. Barlow and J. Hone (2001) Trends in Ecology & Evolution 16, 295–300 • The meaning and impact of the human genome for microbiology, by D.A. Relman and S. Falkow (2001) Trends in Microbiology 9, 206–208 • Intracellular survival strategies of mutualistic and parasitic prokaryotes, by W. Goebel and R. Gross (2001) Trends in Microbiology 9, 267–273

http://parasites.trends.com