Complement receptor 1 is the host erythrocyte
receptor for Plasmodium falciparum PfRh4
Wai-Hong Thama, Danny W. Wilsona, Sash Lopatickia, Christoph Q. Schmidtb, Patience B. Tetteh-Quarcoob,1,
Paul N. Barlowb, Dave Richarda, Jason E. Corbina, James G. Beesona,c, and Alan F. Cowmana,c,2
aThe Walter and Eliza Hall Institute of Medical Research, Parkville 3052, Australia;bSchools of Chemistry and Biological Sciences, University of Edinburgh,
Edinburgh EH9 3JJ, United Kingdom; andcDepartment of Medical Biology, University of Melbourne, Parkville 3052, Australia
Edited* by Louis H. Miller, National Institutes of Health, Rockville, MD, and approved August 27, 2010 (received for review June 10, 2010)
Plasmodium falciparum is responsible for the most severe form of
malaria diseasein humans,causingmorethan 1 milliondeathseach
year. As an obligate intracellular parasite, P. falciparum’s ability to
invade erythrocytes is essential for its survival within the human
host. P. falciparum invades erythrocytes using multiple host recep-
tor–parasite ligand interactions known as invasion pathways. Here
P. falciparum ligand essential for sialic acid–independent invasion.
strongly correlated with the CR1 level on the erythrocyte surface.
Parasite invasion via sialic acid–independent pathways is reduced
in low-CR1 erythrocytes due to limited availability of this receptor
on the surface. Furthermore, soluble CR1 can competitively block
binding of PfRh4 to the erythrocyte surface and specifically inhibit
sialic acid–independent parasite invasion. These results demon-
strate that CR1 is an erythrocyte receptor used by the parasite li-
gand PfRh4 for P. falciparum invasion.
malaria|red blood cell|merozoite|reticulocyte-binding-like homologue
P.falciparuminvade erythrocytesthrough amultistepprocessthat
involves initial contact with the erythrocyte, apical reorientation,
and formation of a tight junction that moves progressively toward
the posterior end of the parasite until host cell membrane fusion
is complete (see ref. 1 for a review). These steps in invasion are
dependent on specific interactions between multiple parasite in-
vasion ligands and their respective human erythrocyte receptors,
which have been defined as distinct invasion pathways.
In P. falciparum, two gene families encode important proteins
used in invasion: the erythrocyte-binding–like antigens (EBAs:
reticulocyte-binding–like homolog proteins (RBPs, or PfRhs:
PfRh1, PfRh2a, PfRh2b, PfRh4, and PfRh5) (2–5). During in-
and are able to bind erythrocytes. Invasion pathways have been
identified by examining the entry of merozoites into erythrocytes
that have deficient or mutant host receptors or that have been
treated with enzymes that modify the properties or presence of
erythrocyte surface proteins. The most common enzyme treat-
ments involve neuraminidase, which removes sialic acid residues,
backbones of proteins. At present, only a handful of erythrocyte
receptors that bind to P. falciparum invasion ligands have been
EBL-1, and glycophorin C for EBA-140 (6–8). All three of these
interactions are sensitive to neuraminidase treatment of eryth-
rocytes and thus are involved in sialic acid–dependent invasion
a receptor for the sialic acid–independent invasion pathways in
multiple laboratory strains and wild isolates, although the parasite
ligand with which it interacts has not yet been identified (9). CR1
rythrocyte invasion is essential for the survival of Plasmodium
falciparum within the human host. The merozoite forms of
also mediates rosetting through its interaction with PfEMP-1,
a parasite-derived variant erythrocyte membrane protein (10).
On the erythrocyte surface, CR1 is present as a ∼190- to 280-kDa
single-chain transmembrane glycoprotein bearing the Knops
blood group (11).
PfRh4 is essential in the sialic acid-independent pathway as
to switch invasion pathways to allow invasion into neuraminidase-
treated erythrocytes (12). Growth assays in the presence of anti-
PfRh4 antibodies have shown that PfRh4 is the major ligand re-
sponsible for invasion via the sialic acid–independent pathways
(50–80%, depending on the parasite strain used) (13). By activat-
ing PfRh4 expression, the parasite is able to switch receptor usage
from sialic acid–dependent to sialic acid-independent pathways,
thereby providing a mechanism for the parasite to invade via dif-
essential component of the life cycle of P. falciparum, the use of
multiple redundant invasion pathways and the ability to switch
pathways through differential expression of parasite ligands pro-
vides the parasite with mechanisms to increase successful invasion
in the face of host immune responses and erythrocyte receptor
polymorphisms in malaria endemic regions (SI Text, Fig. S1).
Identifying the parasite ligand–host erythrocyte receptors used
in invasion is a crucial step toward a complete understanding of
the full repertoire of invasion pathways available to P. falciparum.
Because PfRh4 is a key player in phenotypic variation of invasion
and sialic acid–independent pathway, we sought to identify its
host erythrocyte receptor. We observed that erythroid CR1 is sus-
ceptible to trypsin and chymotrypsin treatment, an enzyme profile
consistent with that obtained for the abolishment of PfRh4
erythrocyte binding (13, 14). Here we show that PfRh4 binds to
a functional sialic acid–independent pathway for P. falciparum
invasion of human erythrocytes.
PfRh4 Binds to CR1 on the Erythrocyte Surface. To examine whether
CR1 is a receptor for PfRh4, we incubated erythrocytes with anti-
CR1 antibodies, then performed an erythrocyte-binding assay
Author contributions: W.-H.T., S.L., P.N.B., J.G.B., and A.F.C. designed research; W.-H.T.,
S.L., D.W.W., C.Q.S., P.B.T.-Q., and J.E.C. performed research; S.L. and D.R. contributed
new reagents/analytic tools; W.-H.T., D.W.W., C.Q.S., P.B.T.-Q., P.N.B., J.G.B., and A.F.C.
analyzed data; and W.-H.T., P.N.B., J.G.B., and A.F.C. wrote the paper .
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
Freely available online through the PNAS open access option.
1Present address: Department of Microbiology, University of Ghana Medical School,
2To whom correspondence should be addressed. E-mail: email@example.com.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
| October 5, 2010
| vol. 107
| no. 40
using culture supernatants containing P. falciparum invasion
ligands released from merozoites (Fig. 1A). Using erythrocytes
precoated with anti-CR1 antibodies, we found that PfRh4 eryth-
rocyte binding was reduced, whereas EBA-175 binding was not
perturbed (Fig. 1A). The addition of increasing amounts of anti-
CR1 antibodies resulted in enhanced inhibition of PfRh4 eryth-
rocyte binding, with >90% of binding blocked at an antibody dose
of 0.0107 mg/mL (Fig. 1A). Preincubation of erythrocytes with
glycophorin A/B or decay accelerating factor (DAF) antibodies
did not affect PfRh4 binding, indicating that the inhibition ob-
served with anti-CR1 antibodies is specific to PfRh4 (Fig. 1B).
We also tested the ability of soluble recombinant CR1 to inhibit
binding of PfRh4 to the surface of human erythrocytes. Soluble
CR1(sCR1) contains the extracellular domain butlacks thetrans-
membrane and cytoplasmic domains (15). Culture supernatants
were preincubated with sCR1 before being added to erythrocyte-
native PfRh4 binding to CR1 on erythrocytes, with >90% inhib-
ition obtained at 0.04 mg/mL (Fig. 1C). Probing of the same
binding eluates with EBA-175 antibodies resulted in no pertur-
bation of EBA-175 erythrocyte binding, indicating that the in-
hibition is specific to PfRh4. Thus, PfRh4 binds to CR1 on the
Recombinant PfRh4 and CR1 Interact Directly. To provide additional
evidence of a PfRh4–CR1 interaction, we performed immuno-
precipitation experiments using sCR1 and recombinant PfRh4,
which is functional, as demonstrated by its ability to bind eryth-
rocytes (Figs. S2 and S3) (13, 15). Using anti-PfRh4 monoclonal
sCR1 was immunoprecipitated (Fig. 2A). Given CR1’s ability to
bind immune complexes, we showed, as a control, that the 10C9
monoclonal antibody does not nonspecifically immunoprecipitate
sCR1 (Fig. 2A). For an antibody control, we used anti-Rh4
monoclonal antibody 2E8, which does not recognize recombinant
PfRh4; this did not immunoprecipitate the CR1–PfRh4 complex
also isolated a complex containing recombinant PfRh4 and sCR1,
but not another hexaHis-tagged control protein, further support-
ing the specificity of the PfRh4–CR1 interaction (Fig. 2B).
based assay, in which sCR1 bound to immobilized recombinant
PfRh4 but not to the control protein, and this binding increased
with increasing concentrations of sCR1 (Fig. 2C and Fig. S3). This
result was reinforced by surface plasmon resonance (SPR)-based
PfRh4 that has been immobilized via amine coupling to a sensor
chip. Measuring the binding of an sCR1 concentration series (at
two different loadings on the chip of recombinant PfRh4) yielded
a Kdof 2.9 ± 0.2 μM for the sCR1–PfRh4 interaction (Fig. 2D).
In an attempt to delineate the region within CR1 involved in
PfRh4 binding, we measured the CR1–PfRh4 interaction in the
presence of anti-CR1 monoclonal antibodies mapped to the ex-
tracellular domain of CR1 (Fig. S3) (16). In the most common
isoform of CR1, the extracellular domain comprises 30 short
consensus repeat modules (SCRs). Based on a high degree of
internal homology, all except the last two carboxy terminal SCRs
which is composed of seven SCRs. The subclass for the mouse
was IgG2a. The addition of anti-CR1 monoclonal antibodies
or Ig2a mouse isotype resulted in three distinct phenotypes for
CR1–PfRh4 interaction in the ELISA-based assay. First, the ad-
dition of 4D6 (which recognizes SCRs 3, 10, and 17) or Ig2a
mouse isotype did not affect this interaction. Second, the CR1–
PfRh4 interaction was enhanced in the presence of 1B4, 3D9
(which also recognize SCRs 3, 10, and 17), and HB8592. Third,
the CR1–PfRh4 interaction was disrupted by 7G9 and E11
monoclonal antibodies (Fig. 2E). These monoclonal antibodies
Although HB8592 recognizes the same class of epitopes as E11
(SCRs 5–7, 12–14, 19–21, and 26–28), it did not inhibit the in-
teraction between sCR1 and PfRh4. This may be because E11 is
reactivity only to recombinant LHR D (16). Although this finding
implicates one or more of SCRs 5–7, 12–14, and 19–21 of CR1 as
the region involved in PfRh4 binding, the binding of antibodies to
CR1 SCRs adjacent to (rather than within) ligand-binding sites
for the inhibition of CR1–PfRh4 interaction, the addition of E11
and 7G9 produced a reduction in PfRh4 binding in standard
erythrocyte-binding assays (Fig. 2F).
Levels of PfRh4 Binding Were Strongly Correlated with CR1 Level on
the Erythrocyte Surface. CR1 levels on erythrocytes vary between
individuals in the range of 50–1,200 molecules per cell. In Cau-
casian populations, single nucleotide polymorphism (SNP) within
exon 22 in the CR1 gene is linked to high CR1 expression (H, high
allele) or low CR1 expression (L, low allele) (17). Homozygous
HH individuals have higher erythrocyte surface levels of CR1,
homozygous LL individuals have <200 molecules per erythrocyte,
whether differential expression of CR1 on erythrocytes correlated
with PfRh4 binding, we analyzed the CR1 phenotype of blood
samples from Australian residents. CR1 erythrocyte levels from
these individuals showed an association with exon 22 genotyping
(Fig. 3A). We observed markedly reduced native PfRh4 binding in
erythrocytes from a LL individual compared with those from an
Fig. S4D). We probed the same binding eluates with EBA-175
antibodies, but found no difference in EBA-175 binding.
To measure the level of PfRh4 erythrocyte binding in more
blood samples, we developed a FACS-based assay for erythrocyte
Anti-CR1 antibody (E11) inhibits PfRh4 erythrocyte binding (Left). Increasing
concentrations of anti-CR1 monoclonal enhanced the reduction in PfRh4
erythrocyte binding (Right). Anti-CR1 monoclonal antibodies at 0–0.0107 mg/
mL were incubated with erythrocytes before the addition of culture super-
natants. (B) PfRh4 erythrocyte binding is not affected by preincubation of
glycophorin A/B (glyA/B) or DAF monoclonal antibodies at final concen-
trations of 0.03 mg/mL. (C) Native PfRh4 binding to CR1 is inhibited by sCR1.
Competitive binding assays with sCR1 were performed by incubating sCR1
with invasion supernatants at the stated final concentrations (0–0.04 mg/mL).
In A and C, the gray numbers under each top panel represent the percentage
of PfRh4 binding relative to the no-antibody lane or the no-sCR1 lane, re-
spectively. In all panels, immunodetection of parasite proteins with anti-
PfRh4 and anti–EBA-175 antibodies after erythrocyte binding is shown.
Anti-CR1 antibodies and sCR1 inhibit PfRh4 erythrocyte binding. (A)
| www.pnas.org/cgi/doi/10.1073/pnas.1008151107Tham et al.
binding using recombinant PfRh4 (Fig. S2) (18). The features of
recombinant PfRh4 erythrocyte binding in a FACS-based assay
are similar to those of standard binding assays in terms of enzyme
sensitivity (Fig. S2). Analysis of 100 individual blood samples
revealed a strong correlation between the percentage of eryth-
rocytes that bound recombinant PfRh4 and the CR1 level on the
erythrocyte surface (r2= 0.8236; Fig. 3C and Fig. S2). In contrast,
there was no significant correlation between recombinant PfRh4
This indicates that the level of native or recombinant PfRh4
binding to erythrocytes is dependent on the amount of CR1
expressed on the erythrocyte.
Sialic Acid-Independent Invasion Was Reduced in Low-CR1 Ery-
throcytes and Inhibited in the Presence of Soluble CR1. To exam-
ine the role of CR1 in P. falciparum invasion, we screened 400
blood samples from nonoverlapping individuals to identify addi-
tional low-CR1 erythrocytes. We selected the 10 samples with the
highest CR1 expression and the 10 samples with the lowest CR1
expression on the erythrocyte surface; in most cases, CR1 phe-
PfRh4 and sCR1. The anti-PfRh4 monoclonal 10C9 does not immunoprecipitate sCR1 in the absence of recombinant PfRh4. (B) Anti-CR1 monoclonal antibody
HB8592 immunoprecipitates sCR1 and recombinant PfRh4, but not another hexaHis-tagged protein (control). (C) An ELISA- based assay measuring the in-
teraction between sCR1 and recombinant PfRh4. Microtiter wells were coated with recombinant PfRh4 or control hexaHis-tagged protein at 0.5 μg/well. sCR1
was added at 0–1 μg/well. Bound CR1 was detected with an anti-CR1 monoclonal HB8592. (D) Use of SPR to measure the dissociation constant (Kd) of sCR1 for
PfRh4. (Left) Duplicate sensorgrams for a range of increasing sCR1 concentrations (0.5, 1.0, 2.5, 5.0, and 10 μM, from bottom to top) flowed over a CM5 chip
surface with a loading (via amine coupling) of 1,476 response units (RU) of recombinant PfRh4. (Right) Plots of the RUs obtained versus sCR1 concentrations
on two different flow cells, coupled with 408 RUs (lower curve) and 1476 RUs (upper curve) of recombinant PfRh4. The dashed vertical line indicates the Kd
fitted to both plots simultaneously. (E) Binding of sCR1 to recombinant PfRh4 was inhibited by anti-CR1 monoclonal antibodies. Microtiter plates were coated
with saturating concentration of recombinant PfRh4 (5 μg/well). Anti-CR1 monoclonal antibodies and Ig2a mouse isotype at concentrations of 0–1 μg/well
were incubated with sCR1 (0.2 μg/well) before being added to the wells. Interaction between sCR1 and PfRh4 was detected using anti-CR1 antibody 6B1
conjugated directly to HRP. 6B1 shares no similar epitopes with the other anti-CR1 antibodies. All ELISA experiments (C and E) were repeated, and similar
results were obtained. In C and E, the y axis indicates absorbance measured at 405 nm. Error bars represent the range of duplicate readings. (F) Anti-CR1
antibodies inhibit native PfRh4 erythrocyte binding in an epitope-specific manner. For erythrocyte-binding inhibition assays, the monoclonal antibodies were
incubated with erythrocytes at 0.03 mg/mL before the culture supernatants were added. Lane 0 has no antibody added. Immunodetection of parasite
proteins with anti-PfRh4 and anti–EBA-175 antibodies after erythrocyte binding is shown.
PfRh4 and CR1 directly interact. (A) Immunoprecipitation with anti-PfRh4 monoclonal antibody 10C9 isolates a complex containing both recombinant
Tham et al.PNAS
| October 5, 2010
| vol. 107
| no. 40
notype was associated with exon 22 genotyping (Fig. S4). These
erythrocyte samples were within normal ranges for blood indices
and glucose-6-phosphate dehydrogenase (G6PD) activity and
werewild type forotherblood polymorphisms,including Gerbich,
South Asian Ovalocytosis (SAO), and α-thalassemia (Fig. S5).
Although low-CR1 erythrocytes had higher levels of G6PD ac-
tivity, this phenotype should not negatively affect parasite growth.
Furthermore, none of the blood samples were microcytic or
macrocytic. The level of recombinant PfRh4 binding was again
strongly associated with CR1 expression, whereas binding of
recombinant PfRh4 and EBA-175 did not correlate with levels of
glycophorin C and CR1 expression, respectively (Figs. S4 and S5).
To examine the importance of CR1 as a receptor for the PfRh4
invasion pathway, we evaluated the invasion of parasite strains
W2mefΔ175 and 3D7 into neuraminidase-treated erythrocytes
(Fig. 4A). These strains have been shown to efficiently invade
neuraminidase-treated erythrocytes, and PfRh4 is known to be an
essential invasion ligand for this sialic acid–independent path-
way (12, 19). Thus, if CR1 is the receptor for PfRh4, then this
invasion pathway will be less efficient in the presence of neur-
aminidase-treated low-CR1 erythrocytes compared with invasion
into neuraminidase-treated high-CR1 erythrocytes, due to the
reduced availability of its receptor. Because both high-CR1 and
low-CR1 erythrocytes showed variation in parasite growth rate,
the efficiency of sialic acid-independent invasion in each sample
(i.e.,the invasion ratio) wascalculated as a ratioof thepercentage
of parasitemia in neuraminidase-treated erythrocytes divided by
the percentage of parasitemia in untreated erythrocytes using the
same strain. These results are shown for high-CR1 and low-CR1
erythrocytes in Fig. 4A. For both W2mefΔ175 and 3D7, invasion
into neuraminidase-treated low-CR1 erythrocytes was reduced
compared with invasion into neuraminidase-treated high-CR1
erythrocytes (P = 0.0009 and 0.003, respectively). This decreased
efficiency of the PfRh4 invasion pathway results from the limited
availability of CR1 on the erythrocyte surface.
and 3D7 in invasion assays into untreated or neuraminidase-treated
erythrocytes (Fig. 4B). Anti-PfRh4 antibodies were shown to inhibit
invasion of strains 3D7 (untreated), W2mefΔ175 (neuraminidase-
treated) and 3D7 (neuraminidase-treated) to 20%, 60%, and 80%,
erythrocyte surface. (A) Erythrocyte CR1 levels in relation to CR1 genotyping
at exon 22 for 80 samples. Each point represents the average of MFI from
duplicate readings. H, the high-CR1 allele; L, the low-CR1 allele; n, number of
samples; n/a, not applicable. (B) Binding of native PfRh4 to erythrocytes from
LL and HH individuals. Immunodetection of parasite proteins with anti-
PfRh4 and anti–EBA-175 antibodies after erythrocyte binding is shown. (C)
The percentage of recombinant PfRh4-bound erythrocytes (x axis) correlates
with the CR1 level on the erythrocyte surface (y axis). Recombinant PfRh4
was added at 0.2 mg/mL to erythrocytes before proceeding with the FACS-
based erythrocyte-binding assay, and binding was detected using anti-PfRh4
monoclonal antibody. (D) Percentage of recombinant PfRh4-bound eryth-
rocytes (x axis) does not correlate with glycophorin C expression (y axis). In C
and D, r2is a measure of the goodness of fit of linear regression.
The level of PfRh4 binding correlates with CR1 expression on the
and inhibited in the presence of sCR1. (A) Parasite growth is reduced into
neuraminidase-treated low-CR1 erythrocytes. The efficiency of sialic acid-
independent invasion in each sample (i.e., the invasion ratio) was calculated
as the ratio of the percentage of parasitemia in neuraminidase (Nm)-treated
erythrocytes divided by the percentage of parasitemia in untreated eryth-
rocytes using the same strain. These results are shown for invasion into high-
CR1 erythrocytes (H, black circles) and low-CR1 erythrocytes (L, white circles)
for both W2mefΔ175 (left y axis) and 3D7 (right y axis). The mean ± SEM
invasion rates are shown. P values were determined using the unpaired
Student t test. (B) The PfRh4 invasion pathway is inhibited in the presence of
sCR1. Parasite strains W2mef, W2mefΔRh4, W2mefΔ175, and 3D7 were
tested in growth assays into untreated or Nm-treated erythrocytes in the
presence of 0.5 mg/mL of sCR1 (white bars) or control protein (black bars).
(C) Inhibition of parasite growth by sCR1 is concentration-dependent. In-
vasion of 3D7 into untreated (un) or Nm-treated erythrocytes was measured
in the presence of sCR1at the stated 0–0.5 mg/mL final concentration. In B
and C, growth is measured as the percent of noninhibitory PBS control, and
error bars represent the SEM from two separate experiments (in triplicate).
Sialic acid-independent invasion is reduced in low-CR1 erythrocytes
| www.pnas.org/cgi/doi/10.1073/pnas.1008151107Tham et al.
respectively, but no inhibition was observed for W2mefΔRh4, Download full-text
W2mef, and W2mefΔ175 into untreated erythrocytes (13). This
indicates that invasion of W2mefΔRh4, W2mef, and W2mefΔ175
into untreated erythrocytes is not dependent on PfRh4, whereas
invasion of 3D7 (untreated), W2mefΔ175 (neuraminidase-treated),
and 3D7 (neuraminidase-treated) is increasingly dependent on
rocytes by both 3D7 and W2mefΔ175 was greatly reduced in the
presence of sCR1 (87% inhibition, P = 0.001 and 81% inhibition,
of control protein. This inhibition of parasite invasion could be ti-
trated, with decreasing amounts of sCR1 resulting in less invasion
antibodies, although the difference was not statistically significant
(20% inhibition; P = 0.09). As controls, we observed no significant
erythrocytes with W2mef, W2mefΔRh4, and W2mefΔ175, strains
0.352, respectively). These resultsdemonstrate that CR1 is the host
erythrocyte receptor used by PfRh4,and that this pathway functions
directly in P. falciparum invasion.
Invasion of P. falciparum into human erythrocytes requires spe-
cific ligand–receptor interactions. The PfRh family of proteins is
a group of key ligands important to invasion by direct binding to
the erythrocyte. Up to now, no receptors have been identified,
however. Here we show that CR1 serves as a receptor for P. fal-
ciparum invasion via direct binding of the parasite ligand PfRh4.
The level of PfRh4 binding to erythrocytes is strongly correlated
with the amount of CR1 molecules on the erythrocyte surfaces. In
the presence of sCR1, low inhibition of parasite invasion into
untreated erythrocytes was observed in sialic acid–independent
strains 3D7 and W2mefΔ175. This is related to the redundancy
of EBA and PfRh proteins, so that inhibition of one pathway is
compensated for by the function of others. Treatment of eryth-
rocytes with neuraminidase resulted in a blockage of invasion by
such ligands as EBA-175, EBA-181, EBA-140, and PfRh1, which
require sialic acid-containing receptors. As a result, parasite in-
vasion into neuraminidase-treated erythrocytes was increasingly
reliant on the PfRh4–CR1 pathway, as demonstrated by the
consistent with data reported by Spadafora et al. (9) showing that
the addition of sCR1 inhibited parasite invasion into neuramini-
dase-treated erythrocytes in sialic acid-independent strains, such
as 7G8 and 3D7. sCR1 was not able to inhibit invasion in strains
that either lack PfRh4 expression (W2mef) or possess a genetic
knockout of the PfRh4 gene (W2mefΔRh4), providing convinc-
ing evidence that the reduction in invasion was due to the
sCR1–PfRh4 interaction. We also show that binding of PfRh4 to
erythrocytes, as well as PfRh4–CR1 interactions, can be blocked
by the presence of sCR1 or anti-CR1 monoclonal antibodies,
thus providing a molecular basis for the inhibition of invasion.
These results indicate that the PfRh4–CR1 interaction is a func-
tional sialic acid–independent invasion pathway for the entry of
P. falciparum into human erythrocytes. Although neuraminidase-
treated erythrocytes do not exist in the field, this in vitro situation
mimics situations in which host immune responses selectively
block the function of specific parasite invasion pathways or eryth-
rocyte polymorphisms in surface proteins result in a limited re-
CR1 has been shown to mediate rosetting through its in-
teraction with PfEMP-1, a parasite-derived variant erythrocyte
membranevirulence protein (10). Rosetting occurs wheninfected
erythrocytes adhere to uninfected erythrocytes, resulting in
“clumps.” The formation of rosettes can be disrupted by the ad-
dition of sCR1 and is reduced in the presence of low-CR1 ery-
throcytes (10). The CR1 region required for rosetting is SCRs
10 and 17. Anti-CR1 monoclonal antibodies that map to those
regions (1B4, 3D9, and 4D6) do not inhibit either PfRh4 eryth-
rocyte binding or the CR1–PfRh4 interaction, suggesting that the
regions of binding to PfEMP-1 and PfRh4 might be distinct sites
within CR1 (20). However, anti-CR1 monoclonal antibody J3B11
has been shown to inhibit parasite invasion into neuraminidase-
treated erythrocytes and also to interfere with PfEMP-1 binding
(9, 20). A major caveat with all of these experiments is that anti-
CR1 monoclonal antibodies that recognize the same epitopes
show contrasting results in terms of inhibition of rosetting, PfRh4
erythrocyte binding, and invasion assays (9, 20). In addition, it has
been shown that binding of anti-CR1 antibodies to SCRs adjacent
to (rather than within) ligand-binding sites can perturb function
(16). Finer epitope mapping of the anti-CR1 monoclonal anti-
bodies and further studies using functional protein domains of
CR1will help elucidatethepreciselocationwithinCR1thatbinds
PfRh4. Of note, P. falciparum uses the same host erythrocyte
receptorfortwodistinct functions important forthevirulence and
survival of the parasite, which probably exploits the essential na-
ture of CR1 on erythrocytes, given the absence of CR1 null
erythrocytes in the human population.
Field isolates from India, Gambia, Brazil, Tanzania, and Kenya
use many different invasion pathways, as demonstrated by the
ability of these field parasites to invade a wide variety of enzyme-
treated erythrocytes (21–25). Interestingly, a majority of Kenyan
P.falciparumisolates invade via a neuraminidase-resistant, trypsin-
sensitive, and chymotrypsin-sensitive pathway—an enzyme profile
identical to that observed for PfRh4 erythrocyte binding (19 out
of 31 isolates) (22). It is highly likely that PfRh4 is involved in this
invasion pathway, given the fact that no other PfRh or EBA has
a comparable enzyme profile. In support of this possibility, Spa-
dafora et al. (9) reported the inhibition of invasion of three clinical
isolates into both intact and neuraminidase-treated erythrocytes
by the addition of sCR1. Determining the prevalence of the CR1–
PfRh4 pathway in field isolates would require growth assays on
more samples in the presence of inhibitory anti-PfRh4 antibodies
or with the addition of sCR1 (13).
To prevent invasion, the human population relies on two pro-
tective mechanisms: a polymorphic erythrocyte surface and anti-
body responses to block parasite ligand function. Blood poly-
morphisms prevalent in malaria-endemic regions, such as Gerbich
invasion via EBA-140 (26). Human sera from malaria-exposed
individuals contain antibodies to P. falciparum ligands and inhibit
invasion in both sialic acid-dependent and –independent pathways
(27). Thus, receptor polymorphisms and host immune responses
against individual ligands might lead to inhibition of particular in-
vasion pathways. The population of Madang, PNG, has the lowest
CR1 levels reported to date (28). Our results indicate that the
PfRh4–CR1 invasion pathway was less effective in the presence of
low-CR1 erythrocytes, most likely due to limited receptor avail-
ability. However, studies in PNG have found no evidence for dif-
ferences in parasite densities among HH, HL and LL genotypes
forCR1,although theHLgenotype issignificantly protectedfrom
severe malaria (29). The lack of correlation between protection
for other blood polymorphisms, including α-thalassemia and
SAO, and remains a confounding dilemma in the field (30, 31). In
terms of CR1 polymorphisms, the lack of measurable differences
in parasite densities is likely due to the fact that HH, HL, and LL
erythrocytes from PNG individuals all have low CR1 levels (<400
molecules per cell) compared with Caucasian individuals. The
importance of PfRh4 as an invasion ligand in the PNG population
is indicated by the strong reactivity to PfRh4 of sera from immune
individuals (13). To date, no study has investigated the invasion
phenotypes of PNG field isolates, and further work is needed to
determine the relevance of these protective mechanisms.
Tham et al.PNAS
| October 5, 2010
| vol. 107
| no. 40
Here we show that CR1 is a receptor used for P. falciparum
invasion into erythrocytes. CR1 belongs to the regulator of com-
plement activation (RCA) protein family and has been charac-
terized as a negative regulator of complement activation. Several
pathogens are known to bind to RCA proteins as a means to
facilitate entry into host cells or to down-regulate comple-
ment activation (reviewed in ref. 32). RCA proteins, such as
DAF, are present on erythrocytes and also may be involved in
P. falciparum invasion.
Materials and Methods
See SI Materials and Methods for more detailed information.
Erythrocyte-Binding Assays. Erythrocyte-binding assays and enzymatic treat-
ment of erythrocytes were performed as described previously (13). For the
antibody inhibition of erythrocyte binding, anti-CR1, anti-glycophorin A/B
monoclonal, and anti-DAF monoclonal antibodies were preincubated with
erythrocytes before the addition of culture supernatants. Competitive
binding assays with sCR1 were performed by incubating sCR1 with culture
supernatants before proceeding with standard erythrocyte-binding assays.
The FACS-based erythrocyte-binding assay for P. falciparum is a modification
of the protocol used for Duffy antigen binding (18).
In Vitro Immunoprecipitation. Recombinant proteins used in immunopreci-
pitations were incubated together in T-NET buffer [1% Triton X-100, 150 mM
NaCl, 10 mM EDTA, and 50 mM Tris (pH 7.4)]. Immunoprecipitation was per-
formed using anti-CR1 (HB8592) or anti-PfRh4 (10C9 and 2E8) monoclonal
antibody coupled to Protein G/A Sepharose beads and then eluted in glycine
elution buffer. Eluted proteins were resuspended in nonreducing sample
buffer, run on SDS/PAGE gels, and evaluated by immunoblot analysis.
ELISA. Microtiter plates were coated with recombinant fusion protein,
blocked, and incubated with ligand. Protein interactions were detected using
a specific monoclonal antibody and an HRP-conjugated secondary antibody.
Azino-bis-3-ethylbenthiazoline-6-sulfonic acid was used to detect HRP ac-
tivity, and optical density was measured at 405 nm.
SPR. Binding of sCR1 for PfRh4 was monitored by SPR using the Biacore T100
system (GE Healthcare). The sensor chip surfaces were prepared by immobi-
lizing recombinant PfRh4 at different loading densities on two of the four
flow cells of a Biacore Series S carboxymethylated dextran (CM5) sensor chip,
using standard amine coupling. Experiments were performed at 25 °C at and
a flow rate of 30 μL/minute. Duplicate injections of sCR1 samples in 10 mM
Hepes-buffered 150 mM saline with 3 mM EDTA and 0.005% (vol/vol) poly-
sorbate 20 (pH 7.4) were performed at the concentrations indicated in Fig. 2.
Data were processed using Biacore T100 evaluation software version 2.0.
Dissociation constants were calculated by fitting steady-state binding levels
derived from the background-subtracted traces to a one-to-one binding
Growth Assays. Parasite growth assays were performed over two cycles of
parasite growth as described previously (27).
ACKNOWLEDGMENTS. We thank Dr. Henry C. Marsh (Celldex Therapeutics,
Needham, MA) for providing sCR1, CR1 monoclonal antibodies 4D6 and
6B1, and comments on the manuscript. We thank Ronald Taylor (University
of Virginia School of Medicine, Charlottesville, VA) and Thalachallour
Mohanakumar (University of Virginia School of Medicine, Charlottesville,
VA) for CR1 monoclonal antibodies 1B4, 3D9, 7G9, 9H3, and KuN241; the
Australian Red Cross (Dr. Jenny Condon) for the blood samples; The Walter
and Eliza Hall Institute’s Monoclonal Facility (The Walter and Eliza Hall
Institute of Medical Research, Melbourne) for antibody production; and
Stuart Wyithe (Department of Physics, University of Melbourne, Mel-
bourne) for statistical analysis. Infrastructure was supported by Victoria
State Government Offer Information Statement and National Health and
Medical Research Council Independent Research Institute Infrastructure
Support Scheme grants. A.F.C. is a Howard Hughes International Scholar
and an Australia Fellow of the National Health and Medical Research Coun-
cil. J.G.B. is supported by a National Health and Medical Research Council
Career Development award. This work was supported by the National
Health and Medical Research Council, the Darwin Trust of Edinburgh (stu-
dentship to P.B.T-Q.), and the Wellcome Trust.
1. Cowman AF, Crabb BS (2006) Invasion of red blood cells by malaria parasites. Cell 124:
2. Peterson DS, Wellems TE (2000) EBL-1, a putative erythrocyte-binding protein of
Plasmodium falciparum, maps within a favored linkage group in two genetic crosses.
Mol Biochem Parasitol 105:105–113.
3. Adams JH, et al. (1992) A family of erythrocyte-binding proteins of malaria parasites.
Proc Natl Acad Sci USA 89:7085–7089.
4. Rayner JC, Galinski MR, Ingravallo P, Barnwell JW (2000) Two Plasmodium falciparum
genes express merozoite proteins that are related to Plasmodium vivax and
Plasmodium yoelii adhesive proteins involved in host cell selection and invasion. Proc
Natl Acad Sci USA 97:9648–9653.
5. Rayner JC, Vargas-Serrato E, Huber CS, Galinski MR, Barnwell JW (2001) A Plasmodium
falciparum homologue of Plasmodium vivax reticulocyte binding protein (PvRBP1)
defines a trypsin-resistant erythrocyte invasion pathway. J Exp Med 194:1571–1581.
6. Maier AG, et al. (2003) Plasmodium falciparum erythrocyte invasion through gly-
cophorin C and selection for Gerbich negativity in human populations. Nat Med 9:
7. Mayer DC, et al. (2009) Glycophorin B is the erythrocyte receptor of Plasmodium
falciparum erythrocyte-binding ligand, EBL-1. Proc Natl Acad Sci USA 106:5348–5352.
8. Sim BKL, Chitnis CE, Wasniowska K, Hadley TJ, Miller LH (1994) Receptor and ligand
domains for invasion of erythrocytes by Plasmodium falciparum. Science 264:
9. Spadafora C, et al. (2010) Complement receptor 1 is a sialic acid–independent
erythrocyte receptor of Plasmodium falciparum. PLoS Pathog 6:e1000968.
10. Rowe JA, Moulds JM, Newbold CI, Miller LH (1997) P. falciparum rosetting mediated
by a parasite-variant erythrocyte membrane protein and complement-receptor 1.
11. Krych-Goldberg M, AtkinsonJP (2001)
complement receptor type 1. Immunol Rev 180:112–122.
12. Stubbs J, et al. (2005) Molecular mechanism for switching of P. falciparum invasion
pathways into human erythrocytes. Science 309:1384–1387.
13. Tham WH, et al. (2009) Antibodies to reticulocyte-binding protein-like homologue 4
inhibit invasion of Plasmodium falciparum into human erythrocytes. Infect Immun 77:
14. Gaur D, et al. (2007) Recombinant Plasmodium falciparum reticulocyte homology
protein 4 binds to erythrocytes and blocks invasion. Proc Natl Acad Sci USA 104:
15. Weisman HF, et al. (1990) Soluble human complement receptor type 1: In vivo
inhibitor of complement suppressing post-ischemic myocardial inflammation and
necrosis. Science 249:146–151.
16. Nickells M, et al. (1998) Mapping epitopes for 20 monoclonal antibodies to CR1. Clin
Exp Immunol 112:27–33.
17. Xiang L, Rundles JR, Hamilton DR, Wilson JG (1999) Quantitative alleles of CR1:
Coding sequence analysis and comparison of haplotypes in two ethnic groups. J
18. Tran TM, et al. (2005) Detection of a Plasmodium vivax erythrocyte-binding protein by
flow cytometry. Cytometry A 63:59–66.
19. Duraisingh MT, Maier AG, Triglia T, Cowman AF (2003) Erythrocyte-binding antigen
175 mediates invasion in Plasmodium falciparum utilizing sialic acid–dependent
and –independent pathways. Proc Natl Acad Sci USA 100:4796–4801.
20. Rowe JA, et al. (2000) Mapping of the region of complement receptor (CR) 1 required
for Plasmodium falciparum rosetting and demonstration of the importance of CR1 in
rosetting in field isolates. J Immunol 165:6341–6346.
21. Lobo CA, et al. (2004) Invasion profiles of Brazilian field isolates of Plasmodium
falciparum: Phenotypic and genotypic analyses. Infect Immun 72:5886–5891.
22. Deans AM, et al. (2007) Invasion pathways and malaria severity in Kenyan
Plasmodium falciparum clinical isolates. Infect Immun 75:3014–3020.
23. Gomez-Escobar N, et al. (2010) Erythrocyte invasion and merozoite ligand gene
expression in severe and mild Plasmodium falciparum malaria. J Infect Dis 201:
24. Bei AK, et al. (2007) Variant merozoite protein expression is associated with
erythrocyte invasion phenotypes in Plasmodium falciparum isolates from Tanzania.
Mol Biochem Parasitol 153:66–71.
25. Okoyeh JN, Pillai CR, Chitnis CE (1999) Plasmodium falciparum field isolates commonly
use erythrocyte invasion pathways that are independent of sialic acid residues of
glycophorin A. Infect Immun 67:5784–5791.
26. Serjeantson SW, White BS, Bhatia K, Trent RJ (1994) A 3.5-kb deletion in the
glycophorin C gene accounts for the Gerbich-negative blood group in Melanesians.
Immunol Cell Biol 72:23–27.
27. Persson KE, et al. (2008) Variation in use of erythrocyte invasion pathways by
Plasmodium falciparum mediates evasion of human inhibitory antibodies. J Clin
28. Cockburn IA, et al. (2004) A human complement receptor 1 polymorphism that
reduces Plasmodium falciparum rosetting confers protection against severe malaria.
Proc Natl Acad Sci USA 101:272–277.
29. Fowkes FJ, et al. (2008) Host erythrocyte polymorphisms and exposure to Plasmodium
falciparum in Papua New Guinea. Malar J 7:1.
30. Genton B, et al. (1995) Ovalocytosis and cerebral malaria. Nature 378:564–565.
31. Allen SJ, et al. (1997) α+-Thalassemia protects children against disease caused by other
infections as well as malaria. Proc Natl Acad Sci USA 94:14736–14741.
32. Lambris JD, Ricklin D, Geisbrecht BV (2008) Complement evasion by human
pathogens. Nat Rev Microbiol 6:132–142.
| www.pnas.org/cgi/doi/10.1073/pnas.1008151107Tham et al.