Genetic susceptibility to systemic lupus
erythematosus protects against cerebral
malaria in mice
Michael Waisberga, Tatyana Tarasenkoa, Brandi K. Vickersa, Bethany L. Scotta, Lisa C. Willcocksb, Alvaro Molina-Cruzc,
Matthew A. Piercea, Chiung-yu Huangd, Fernando J. Torres-Veleze, Kenneth G. C. Smithb, Carolina Barillas-Muryc,
Louis H. Millerc,1, Susan K. Piercea, and Silvia Bollanda
Laboratories ofaImmunogenetics andcMalaria and Vector Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health,
Rockville, MD 20852;bCambridge Institute for Medical Research and the Department of Medicine, Addenbrooke’s Hospital, University of Cambridge School
of Clinical Medicine, Cambridge CB2 2OY, United Kingdom;dBiostatistics Research Branch, National Institute of Allergy and Infectious Diseases, National
Institutes of Health, Bethesda, MD, 20817; andeComparative Medicine Branch, National Institute of Allergy and Infectious Diseases, National Institutes of
Health, Bethesda, MD, 20892
Contributed by Louis H. Miller, December 9, 2010 (sent for review October 14, 2010)
Plasmodium falciparum has exerted tremendous selective pressure
on genes that improve survival in severe malarial infections. Sys-
temic lupus erythematosus (SLE) is an autoimmune disease that is
sixto eight times more prevalent in women ofAfrican descentthan
netic susceptibility to SLE protects against cerebral malaria. Mice
that are prone to SLE because of a deficiency in FcγRIIB or overex-
pression of Toll-like receptor 7 are protected from death caused by
cerebral malaria. Protection appears to be by immune mechanisms
that allow SLE-prone mice better to control their overall inflamma-
tory responses to parasite infections. These findings suggest that
the high prevalence of SLE in women of African descent living out-
side of Africa may result from the inheritance of genes that are
beneficial in the immune control of cerebral malaria but that, in
the absence of malaria, contribute to autoimmune disease.
dium falciparum (Pf), the most deadly species of Plasmodium
that prevails in Africa. The high mortality associated with
malaria has exerted an enormous selective pressure on genes
that protect against severe malaria even though they may be
deleterious in other regards (1). This phenomenon perhaps is
best illustrated by the high prevalence in Africa of the hemo-
globin S allele (Hb-S). Although Hb-S homozygosity is lethal in
children in West Africa (causing sickle cell anemia), Hb-S is
maintained at an 18% gene frequency because in the heterozy-
gous state Hb-S confers protection against severe malaria and
death (2). Pf infections result in a spectrum of diseases ranging
from mild, uncomplicated malaria to severe malaria including
cerebral malaria and respiratory distress, the primary causes of
death from malaria in African children (3). Children in Africa
appear to acquire immunity to severe malaria by age 5 y, but
immunity to mild malaria requires additional years of repeated
exposures (4), suggesting that different immune mechanisms
may be at work in controlling severe versus mild malaria. Be-
cause severe malaria kills children, it is anticipated that any gene
that contributes to immune protection from severe disease would
be selected for. Such genes would be beneficial to individuals
living in malaria-endemic areas, who frequently are infected with
Pf. What is not known is whether such genes might have dele-
terious effects in individuals of African descent who no longer
live in a malaria-endemic environment. Although we assume that
the genetic selective pressure of malaria would be gender neu-
tral, one of the best examples of the deleterious effects immune
selection in Africans may be systemic lupus erythematosus
(SLE). SLE is an autoimmune disease that is six to eight times
more prevalent in women of African descent living outside
Africa than in women of European descent (5), even though
little autoimmune disease is reported in Africa (6). The in-
creased risk of SLE in women of African descent correlates with
alaria, an infectious disease that kills nearly 1 million
children each year in Africa alone, is caused by Plasmo-
the portion of the genome that is of West African ancestry, in-
dicating a genetic basis for this SLE susceptibility (7). To date,
however, the West African genes that account for the high risk of
SLE in women of African descent have not been identified (5).
Together, these observations suggest that SLE-susceptibility
genes may protect against severe malaria but that in the absence
of continual exposure to Pf such genes may contribute to hy-
perimmune responses typical of systemic autoimmune disease.
To test directly if SLE susceptibility protects against severe,
lethal malaria, we evaluated three mouse strains with defined
genetic alterations that have been shown to develop lupus glo-
merulonephritis with pathology similar to human SLE in
a mouse model of cerebral malaria. We analyzed FcγRIIB−/−.yaa
mice that are deficient in FcγRIIB, an inhibitory receptor that is
central to the control of humoral immune responses (8).
FcγRIIB was of particular interest because a human allele that
encodes a polymorphism in the transmembrane domain in
FcγRIIB that results in a loss of function (9, 10) is significantly
more common in Africans (11, 12) and in African Americans (9)
than in Europeans and is associated with SLE in Asia (13). In-
deed, FcγRIIB deficiency in mice has been shown to reduce the
severity of nonfatal Plasmodium chabaudi infections that corre-
lated with increased levels of the proinflammatory cytokine
TNF-α and with increased antibody levels (11). Willcocks et al.
(12) recently provided evidence that a loss-of-function poly-
morphism in FcγRIIB is associated with protection from severe
malaria in African children. In addition to the FcγRIIB de-
ficiency, FcγRIIB−/−.yaa mice have the Y chromosome-linked
genetic modifier Yaa, a duplication in the gene that encodes
Toll-like receptor 7 (TLR7) (14), a member of one family of the
innate immune system’s pathogen-associated molecular pattern
recognition receptors. We also evaluated mice deficient only in
FcγRIIB (FcγRIIB−/−mice) (15) and mice with multiple (ap-
proximately six or seven) copies of the gene encoding TLR7
(TLR7.tg mice) (16). Here we provide evidence that these SLE-
prone mice are protected against cerebral malaria. Overall, SLE-
prone mice appeared to be protected because of their ability to
control their inflammatory responses to parasite infection. These
results support the view that the high risk of SLE in women of
African descent is related to the protective value of SLE high-
risk genotypes in severe malaria.
Author contributions: M.W., T.T., K.G.C.S., C.B.-M., L.H.M., S.K.P., and S.B. designed re-
search; M.W., T.T., B.K.V., B.L.S., L.C.W., A.M.-C., M.A.P., and F.J.T.-V. performed research;
M.W., T.T., C.-y.H., F.J.T.-V., K.G.C.S., C.B.-M., L.H.M., S.K.P., and S.B. analyzed data; and
M.W., K.G.C.S., C.B.-M., L.H.M., S.K.P., and S.B. wrote the paper.
The authors declare no conflict of interest.
Freely available online through the PNAS open access option.
1To whom correspondence should be addressed: E-mail: firstname.lastname@example.org.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
| January 18, 2011
| vol. 108
| no. 3www.pnas.org/cgi/doi/10.1073/pnas.1017996108
Mice with a Genetic Predisposition to SLE Are Protected from
Cerebral Malaria. Infection with the ANKA strain of Plasmo-
dium berghei (P. berghei ANKA) that causes cerebral malaria in
susceptible strains of mice is characterized by rapid death, within
≈10 d of infection, at a relatively low parasitemia accompanied
by brain microhemorrhages (17). In resistant strains of mice,
P. berghei ANKA infections result in later death, ≈15 d after
infection, caused by high parasitemia and severe anemia without
evidence of brain microhemorrhages. At equivalent levels of
parasitemia, susceptible strains of mice also show parasitized
RBCs in the brain, but resistant mice do not (17).
WT mice, B6 mice, and FcγRIIB−/−.yaa mice (SLE-susceptible
mice on the B6 background) were infected with P. berghei ANKA
at age 10–12 wk. Infected FcγRIIB−/−.yaa mice died significantly
later than WT mice (P < 0.0001) (Fig. 1A). The median survival
time was 15 d in 12 P. berghei ANKA-infected FcγRIIB−/−.yaa
mice and 9.5 d in infected WT mice. No FcγRIIB−/−.yaa mouse
survived the infection, as was expected, because even cerebral
malaria-resistant mice with P. berghei ANKA infections die be-
cause of severe anemia. FcγRIIB−/−.yaa and WT mice developed
parasitemia at similar rates (Fig. 1B), but the WT mice died
at significantly lower parasitemias (% parasitized RBC at
death = 19 ± 15) than FcγRIIB−/−.yaa mice (% parasitized RBC
at death = 42 ± 16) (P = 0.002).
After infection, FcγRIIB−/−.yaa animals developed spleno-
megaly with spleen weights significantly greater than infected
WT mice measured at day 7 postinfection (Fig. 1D). Spleen size
measured at autopsy correlated well with survival times for
infected mice (Spearman’s ρ = 0.73) (Fig. 1D). We also mea-
sured postmortem weights of brains and kidneys and found that
the brains of FcγRIIB−/−.yaa animals were lighter than those of
WT mice (Fig. 1C) (P = 0.001), possibly suggesting less edema.
We did not observe significant differences in the weights of
kidneys (P = 0.54). In contrast to WT mice, all FcγRIIB−/−.yaa
animals had developed moderate kidney inflammation (an early
sign of SLE) at the time of infection (Fig. S1).
In addition to causing rapid death at relatively low para-
sitemias, cerebral malaria is characterized by microhemorrhages
in the brain (17). Analyses of brain pathology at the time of
autopsy showed that the number of microhemorrhages was sig-
nificantly greater in WT mice than in FcγRIIB−/−.yaa mice (Fig.
2). An example of a microhemorrhage in the brain of an infected
WT mouse is shown in Fig. 2, A and B. Microhemorrhages were
quantified and scored on a scale of 0–3 based on the number of
microhemorrhages observed in four tissue sections. The scores of
WT mice were significantly higher than those of FcγRIIB−/−.yaa
mice (Fig. 2C), and the brain microhemorrhage scores were in-
versely correlated with survival (Spearman’s ρ = 0.71). Collec-
tively, the low brain microhemorrhage scores, prolonged
survival, and relatively high parasitemias at death indicate that
the FcγRIIB−/−.yaa mice are resistant to cerebral malaria and die
of severe anemia and high parasitemia. We also observed that
more FcγRIIB−/−.yaa mice than WT mice developed focal men-
ingeal mononuclear infiltrates, although the difference did not
reach statistical significance (P = 0.09) (Fig. 2 E–G). Mononuclear
infiltrates into the meninges have been reported in susceptible B6
mice protected from lethal P. berghei ANKA cerebral malaria by
vaccination with killed blood-stage parasites (18).
The results presented thus far show that FcγRIIB−/−.yaa mice
are resistant to cerebral malaria but are unable to control high
parasitemia in P. berghei infections. We directly evaluated the
susceptibility of FcγRIIB−/−.yaa mice to infection with the 17XL
strain of Plasmodium yoelii (P. yoelii 17XL) in which the primary
cause of death is severe anemia resulting from high parasitemia,
with no cerebral involvement. FcγRIIB−/−.yaa mice had no sur-
vival advantage over WT mice in P. yoelii 17XL infections (Fig.
3A). FcγRIIB−/−.yaa and WT mice died at the same rate, with
100% of animals dying by day 8 after infection with similar
parasitemias. FcγRIIB−/−.yaa mice had higher spleen weights than
WT mice (P = 0.03), but these weights did not correlate with
survival (Spearman’s ρ = 0.37; P = 0.12). Brains of FcγRIIB−/−
animals were lighter than those of WT mice (P = 0.005), but
kidney weights in the two groups were not different (Fig. S2).
Increased TLR7 Gene Copy Number and FcγRIIb Deficiency Independently
Protect Against Cerebral Malaria. To determine if the protection
against cerebral malaria observed in FcγRIIB−/−.yaa mice was
caused by the absence of FcγRIIB or the increase in TLR7 gene
copy number, we determined the survival times for TLR7.tg and
FcγRIIB−/−mice in P. berghei ANKA infections. Both TLR7.tg and
FcγRIIB−/−mice develop SLE-like diseases, but disease is somewhat
less severe in TLR7.tg mice than in FcγRIIB−/−.yaa mice as judged
by mean survival time (FcγRIIB−/−.yaa, 4–5 mo; TLR7.tg, 7–8 mo;
and FcγRIIB−/−, 8–9 mo). TLR7.tg mice tended to survive P. berghei
ANKA infections longer than control WT mice (Fig. 3B): TLR7.tg
mice had a median survival time of 10 d following infection com-
pared with 7 d for WT control mice. To determine if TLR7.tg mice
were protected from cerebral disease, the brains were examined at
autopsy. The analysis showed microhemorrhages significantly more
earlier and at lower parasitemias than FcγRIIB−/−.yaa
mice. FcγRIIB−/−.yaa ● (n = 12) and WT mice □ (n = 12)
were infected with 1 × 106P. berghei ANKA iRBCs on
day 0. (A) The percent of FcγRIIB−/−.yaa and WT mice
that survived over time is given in Kaplan–Meier
curves. Death was defined as parasitemia ≥60%,
moribundity, or death (P < 0.0001). (B) Parasitemias,
determined by Giemsa-stained smears of tail blood,
are given for FcγRIIB−/−.yaa mice and WT mice over
time. (C) The weights of brains, kidneys, and spleens
of mice determined at autopsy. (D) The weights of
spleens of FcγRIIB−/−.yaa and WT mice determined at
day 7 postinfection. *P < 0.05; **P < 0.01.
In P. berghei ANKA infections, WT mice die
Waisberg et al. PNAS
| January 18, 2011
| vol. 108
| no. 3
in WT than in TLR7.tg mice (Fig. 2D). Taken together with the
observation that TLR7.tg and WT mice developed parasitemias at
similar rates (Fig. 3B) but that TLR7.tg mice died with significantly
higher parasitemias than WT mice (% parasitemia at death 56 ± 8
for TLR7.tg mice vs. 28 ± 18 for WT mice; P = 0.001), these data
indicate that TLR7.tg mice are resistant to cerebral malaria and die
of severe anemia caused by high parasitemia. On autopsy, TLR7.tg
mice also had larger spleens than control mice (P = 0.003), and
spleen size correlated well with survival times (Spearman’s ρ =
0.79) (Fig. S3). FcγRIIB-deficient mice also survived P. berghei
survival times (Spearman’s ρ = 0.80) (Fig. S3), developed para-
sitemias at similar rates (Fig. 3B) but died with higher parasitemias
than WT mice (% parasitemia at death = 63 ± 18 vs. 37 ± 30; P =
0.03). Neither TLR7.tg nor FcγRIIB−/−mice presented signs of overt
autoimmune disease, including kidney inflammation and autoanti-
bodies, at the age they were used in our experiments (8).
Characterization of Immune Responses in FcγRIIB−/−.yaa Mice. At
present the immune mechanisms at play in cerebral malaria in
either humans or in mouse models are not well understood. In
mouse models there is considerable evidence that immune
responses to infection, which play a role in the control of para-
sitemia, also result in strong inflammatory responses and high
levels of inflammatory cytokines. The inflammatory response and
the sequestration of infected RBCs (iRBCs) in the brain con-
tribute to cerebral malaria (19). We measured a number of
cytokines in the peripheral blood of animals before and 5 d after
infection with P. berghei ANKA to determine if any cytokines
were associated with protection against cerebral malaria (Fig. 4
and Table S1). Before infection, only the levels of GM-CSF and
IL-10 were significantly different in FcγRIIB−/−.yaa and WT mice
(P = 0.002), with FcγRIIB−/−.yaa mice having higher levels of
both cytokines (Fig. 4 and Table S1). After infection, however,
than FcγRIIB−/−.yaa or TLR7.tg mice. (A and B) Typical brain sections of WT
(n = 12) P. berghei-infected mice. The arrows in A indicate typical micro-
hemorrhages found in WT brains; these microhemorrhages are magnified in
B. (C) Microhemorrhages were quantified, given scores of 0–3 based on the
number of hemorrhages in four tissue sections (0 = none; 1 = fewer than
five; 2 = 5–10; 3 = more than 10 microhemorrhages). The average scores of
FcγRIIB−/−.yaa and WT mice are given (P = 0.04). Shown are the mean and SD.
*P < 0.05. (D) Average microhemorrhage scores for TLR7.tg and WT mice (P =
0.046). (E and F) Typical brain sections of FcγRIIB−/−.yaa mice infected with
P. berghei (n = 12). Arrows in E indicate typical meningeal infiltrates found
in FcγRIIB−/−.yaa brains; these infiltrates are magnified in F. (G) Meningeal
infiltrates were evaluated and given a binary score (0 = absent; 1 = present).
The percent of WT and FcγRIIB−/−.yaa mice showing focal meningeal infil-
trates in brain sections is given.
At death, WT mice have significantly more brain microhemorrhages
infection, and FcγRIIB-deficiency and TLR7 overexpression independently
protect against cerebral malaria. (A) FcγRIIb−/−.yaa (n = 10) and WT mice (n =
9) were infected with 1 × 104P. yoelii 17 XL iRBCs. (Left) Percent of FcγRIIB−/−.
yaa and WT mice that survived over time after infection is shown in Kaplan–
Meier curves (P = 0.80). (Right) Parasitemias are given with time after in-
fection. (B) TLR7.tg (n = 10) and WT mice (n = 10) were infected with 1 × 106
P. berghei ANKA iRBCs. (Left) Percent of mice surviving over time is shown in
Kaplan–Meier curves (P = 0.07). (Right) Parasitemias are given with time
after infection. (C) FcγRIIB−/−and WT mice were infected with 1 × 106P.
berghei ANKA iRBCs. (Left) Percent of FcγRIIB−/−and WT mice that survived
over time after infection is shown in Kaplan–Meier survival curves (*P =
0.003). (Right) Parasitemias with time after infection.
FcγRIIB−/−.yaa and WT mice are equally susceptible to P. yoelli XL17
| www.pnas.org/cgi/doi/10.1073/pnas.1017996108 Waisberg et al.
the levels of GM-CSF were equivalent in FcγRIIB−/−.yaa and
WT mice, and the levels of IL-10 in WT mice rose to levels
greater than those in either uninfected or infected FcγRIIB−/−.
yaa mice (P = 0.01). After infection WT mice also had signifi-
cantly higher levels of IFN-γ, IL-12p70, IL-17, IL-6, monocyte
chemotactic protein-1 (MCP-1), and regulated upon activation,
normal T-cell expressed and secreted (RANTES) than FcγRIIB−/−.
yaa mice. It is of interest that after infection FcγRIIB−/−.yaa mice
showed pronounced T helper 17 (Th17) skewing as compared with
infected WT mice. It may be that the higher levels of the anti-in-
flammatory cytokine IL-10 before infection and the lower levels of
the pro-inflammatory cytokines IL-6, MCP-1, RANTES, and IFN-
γ postinfection in FcγRIIB−/−.yaa mice played a role in their pro-
tection from cerebral malaria.
To identify alterations in immune cell populations that might
correlate with protection from cerebral malaria, we carried out
extensive phenotyping by flow cytometry of immune cells in the
spleens of FcγRIIB−/−.yaa and WT mice before and 7 d after in-
fection with P. berghei ANKA (Fig. 5A and Table S2). Most cell
subpopulations, including B220+GL7+germinal center B cells,
B220intCD138+plasma cells, B220+CD69+activated B cells,
cytotoxic T cells, were increased in FcγRIIB−/−.yaa mice as com-
pared with WTmice before infection but after infection increased
to similar levels in both infected WT and FcγRIIB−/−.yaa mice.
However, there were fewer natural killer cells in infected
FcγRIIB−/−.yaa mice than in infected WT mice, and infection re-
duced the percent of forkhead box P3 (FoxP3)-positive regulatory
T cells (Tregs) in WT mice but had no effect in FcγRIIB−/−.yaa
mice. The pattern of expression of inducible costimulator (ICOS)
was of interest (Fig. 5B). Strikingly, most CD4+T cells from un-
infected FcγRIIB−/−.yaa animals expressed ICOS at intermediate
yaa mice before infection. Infection of FcγRIIB−/−.yaa mice had
little effect on ICOS expression (Fig. 5B), with the majority of
T cells continuing to express intermediate levels of ICOS. In
contrast, T cells in uninfected WT mice did not express ICOS, but
infection resulted in an increase in ICOS expression to high levels
in 50% of the WT CD4+T cells. ICOShighT cells synthesize IL-2,
IL-3,andIFN-γ (20), consistent withthe observation that infected
WT mice produce more IFN-γ than do uninfected mice. Similarly
the small number of ICOShighT cells in infected FcγRIIB−/−.yaa
mice may explain why those mice produced less IFN-γ than
infected WT mice. Although uninfected FcγRIIB−/−.yaa mice had
significantly more CD4+CD44+activated T cells than did WT
cells in both FcγRIIB−/−.yaa and WT mice (Fig. 5B). Collectively,
the characteristics of the splenic immune cell populations suggest
that cells from FcγRIIB−/−.yaa mice had features of a controlled
chronic activation before infection that were absent in WT mice
and that infection of FcγRIIB−/−.yaa mice resulted in reduced in-
flammatory responses as compared with WT mice.
The lower levels of inflammatory cytokines, particularly MCP-
1 and RANTES, observed in the peripheral blood of infected
FcγRIIB−/−.yaa as compared with WT mice could indicate re-
duced recruitment of inflammatory and cytotoxic immune cells
to the brain and might explain the reduced hemorrhagic pa-
thology of FcγRIIB−/−.yaa mice. We examined the immune cell
populations in the brains of WT and FcγRIIB−/−.yaa mice before
and 7 d after infection with P. berghei ANKA. We observed that
CD4+and CD8+T cells were recruited to the brains of both WT
and FcγRIIB−/−.yaa infected mice (Fig. 6A). To determine
whether the T cells present in the brains of WT and FcγRIIB−/−.
yaa mice were differentially activated, we measured the mRNA
levels of granzyme B, IL-10, IFN-γ, IL-6, and IL-12 in the brains
of these mice before and after infection. We found no significant
differential expression of these mRNAs in WT and FcγRIIB−/−.
log10( (pg mL) )
GMCSF IFNγ γ
Day 5 post-infection
log10( (pg mL) )
GMCSF IFNγ γ
Fcγ γRII .yaa
Fcγ γRII .yaa
P. berghei ANKA-infected FcγRIIB−/−.yaa (n =
10) and WT (n = 10) mice. Serum cytokine
levels were measured before infection with
P. berghei ANKA iRBCs and 5 d postinfection
using a multiplex assay. Shown are the log of
the concentrations of GM-CSF, INF-γ, IL-10,
IL-12p70, IL-17, IL-6, MCP-1, and RANTES in
pg/mL Thick lines within the boxes represent
themedians ofthe data, theboxes represent
the upper and lower quartiles, dotted lines
represent the largest and smallest non-out-
lier observations, and open dots represent
outliers (n = 10) for each group.
Cytokine profiles in uninfected and
cerebral malaria. (A) Immune cell populations were analyzed by flow
cytometry. Average values of three mice per group are shown. Values are
given relative to the number of splenocytes in uninfected WT mice or the
percentage in WT mice of each subpopulation Absolute cell numbers are
provided in Table S2. (B) The histograms show fluorescence intensity for ICOS
(Left) and CD44 (Right) gating on CD4+T cells. One representative experi-
ment of six is shown.
Alterations in spleen cell phenotypes related to protection from
Waisberg et al.PNAS
| January 18, 2011
| vol. 108
| no. 3
yaa mice and provide the results for granzyme B to illustrate this
point (Fig. 6). Granzyme B levels are low in the brain of un-
infected WT and FcγRIIB−/−.yaa mice and rise to similar levels in
malaria-infected mice. Thus, although similar numbers of T
cells in similar activation states are recruited to the brains of WT
and FcγRIIB−/−.yaa mice, cerebral disease is controlled only in
The number of F4/80+macrophages in the brain did not in-
crease appreciably in infected WT animals but was somewhat
increased in FcγRIIB−/−.yaa mice both before and after infection.
It has been reported that M2 macrophages can decrease T cell-
mediated central nervous system disease (21), causing an in-
crease in the number of Tregs and increases in both IL-10 and
GM-CSF, characteristics that we observed in FcγRIIB−/−.yaa
mice before infection (Fig. 4). M2 macrophages are character-
ized by high levels of arginase mRNA (22). We detected M2
macrophages in the brains of FcγRIIB−/−.yaa mice at autopsy and
found a twofold increase in arginase mRNA in both infected and
uninfected FcγRIIB−/−.yaa mice relative to WT mice (Fig. 6B),
implicating M2 macrophages in the protection of FcγRIIB−/−.yaa
mice from cerebral malaria.
Here we provide evidence that genetic alterations that are
responsible for the development of an SLE-like autoimmune
disease in mice, namely an FcγRIIB-deficiency and TLR7 du-
plication, are protective against cerebral malaria. FcγRIIB−/−is
a potent SLE-susceptibility gene capable of interacting with
a variety of other loci to modify the initiation and progression of
autoimmune disease (8). A single nucleotide polymorphism in
the gene encoding the human FcγRIIB that abrogates receptor
function is strongly associated with susceptibility to SLE in both
Caucasians and Southeast Asians (13). Willcocks et al. (12) re-
cently reported that the minor allele of this polymorphism is
more common in Southeast Asians and Africans than in Cau-
casians and that homozygosity for the minor allele is associated
with substantial protection from severe malaria in children in
Kenya (12), although this study was not powered to detect
associations with a specific type of severe malaria, for example
cerebral malaria versus severe malaria anemia. Consistent with
the observation by Willcocks et al., it was shown earlier that
macrophages from individuals homozygous for this poly-
morphism efficiently engulfed Pf trophozoites (11). Collectively,
these findings and those reported here may help explain the high
prevalence of SLE among individuals of African descent living
Despite the prevalence of SLE in women of African descent,
current evidence indicates that there is little autoimmunity in
Africa (6, 23). Indeed, SLE is highly prevalent only in Africans
living outside Pf-endemic areas. These observations suggest that
chronic reinfection with Pf, as occurs in Africa, may attenuate
the potential of SLE high-risk genes to cause autoimmunity. If
this speculation is correct, it would be predicted that malaria
infections attenuate autoimmunity. Using the spontaneous au-
toimmune disease models of New Zealand Black (NZB) and
(NZB × New Zealand White) F1 hybrid mice and the nonlethal
mouse strains of P. yoelii, Greenwood et al. (24) provided evi-
dence that malaria infections have some protective effects
against the development of autoimmune disease. Thus, malaria
may suppress autoimmunity in individuals with SLE-suscepti-
bility alleles. It will be of interest to determine what features
of malaria parasite infections are responsible for attenuating
from cerebral malaria. In mice, cerebral malaria is characterized
brain barrier, sequestration of iRBCs in the brain microvessels,
accumulation of leukocytes in the brain, and the production of
proinflammatory cytokines (25, 26). The analyses of a variety of
mice deficient in cytokines (27, 28), chemokines and chemokine
receptors (29–31), and TLRs (32) have implicated a variety of
led to a clear picture of the underlying cause of disease pathology.
Althoughthegenetic alterations intheSLE-prone micedescribed
here are well defined, the cellular and molecular mechanism by
which these alterations result in autoimmunity is not fully un-
derstood.The immune parameterdata provided heresuggest that
FcγRIIB−/−.yaa mice develop chronic inflammation with a Th17
skewing and a concomitant induction of regulatory pathways that
reduce the highly Th1-biased inflammatory responses in cerebral
malaria.We speculate thatFcγRIIB−/−.yaamicemay beprotected
from cerebral malaria by decreasing the overall inflammatory re-
sponse to the parasite. The feedback regulatory pathway that
reduces acute inflammatory responses in FcγRIIB−/−.yaa mice
could involve IL-10, because it has been shown that this pathway
provides a feedback control of autoimmune responses (33). Reg-
macrophages have been shown to attenuate Tcell-mediated brain
disease in experimental autoimmune encephalomyelitis (21).
In conclusion, the results presented here provide evidence for
a relationship between SLE high-risk alleles and protection
against cerebral malaria. At present, the association of genes
with SLE susceptibility in women of African descent remains
largely unexplored. Once identified, such genes could be tested
for their protective value against cerebral malaria in African
Relative Granzyme B
% Brain Mononuclear
Mononuclear cells isolated from brains of uninfected (n = 5) and infected (n =
4) WT and uninfected (n = 5) and infected (n = 5) FcγRIIB−/−.yaa mice were
characterized by flow cytometry. Shown are the percent of live mononuclear
brain cells that were CD4+T cells, CD8+T cells, or F4/80+macrophages. Boxes
show the median and 25th and 75th percentiles; whiskers show the largest and
smallest values. *P < 0.05. (B) Granzyme B mRNA expression was measured in
brain samples of infected (n = 5) and uninfected (n = 2) WT mice and in
infected (n = 4) and uninfected (n = 3) FcγRIIB−/−.yaa mice. (C) Arginase mRNA
expression was measured in brain samples of infected (n = 5) and uninfected
(n = 2) WT mice and in infected (n = 4) and uninfected (n = 3) FcγRIIB−/−.yaa
mice. For B and C, expression of the L32 housekeeping gene was used as
standard in the calculations. Shown are means and SD. *P < 0.05.
Recruitment of immune cells to the brains of infected mice. (A)
| www.pnas.org/cgi/doi/10.1073/pnas.1017996108Waisberg et al.
children. Future progress in deciphering the cellular and mo-
lecular mechanisms underlying the relationships between host
susceptibility to autoimmunity and protection from cerebral
malaria may lead to new strategies for therapeutics both for SLE
and for cerebral malaria.
Materials and Methods
All experiments were approved by the National Institute of Allergy and In-
fectious Diseases Animal Care and Use Committee.
Animals. Male C57BL/6 mice (10- to 12-wk-old) were obtained from Jackson
Laboratories. FcγRIIB−/−.yaa, FcγRIIB−/−, and TLR7.tg mice were bred at the
Laboratory of Immunogenetics in the National Institute of Allergy and In-
fectious Diseases animal facility.
Malaria Infections. Mice were infected with P. berghei ANKA or P. yoelii 17XL
by injecting i.p. 1 × 106P. berghei ANKA or 1 × 104P. yoelii 17XL iRBCs
obtained from C57BL/6 or BALB/c mice, respectively. Parasitemias in infected
mice were quantified by examining Giemsa-stained blood smears.
Cytokine Measurements. Blood was collected on days 0, 3, and 5 postinfection,
and sera were stored at −80 °C until analyzed for IL-1–6, IL-10, IL-12p70, IL-
17, macrophage inflammatory protein 1α (MIP-1α), MCP-1, RANTES, GM-CSF,
IFN-γ, and TNF-α using the Q-Plex Human Citokyne kit (Quansys Biosciences)
according to the manufacturer’s instructions.
Real-Time PCR. Cells were lysed in TRIzol (Invitrogen), and total RNA was
isolated using the RNeasy Mini Kit (QIAGEN). One microgram of total RNA
was reverse transcribed using the iScript cDNA Synthesis Kit (Bio-Rad).
Quantitative PCR was performed using the real-time iCycler PCR platform
(Bio-Rad). The sequences of primers for arginase 1 (Arg1) were described
previously (34). Arg1 mRNA levels were normalized to that of L32 ribosomal
RNA. PCR primers for L32 were 5′-AGAGGACCAAGAAGTTCATCAGGC-3′
Pathology. Animals that reached 60% parasitemia, that became moribund, or
that were found dead were dissected, and kidneys, cerebrum, and spleens
were fixed in 10% neutral buffered formalin. The tissues were weighed,
sectioned, andstained with hematoxylin-eosin. The tissues were evaluated by
pathologists blinded to the experimental design. Four tissue sections per
animal were analyzed, and brain hemorrhages were scored on a scale from
0 to 3 where 0 represented no hemorrhages, 1 represented >5 hemorrhages,
2 represented 5–10 hemorrhages, and 3 represented >10 hemorrhages. The
degree of kidney inflammation was scored from 0 to 3 where 0 represents
no inflammation, 1 represents mild inflammation, 2 represents moderate
inflammation, and 3 represents severe inflammation. Meningeal infiltrates
were evaluated on a binary scale in which 0 indicated absent, and 1 indi-
Flow Cytometry Analysis. Single-cell suspensions from spleens were prepared
from infected and uninfected FcγRIIB−/−.yaa and WT mice. Mononuclear cells
were isolated from the brains of mice infected with malaria on day 7 post-
infection. The brains were homogenized, and mononuclear cells were iso-
lated using 40–70% discontinuous Percoll gradients (35). Total cell numbers
were determined by counting on a hemocytometer and were analyzed by
FACS. Antibodies to the following antigens were used for FACS analysis:
B220, CD4, CD8, CD11b, GL7, Fas, CD21, CD23, CD25, CD69, CD44, CD45RB,
CD62L, NK1.1, and ICOS (BD Pharmingen). For forkhead box P3 (FoxP3)
staining, an anti-mouse FoxP3 staining kit from Ebioscience, was used.
Samples were analyzed on a FACSCalibur flow cytometer (BD Biosciences),
and data were analyzed using FlowJo software (Tree Star Technologies).
Statistical Analysis. A detailed description of the statistical methods is pro-
vided in SI Materials and Methods.
ACKNOWLEDGMENTS. This study was supported by the Intramural Research
Program of the National Institutes of Health, National Institute of Allergy
and Infectious Diseases.
1. Kwiatkowski DP (2005) How malaria has affected the human genome and what
human genetics can teach us about malaria. Am J Hum Genet 77:171–192.
2. Molineaux L (1985) The impact of parasitic diseases and their control on mortality,
with emphasis on malaria and Africa. Health Policy, Social Policy, and Mortality
Prospects, eds Vallin J, Lopez A (Ordina Editions, Liege, Belgium), pp 13–44.
3. Marsh K, et al. (1995) Indicators of life-threatening malaria in African children. N Engl
J Med 332:1399–1404.
4. Langhorne J, Ndungu FM, Sponaas AM, Marsh K (2008) Immunity to malaria: More
questions than answers. Nat Immunol 9:725–732.
5. Molokhia M, McKeigue P (2006) Systemic lupus erythematosus: Genes versus
environment in high risk populations. Lupus 15:827–832.
6. Greenwood BM, Herrick EM, Voller A (1970) Can parasitic infection suppress
autoimmune disease? Proc R Soc Med 63:19–20.
7. Molokhia M, et al. (2003) Relation of risk of systemic lupus erythematosus to west
African admixture in a Caribbean population. Hum Genet 112:310–318.
8. Bolland S, Yim YS, Tus K, Wakeland EK, Ravetch JV (2002) Genetic modifiers of
systemic lupus erythematosus in FcgammaRIIB(-/-) mice. J Exp Med 195:1167–1174.
9. Li X, et al. (2003) A novel polymorphism in the Fcgamma receptor IIB (CD32B)
transmembrane region alters receptor signaling. Arthritis Rheum 48:3242–3252.
10. Floto RA, et al. (2005) Loss of function of a lupus-associated FcgammaRIIb
polymorphism through exclusion from lipid rafts. Nat Med 11:1056–1058.
11. Clatworthy MR, et al. (2007) Systemic lupus erythematosus-associated defects in the
inhibitory receptor FcgammaRIIb reduce susceptibility to malaria. Proc Natl Acad Sci
12. Willcocks LC, et al. (2010) A defunctioning polymorphism in FCGR2B is associated with
protection against malaria but susceptibility to systemic lupus erythematosus. Proc
Natl Acad Sci USA 107:7881–7885.
13. Niederer HA, et al. (2010) Copy number, linkage disequilibrium and disease
association in the FCGR locus. Hum Mol Genet 19:3282–3294.
14. Pisitkun P, et al. (2006) Autoreactive B cell responses to RNA-related antigens due to
TLR7 gene duplication. Science 312:1669–1672.
15. Bolland S, Ravetch JV (2000) Spontaneous autoimmune disease in Fc(gamma)RIIB-
deficient mice results from strain-specific epistasis. Immunity 13:277–285.
16. Deane JA, et al. (2007) Control of toll-like receptor 7 expression is essential to restrict
autoimmunity and dendritic cell proliferation. Immunity 27:801–810.
17. Engwerda CR, et al. (2002) Locally up-regulated lymphotoxin alpha, not systemic
tumor necrosis factor alpha, is the principle mediator of murine cerebral malaria.
J Exp Med 195:1371–1377.
18. Jennings VM, Lal AA, Hunter RL (1998) Evidence for multiple pathologic and
protective mechanisms of murine cerebral malaria. Infect Immun 66:5972–5979.
19. Schofield L, Grau GE (2005) Immunological processes in malaria pathogenesis. Nat Rev
20. Löhning M, et al. (2003) Expression of ICOS in vivo defines CD4+ effector T cells with
high inflammatory potential and a strong bias for secretion of interleukin 10. J Exp
21. Weber MS, et al. (2007) Type II monocytes modulate T cell-mediated central nervous
system autoimmune disease. Nat Med 13:935–943.
22. Ho VW, Sly LM (2009) Derivation and characterization of murine alternatively
activated (M2) macrophages. Methods Mol Biol 531:173–185.
23. Symmons DP (1995) Frequency of lupus in people of African origin. Lupus 4:
24. Greenwood BM, Herrick EM, Voller A (1970) Suppression of autoimmune disease in
NZB and (NZB x NZW) F1 hybrid mice by infection with malaria. Nature 226:266–267.
25. Hunt NH, Grau GE (2003) Cytokines: Accelerators and brakes in the pathogenesis of
cerebral malaria. Trends Immunol 24:491–499.
26. Combes V, De Souza JB, Renia L, Hunt NH, Grau GE (2005) Cerebral malaria: Which
parasite? Which model? Drug Discov Today 2:141–147.
27. Grau GE, et al. (1989) Monoclonal antibody against interferon gamma can prevent
experimental cerebral malaria and its associated overproduction of tumor necrosis
factor. Proc Natl Acad Sci USA 86:5572–5574.
28. Rudin W, et al. (1997) Resistance to cerebral malaria in tumor necrosis factor-alpha/
beta-deficient mice is associated with a reduction of intercellular adhesion molecule-1
up-regulation and T helper type 1 response. Am J Pathol 150:257–266.
29. Belnoue E, et al. (2002) On the pathogenic role of brain-sequestered alphabeta CD8+
T cells in experimental cerebral malaria. J Immunol 169:6369–6375.
30. Campanella GS, et al. (2008) Chemokine receptor CXCR3 and its ligands CXCL9 and
CXCL10 are required for the development of murine cerebral malaria. Proc Natl Acad
Sci USA 105:4814–4819.
31. Nie CQ, et al. (2009) IP-10-mediated T cell homing promotes cerebral inflammation
over splenic immunity to malaria infection. PLoS Pathog 5:e1000369.
32. Togbe D, et al. (2007) Murine cerebral malaria development is independent of toll-
like receptor signaling. Am J Pathol 170:1640–1648.
33. Gabrysová L, et al. (2009) Negative feedback control of the autoimmune response through
antigen-induced differentiation of IL-10-secreting Th1 cells. J Exp Med 206:1755–1767.
34. Nair MG, Cochrane DW, Allen JE (2003) Macrophages in chronic type 2 inflammation
have a novel phenotype characterized by the abundant expression of Ym1 and Fizz1
that can be partly replicated in vitro. Immunol Lett 85:173–180.
35. Ponomarev ED, Dittel BN (2005) Gamma delta T cells regulate the extent and duration
of inflammation in the central nervous system by a Fas ligand-dependent mechanism.
J Immunol 174:4678–4687.
Waisberg et al.PNAS
| January 18, 2011
| vol. 108
| no. 3