Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Lothar Wiese by yaofenjin


									The role of Erythropoietin Treatment in Cerebral Malaria – Studies
                  using an animal model in mice

                             Lothar Wiese
                                PhD Thesis

                      Faculty of Health Sciences
                      University of Copenhagen

                     Department of Clinical Microbiology
               Copenhagen University Hospital (Rigshospitalet)
                       Centre for Medical Parasitology
                         University of Copenhagen

1 Contents
1      Contents.......................................................................................................................................... 2
2      List of abbreviations ...................................................................................................................... 3
3      Original papers ............................................................................................................................... 4
4      Acknowledgements ....................................................................................................................... 5
5      Summary ......................................................................................................................................... 6
6      Resume ........................................................................................................................................... 7
7      Objectives ....................................................................................................................................... 8
8      Introduction: Cerebral malaria – a complication of Plasmodium falciparum malaria ............ 9
8.1    The burden of cerebral malaria ........................................................................................................ 9
8.2    The life cycle of Plasmodium falciparum.......................................................................................... 9
8.3    Acquired immunity to P. falciparum infection ................................................................................. 10
8.4    Difficulties defining the clinical syndrome “Cerebral malaria” in humans ...................................... 11
8.5    The mouse model for cerebral malaria .......................................................................................... 12
8.6    Pathogenesis of cerebral malaria .................................................................................................. 15
8.7    The sequestration of infected red blood cells ................................................................................ 16
8.8    Adjuvant treatment for cerebral malaria......................................................................................... 27
8.9    Erythropoietin ................................................................................................................................. 29
8.10   Metallothionein in cerebral pathology ............................................................................................ 33
9      Neuronal Apoptosis, Metallothionein Expression and Proinflammatory Responses
       During Cerebral Malaria in Mice ................................................................................................. 35
9.1    Abstract .......................................................................................................................................... 35
9.2    Introduction..................................................................................................................................... 36
9.3    Materials and methods................................................................................................................... 37
9.4    Results ........................................................................................................................................... 41
9.5    Discussion ...................................................................................................................................... 44
9.6    Acknowledgements ........................................................................................................................ 48
9.7    Figures ........................................................................................................................................... 49
10     Recombinant human erythropoietin increases survival in a dose- and time-dependent
       manner and reduces neuronal apoptosis in a murine model of cerebral malaria ................ 53
10.1   Abstract .......................................................................................................................................... 53
10.2   Introduction..................................................................................................................................... 53
10.3   Materials and methods................................................................................................................... 55
10.4   Results ........................................................................................................................................... 57
10.5   Discussion ...................................................................................................................................... 59
10.6   Conclusions.................................................................................................................................... 61
10.7   Acknowledgments .......................................................................................................................... 61
10.8   Figures ........................................................................................................................................... 63
11     Unpublished results..................................................................................................................... 70
11.1   Treatment with a newly constructed erythropoietin fragment could not increase survival in mice
       with cerebral malaria ...................................................................................................................... 70
11.2   Cytokines and VEGF in murine CM ............................................................................................... 71
11.3   Treatment with exogenous Metallothionein in mice with CM......................................................... 72
12     Perspectives ................................................................................................................................. 73
12.1   Erythropoietin fragments in the treatment of cerebral malaria....................................................... 73
12.2   The role of the blood brain barrier in cerebral malaria ................................................................... 73
12.3   Screening for changes in mRNA expression in erythropoietin-treated mice with CM ................... 74
12.4   Metallothionein in the treatment of cerebral malaria ...................................................................... 75
13     References .................................................................................................................................... 76

2 List of abbreviations

Ab         Antibody
BBB        Blood brain barrier
CD         Cluster differentiation
CM         Cerebral malaria
CNS        Central nervous system
DNA        Deoxyribonucleic acid
EC         Endothelial cells
ECM        Experimental cerebral malaria
Epo        Erythropoietin
GFAP       Glial fibrillary acidic protein
GPI        Glycosylphosphatidylinositol
HO-1       Haeme oxigenase-1
HZ         Haemozoin
ICAM-1     Intercellular adhesion molecule 1
ICP        Intra-cranial pressure
IL         Interleukin
INF        Interferon
iRBC       Infected red blood cell
kDa        Kilo Dalton
LT-α       Lymphotoxin alfa
MT         Metallothionein
mRNA       Messenger ribonucleic acid
NK-cells   Natural killer cells
PbA        Plasmodium berghei ANKA
PCR        Polymerase chain reaction
PfEMP-1    P. falciparum erythrocyte membrane protein-1
RBC        Red blood cell
rhEpo      Recombinant human erythropoietin
ROS        Reactive oxygen species
TGF-β      Transforming growth factor-β
TNF        Tumor necrosis factor
TUNEL      Terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-digoxigenin nick end labeling
VEGF       Vascular endothelial growth factor
WHO        World Health Organization
βcR        Common β receptor

3 Original papers

The PhD thesis is based on the following original papers included as separate chapters:

Neuronal Apoptosis, Metallothionein Expression and Proinflammatory Responses During
Cerebral Malaria in Mice
Experimental Neurology 2006 Jul;200(1):216-26

Recombinant human erythropoietin increases survival in a dose and time dependant manner
and reduces neuronal apoptosis in a murine model of cerebral malaria

4 Acknowledgements

Thanks to the many people who have contributed to the work on which this thesis is based.

Special thanks to my supervisors, Jørgen Kurtzhals, Milena Penkowa and Professor Ib
Bygbjerg, for creating the opportunity to work on this highly interesting topic and for supporting
literally every idea that came across in the process.

Especially, I would like to thank Jørgen for his support and ability to listen, not only to
professional topics.

Thanks to Professor Niels Høiby for his financial support all the way through the project.

Thanks to Casper Hempel, Nikolai Kirkby and Åsa Schiött for their work with the rt-PCR
analysis and their assistance with the Luminex machine.

Thanks to Grethe Gomme, Hanne Hadberg, Pernille S. Froh, Ha Nguyen, Marjan Yousefi and
many others for excellent technical assistance.

Thanks to all the people involved for supporting my work in the animal house, and in particular
to Kirsten Hansen and Jørgen Bøgh Stormm.

Thanks to the Department of Biostatistics, University of Copenhagen.

Thanks to Trine Staalsø for many fruitful discussions.

Thanks and all my love to Celine Haastrup and Olivia Wiese Niebuhr for making my days, both
the good ones and the very few not so good ones.

5 Summary

 The role of Erythropoietin Treatment in Cerebral Malaria – Studies in
                                an animal model in mice
Plasmodium falciparum malaria remains a massive burden of disease and death. At a
minimum, between 700,000 and 2.7 million persons die every year from malaria, over 75% of
them are African children. The two major complications, cerebral malaria (CM) and severe
anaemia, are responsible for more than 1 million deaths a year, mainly in children under 5 years
of age in sub-Saharan Africa. A significant number of children surviving CM are likely to be left
with neurological impairment.
Erythropoietin (Epo) was first identified as a haematopoietic growth factor made by the kidneys
and recombinant human Epo (rhEpo) is widely used as an anti-anaemic drug. The observation
that Epo and its receptor are expressed in the brain expanded the biological role of Epo beyond
Reports from animal studies had shown a reduction of neuronal damage and neurological
dysfunction in cerebral ischaemia, as well as neuroprotective effects in models of mechanical
trauma, excitotoxic injury and experimental autoimmune encephalitis. In a first trial in human
stroke patients, high-dose rhEpo treatment was shown to be safe and efficient.
Against this background, I studied the possible neuroprotective effects of Eythropoietin (Epo) in
CM in a widely used mouse model. First, I explored immunohistochemical changes in murine
CM. I was thereby able to describe neuronal apoptosis as a pathophysiological feature of
experimental CM and to propose it as a possible morphological correlate of neurological
impairment in human CM.
Subsequent studies in human recombinant Epo (rhEpo) revealed a time- and dose-dependent
protective effect of the treatment. Survival was increased by 50% in an otherwise lethal model,
and neuronal damage (i.e. the number of apoptotic neurons) was reduced significantly. A direct
effect on the brain could be demonstrated by measurements of gene expression (quantitative
PCR) in the brain.
Perspectives: In ongoing studies we try to elucidate the mechanisms of the positive treatment
effect of rhEpo by examining systemic serum-cytokine levels and gene expression profiles in
the brain (microarray technique). In immunohistochemical studies we want to supplement our
findings with studies of the treatment effect on the blood-brain barrier.
Based on the current findings and the very good safety profile of rhEpo as a treatment in
humans, a first safety and proof of concept trial in human CM patients is planned.

6 Resume

              Erythropoietin behandling ved Cerebral Malaria –

                             Studier i en dyremodel i mus
Plasmodium falciparum malaria er en betydelig årsag til sygdom og årsag til mellem 700.000 og
2,7 millioner malariadødsfald om året. Alene de to komplikationer cerebral malaria (CM) og
svær malaria anæmi tegner sig for mere end ca. 1 million dødsfald, mest afrikanske børn under
5 år. Neurologiske skader er hyppigt set hos børn efter behandling for CM.
Erythropoietin (Epo) blev først opdaget som en hæmatopoietisk vækstfaktor som bliver
produceret i nyrerne, og rekombinant human Epo (rhEpo) er i vidt omfang brugt som anti-
anæmisk lægemiddel ved blandt andet nyresvigt. Opdagelsen af ekspressionen af både Epo og
Epo-receptoren i hjernen har udvidet synet på Epo’s biologiske rolle: Behandling med
rekombinant Epo har vist sig at kunne reducere skader på neuroner og neurologisk dysfunktion
i forskellige dyremodeller for cerebral iskæmi; samt virke neuroprotektiv i modeller for mekanisk
trauma, excitotoksisk skade og eksperimentel autoimmun encephalitis. Derudover viser det
første studie i mennesker med iskæmisk hjerneskade at behandling med høje doser rhEpo er
både sikker og effektiv.
Med dette som baggrund har jeg undersøgt mulige effekter af Epo-behandling ved CM i en
dyremodel i mus. Jeg har først undersøgt immunhistokemiske forandringer i hjernen ved CM.
Her har jeg vist at apoptose af neuroner er del af de patofysiologiske forandringer ved
eksperimentel CM, og jeg kan derfor foreslå dem som et muligt morfologisk korrelat til de
neurologiske skader ved CM i mennesker.
Videregående forsøg har vist beskyttende effekt på hjernen som både er afhængig af dosis og
tidspunktet for behandlingen. Overlevelsen kunne forbedres til 50 % i en ellers dødelig model
for CM, og hjerneskaden - målt som antal af apoptotiske neuroner - var signifikant reduceret. En
virkning af behandlingen kunne også påvises ved hjælp af måling af gen-ekspressionen af bl.a.
cytokiner i hjernen (kvantitativ PCR).
Perspektiver: Mekanismerne bag den positive effekt af rhEpo behandlingen i CM skal nu
undersøges. Ved brug af microarray teknik skal indflydelsen af behandlingen på gen-
ekspression i hjernen sonderes, og derudover skal effekten af behandlingen på blod-
hjernebarrieren kortlægges ved hjælp af immunhistokemiske studier.
På baggrund af de foreliggende resultater og i lyset af den mangeårige og sikre brug af rhEpo
som lægemiddel i mennesker er det første ”proof-of-concept” studie i human CM ved af blive

7 Objectives

The overall aim of the PhD study was to evaluate a possible neuroprotective effect from
treatment with human recombinant erythropoietin in cerebral malaria in a mouse model for CM:

   1. Does treatment with rhEpo influence the clinical course of the disease?
   2. Does treatment with rhEpo influence the histopathological picture in the brain?

8 Introduction: Cerebral malaria – a complication of Plasmodium
falciparum malaria

8.1      The burden of cerebral malaria

Plasmodium falciparum malaria remains a massive burden of disease and death. At a
minimum, between 700,000 and 2.7 million persons die every year from malaria, over 75% of
them are African children (1). It is estimated that there are 515 (range 300–660) million clinical
episodes of clinical P. falciparum malaria worldwide (2), approximately 200 million of those in
sub-Saharan Africa (3). The two major complications, cerebral malaria (CM) and severe
anaemia, are responsible for more than 1 million deaths a year, mainly in children under 5 years
of age in sub-Saharan Africa (2).Very careful estimates demonstrate that CM alone affects
575,000 children under 5 years of age on the African continent annually. In the first place, all
these figures illustrate two points: The problem is big, and it is not even known exactly how big
it is.
Approximately every fifth child admitted to a hospital with the diagnosis CM in African countries
dies, despite of adequate treatment (4;5). The risk of CM is significantly higher in high-endemic
areas than in low-endemic areas (5), and a significant number of children surviving CM are
likely to be left with neurological impairment: Estimates between 2-10% are reported for
neurological complications lasting for more than 6 months after successful treatment (4;6-11).
It is widely agreed that P. falciparum is the only one of the four malaria parasites pathogenic for
humans that can cause CM; nevertheless solitary CM cases are reported in P. vivax infections

8.2      The life cycle of Plasmodium falciparum
The genus Plasmodium is characterised by erythrocytic and exo-erythrocytic schizogony in the
vertebrate host (mammals, birds, lizards) and sexual reproduction in blood-sucking mosquitoes.
Of the more than 200 Plasmodium species, four infect humans: P. falciparum, P. vivax, P. ovale
and P. malariae. Humans are infected with sporozoites, transmitted from infected, female
Anopheles mosquitoes. Sporozoites enter hepatocytes and multiply by schizogony generating
several thousands of merozoites. When released into the blood stream, merozoites enter the
cycle of erythrocytic schizogony by invading RBC. For P. falciparum, each intra-erythrocytic
cycle takes 48 hours and ends with the schizont stage and the release of 6-24 merozoites per
infected RBC (iRBC), which again infect RBC. Some merozoites develop into gametocytes able
to initiate sexual reproduction in the mosquito when taken up during a blood meal (Figure A).
The blood stage cycle causes rapid increase in parasitaemia and clinical disease. In non-

immune, untreated individuals the P. falciparum infection will eventually lead to complications
such as anaemia, respiratory distress, cerebral malaria and multiorgan failure.

                                                                         Figure A: The life cycle of
                                                                         Plasmodium falciparum.
                                                                         Sporozoites enter the liver and
                                                                         multiply. Released into the
                                                                         bloodstream as merozoites they
                                                                         infect red blood cells, thereby
                                                                         entering the asexual,
                                                                         erythrocytic cycle. As a result
                                                                         6-24 merozoites are released
                                                                         into the bloodstream ready to
                                                                         enter new red blood cells
                                                                         unless they develop into
                                                                         gametocytes. Gametocytes are
                                                                         taken up by a female mosquito
                                                                         during blood meal where they
                                                                         complete the life cycle of the
                                                                         parasite by sexual reproduction
                                                                         (from (14)).

8.3   Acquired immunity to P. falciparum infection
Infection by P. falciparum parasites can lead to substantial protective immunity. There is little
doubt that the reason why severe malaria is concentrated among children in areas of intense
and stable P. falciparum transmission is that adults have acquired protective immunity to the
disease. Adults in such areas may continue to experience sporadic parasitemic episodes, but
severe and life-threatening disease is rare (15). Acquired protection from malaria following
natural exposure to P. falciparum parasites is largely mediated by IgG. Clonally variant surface
antigens (VSA) that are present on the surface of iRBC are supposingly the main target of
protective IgG (15;16).
A distinct age distribution of CM and severe malaria anaemia (SMA) with in children being
hospitalised in a high-endemic area has been reported: Most cases of SMA appear within the

first two years of age, while the peak of CM occurs in the age group 3-4 years. It is not clear
whether this age distribution is caused by acquired immunity alone (5;17).

8.4    Difficulties defining the clinical syndrome “Cerebral malaria” in humans
The diagnosis CM in humans cannot rely on pathognomonic features of the disease and must
therefore be based on the exclusion of other pathological conditions. In the first place, this is a
pressing problem for the choice of the treatment and management in clinical settings, but it is
likewise important to clinical studies in which careful and uniform case definitions are crucial for
studies of malaria pathogenesis (18).

                                African children                               Adults
      Clinical features
      Coma                      Develops rapidly often after seizure           Develops gradually following drowsiness,
                                                                               disorientation, delirium and agitation over
                                                                               2-3 days or follows a generalised seizure
      Seizures                  Over 80% present with a history of             Occurs in up to 20%, mostly generalised
                                seizures and 60% have seizures during          tonic-clonic seizures; status epilepticus is
                                hospital admission; recurrent seizures are     rare
                                focal motor in >50%, generalised tonic-
                                clonic in 34%, partial with secondary
                                generalisation in 14% and subtle or
                                electrographic in 15%; status epilepticus is
      Other signs               Pallor, respiratory distress, dehydration,     Jaundice (40-70%), Kussmaul’s breathing,
                                rarely jaundice                                shock and spontaneous bleeding
      Neurological signs        Brainstem signs are present in >30% and         Patients typically have symmetrical
                                are associated with raised ICP, retinal        upper-motor neuron signs; brainstem
                                abnormalities are present in >60%, brain       signs and retinal abnormalities are less
                                swelling on CT is seen in 40%                  common
      Major complications and   Severe anaemia in 20-50%, of whom 30%          Multisystem and organ (circulatory,
      involvement of other      require blood transfusion, severe              hepatic, coagulation, renal and
      organs                    metabolic acidosis (presents as                pulmonary) dysfunction; pulmonary
                                respiratory distress), often associated with   oedema, renal failure, lactic acidosis,
                                hyperlactaemia; hyponatraemia (>50%),          haemoglobinuria have been reported;
                                hypoglycaemia (30%) and changes in             hypoglycaemia is present in only 8%
                                potassium; renal failure and pulmonary
                                oedema are rare
      Neurological squeals      Occurs in 11%, common sequelae are             Few, occurs in >5%; isolated cranial nerve
                                ataxia (2.5%), hemiparesis (4.4%),             palsies, mononeuritis multiplex,
                                quadriparesis (3.5%), hearing (1.9%),          polyneuropathy, extrapyramidal tremor
                                visual (2.3%) and speech (2.1%)                and other cerebellar signs
                                impairments, behavioural difficulties
                                (1.3%) end epilepsy

Table A: Clinical features and outcomes of cerebral malaria in African children and southeast Asian adults (adopted
from (19))

Briefly, CM can be described as a clinical syndrome with unrousable coma not attributable to
any other causes in a patient with falciparum malaria (20). Mainly systemic conditions (including
hypoglycaemia and hyponatriaemia) and bacterial meningitis, but also locally prevalent viral
encephalitides must be excluded. The picture is complicated by the fact that these conditions
frequently occur simultaneously with CM. The difficulties are further illustrated by the fact that

for example a patient in a postictal state after generalised convulsions and a present, but
symptom-free, parasitaemia – a feature frequently seen in malaria-endemic areas - would have
to be included.
Using the WHO clinical criteria, CM is defined as a potentially reversible, diffuse
encephalopathy causing a Glasgow coma score of 11/15 or less (alternatively: unable to
localise pain or a Blantyre Coma Score ≤2), often associated with fitting, in the absence of other
factors that could cause unconsciousness – such as coexistent hypoglycaemia or other CNS
infections - in a patient with asexual forms of P. falciparum on peripheral blood smears (21).
Nevertheless, the clinical manifestations of CM are varied, altered consciousness being the
single most characteristic feature, but any of a variety of other neurological signs as a possible
leading symptom (22).
The problem of differing case definitions is recognised also in published data of paediatric
cerebral malaria, where substantial differences have been recognised (22), making
comparisons between different studies sometimes difficult. In addition, considerable differences
as to the clinical features of CM between children and adults are seen (Table A). It is not
entirely clear whether these differences are associated with immunity or age (19).
In conclusion, it is important to recognise that the broad clinical definition based on exclusion
criteria can harbour several pathological mechanisms, meaning that not all clinically diagnosed
cases of CM in endemic areas will prove to be one; a point recently demonstrated in a
clinicopathological correlation of Malawian children (23). In particular, metabolic acidosis has
been recognised as a pathophysiological feature that cuts across the classical clinical
syndromes of CM and severe malaria anaemia with substantial influence on mortality (24;25).
This hampers interpretation and comparison of clinical studies. Also, it reflects the fact that the
pathophysiology of CM is not fully understood.

8.5     The mouse model for cerebral malaria
There are currently five major investigative approaches to better understanding the
mechanisms involved in the pathogenesis of severe malaria:
      1. clinical studies of patients in malaria-endemic regions;
      2. post-mortem studies of central nervous system tissue from patients with cerebral
         malaria (CM) and controls;
      3. in vitro model systems, for example studies of the adhesion of parasitised erythrocytes
         to receptor molecules;
      4. in vivo model systems (monkeys, mice and hamsters);
      5. genetic studies of susceptibility and resistance determinants in humans and mice.

Each of these approaches has its strengths and weaknesses, and together they have added to
the understanding of CM and have stimulated investigations in other fields (26).
This should be seen on the background that cultural constraints limit the number of studies of
the neuropathology of human CM, and that prospective studies of the relationship between
neurological dysfunction and immunological or other pathophysiological parameters in humans
are clearly unethical. Experimental animal models are therefore a useful tool to unravel the
processes that lead to coma and death (27).

Table B: Comparison of features of cerebral malaria in humans and mice (from (26))
Abbreviations: pRBC, parasitised red blood cell; CM, cerebral malaria; C3, complement fragment 3; TNF, tumour
necrosis factor; TNFR1, tumour necrosis factor receptor-1; IL-1b, interleukin-1b; ICAM-1, intercellular adhesion
molecule-1; VCAM-1, vascular cell adhesion molecule-1; DG, Dürck’s granuloma (a gliotic lesion containing
microglia and astrocytes); NR, not reported.
    Children from sub-Saharan Africa. cAdults from Thailand, Vietnam, India, Colombia or Caucasian travellers.
    Plasmodium berghei ANKA in CBA, A/J or C57 mice. eThe retina can be considered to be a projection of the brain
into the eye; in particular, there are many similarities in structure and embryological origin of the vasculature.

Animal models, and in particular mouse models, have been instrumental in our understanding
of certain aspects of CM (27-29), but experimental models cannot reproduce all the features of

human diseases, and this is also true for CM. However, CM is a syndrome in which the same
altered cell (the endothelial cell) plays a pivotal role both in human and experimental lesions,
and it is now widely agreed that the similarities between pathways of immune responses in mice
and humans justify the use of murine model of CM, even though cautious interpretation of the
results obtained with respect to existing differences is necessary (26;28).
The use of rodent malaria parasites in mice is currently the most widely used model for studying
the pathogenesis of CM. The availability of mice of defined genetic background, and the always
expanding number of transgenic and gene-deficient mice, has allowed a large variety of
experiments (28).
Many studies have used the comparison between resistant (e.g. BALB/c) and susceptible (e.g.
CBA, C57BL) strains of mice. DBA/2 mice develop a milder version of neurological symptoms,
but die later from severe anaemia (28;30). Susceptibility to CM is restricted to strains that are
genetically pre-determined to make a strong proinflammatory response (27). The most widely
used parasite has been P. berghei (ANKA or K173 strains) because of its ability to sequester
within the microcirculation. Susceptible mouse strains infected with P. berghei develop a
neurological syndrome characterised by paralysis, deviation of the head, ataxia, convulsions,
coma and a sudden drop of body temperature (27;28;31). 60-100% of mice infected
intraperitoneally with a relatively low number of iRBC of 104 to 106 die 8-14 days after infection
with relatively low parasitaemia of about 5-15% iRBC. Mice that do not develop CM die within
the third week of anaemia and hyperparasitaemia of more than 50%. Histopathological analysis
reveals vascular plugging, petechial haemorrhages and sequestration of leucocytes with
relatively few sequestered iRBC (28;32;33).
Table B summarises important features of murine CM compared to human CM, focusing on
similarities. Some important differences between CM in humans and mice are:
   •   The sequestering cell type in humans seems to be predominantly iRBC and to a lesser
       extend leukocytes – the situation in mice is vice versa (28).
   •   The PfEMP-1 family of proteins of P. falciparum is not expressed by rodent malaria
       parasites (34): As PfEMP-1 proteins are candidates for a major role in CM pathology
       including sequestration, this is a significant difference and could be the morphological
       basis for lesser iRBC sequestration in mice. On the other hand, it is also possible that
       other proteins expressed by both P. falciparum and rodent parasites play an important
       role in CM.
   •   There is no fever in the mouse models.
   •   Some reports suggest possible phenotypic variation in PbA (35) similar to that seen in P.
       falciparum (36). This would be important to the interpretation of results obtained in

       models using PbA as well as to the comparability of data derived from different
Results from murine models of CM – including the results presented in this work – should be
seen in the light of both similarities and differences with the human disease. The particular
murine model chosen for these experiments (PbA in C56BL/6 mice) has been the model of
choice in most laboratories, which has led to substantial amount of published data (37). It is for
that reason that this model has been favoured and not the more recently described model using
PbA in (Balb/c x C56BL/6)F1 mice, which benefits from a widespread sequestration of iRBC
within the cerebral microvasculature (27).

8.6   Pathogenesis of cerebral malaria
CM is principally seen as an acute encephalopathy with increased serum levels of
proinflammatory cytokines like tumor necrosis factor (TNF), interferon-γ (IFN-γ), and
lymphotoxin (LT) (26;38). A central feature of P. falciparum infection is sequestration of mature
forms of iRBCs and ring stages within the microvasculature of the major organs of the body,
predominantly the brain, heart, lungs and submucosa of the small intestine (27).
Post-mortem examination of the brain reveals haemorrhages, oedema and adherence of
infected red blood cells (iRBCs) to the cerebral microvasculature, mainly postcapillary venules.
The CNS microvasculature may eventually become obstructed by the iRBC adhering to the
endothelial lining, leading to ischemia and metabolic dysfunctioning, which contributes to the
CM pathophysiology (39).
Several hypotheses have been raised to explain the pathophysiology of CM (40-42). However,
none of them can, per se, explain the pathogenesis: It has still not been clarified how the
intraerythrocytic parasite, which sequesters in the cerebral microvasculature, but does not enter
the brain, induces the neurological syndrome CM.
The endothelial lining is likely to play a major role in this respect: Impairment of the BBB due to
endothelial apoptosis caused by direct interference of iRBC (43;44) and CD8+ T-cells (45-47)
with cerebral endothelial cells is believed to play a major role in CM pathology – at least
according to data from animal models.
Presently, severe malaria is recognised as a disorder that affects several tissues and organs,
even when the most marked manifestations may seem to involve a single organ such as the
brain (25). The different chapters under the main headline “Pathogenesis of cerebral malaria” is
a selection of observations on different levels of CM pathophysiology. The sequence of events
has not been settled yet. The brief overview given in the following mainly focuses on topics of
CM pathology with relation to the results of this work.

8.7   The sequestration of infected red blood cells
RBC infected with malaria parasites can acquire the ability to adhere to endothelial cells of post-
capillary venules, a mechanism termed sequestration. Sequestration happens as a result of
cytoadherence of iRBC to endothelial cells in various organs, including heart, lung, brain, liver,
kidney, subcutaneous tissue, and placenta via parasite-derived proteins inserted into the red
cell surface (48-50).
It is commonly accepted that sequestration helps the parasite avoid passage through the spleen
and thereby clearance. A consistent histological finding in paediatric cerebral malaria and in
adults is the presence of both, infected and uninfected RBC, within cerebral blood vessels
How adhesion progresses to pathology is a principal issue that remains unresolved. The fact
that infection with the non-sequestering parasite P. vivax does not lead to CM indicates an
important role for sequestration in the pathology of CM (52), as does the observation that the
number of sequestered parasites seems to increase with disease severity (53).
The effect could be mediated by several mechanisms causing damage to host endothelium and
organs, such as obstruction of blood flow leading to local hypoxia, inducing or blocking of signal
transduction mediated by host receptors and local release of cytokines and/or agents with
neurotransmitter, vasomotor or antigenic activity (25). There is evidence for different
mechanisms that can lead to reduced microvascular blood flow in P. falciparum infections:
Parasite binding to endothelial cells (sequestration), autoagglutination with other iRBC,
adherence to uninfected RBC (rosetting) and reduced deformability of iRBC (19).
Of the parasite-derived proteins that are expressed on the surface of iRBC, the multivariant P.
falciparum erythrocyte membrane protein-1 (PfEMP-1) is best described and is believed to bind
to endothelial cells via CD36, intercellular adhesion molecule-1 (ICAM-1), P-selectin, and
vascular adhesion molecule-1, E-selectin and chondroitin sulphate A (reviewed in (48;54)).
The idea of sequestration being the major cause of CM is challenged by both post-mortem
observations in humans and findings in mouse models: Sequestration seems to be a common
feature of P. falciparum infections both in patients with CM and in asymptomatic individuals. In
mice lacking the CD36 receptor, sequestration of iRBC was abolished without influencing the
clinical course of the disease, as CD36-deficient mice showed similar clinical signs of CM as did
wildtype mice (34). Furthermore, results from in vitro studies argue both for and against a major
role or sequestration in CM pathology. From some studies it seems as if contact of iRBC with
the endothelial cell layer is not necessary to induce alterations: Apoptosis of endothelial cells
could be induced by serum from patients with severe malaria in the presence of neutrophils only
(44). And both membrane-associated as well as soluble iRBC factors mediated the decrease in

electrical resistance as a measure of endothelial cell alteration in another model of the BBB
In other in vitro experiments, only direct contact with iRBCs induced apoptosis of endothelial
cells (43), and it has been proposed that the pathological cascade is initiated when EC are
activated, primarily due to parasite sequestration (45). Effector CD8+ T-cells could
subsequently – guided by locally up-regulated receptors and parasite antigens presented on the
EC surface (56;57) – bind and affect the BBB.

8.7.1   Proinflammatory cytokines in cerebral malaria Tumor necrosis factor and Lymphotoxin-α
A cytokine that has long been seen as playing a central part in the explanation of both
protection and malaria pathogenesis is the Tumor necrosis factor (TNF) (58). With regard to
pathogenesis, TNF levels have been found positively correlated with disease severity, in
particular in fatal cases of CM (59-62), and TNF m-RNA is up-regulated in the human post-
mortem brain of CM patients (63). African children with CM who develop neurological deficits
have higher plasma TNF levels than children without residual defects (64). Elevated levels of
TNF in African children, CM disease severity, and the occurrence of neurological sequelae have
been suggested to be regulated by polymorphisms of the TNF promoter gene (65).
The biological effects of TNF are mediated through interaction with cytokine receptors on target
cells. The soluble forms of TNFR1 and TNFR2 are elevated in plasma of adult and paediatric
malaria patients (66;67) as well as in mice (68). The results of several experiments in mice point
towards a central role of TNFR2 in the pathogenesis of CM: First, TNFR2 (but not TNFR1)
knockout mice were protected against murine CM (68). Second, the up-regulation of ICAM-1 on
brain microvascular endothelial cells has been shown to be mediated through TNFR2
engagement; TNFR2 knockout mice did not up-regulate ICAM-1 and did not show increased
leukocyte sequestration as seen in wildtype mice or in TNFR1 knockout mice (68;69).
Furthermore, studies in TNFR2-deficient chimaeric mice suggest that endothelial TNFR2 -
rather than the receptor on monocytes - is important in the pathogenesis of CM (70).
A clear role for TNF has, on the other hand, been shown in parasite killing (71). In physiological
concentrations, recombinant TNF synergised with INF-γ by increasing the production of toxic
radicals, and the achievement of an optimal level of TNF has been proposed to be central to the
resolution of a malaria infection (71).
However, recent studies in mice showed that Lymphotoxin-α (LT-α) - and not TNF - might be
the principal mediator of murine CM, as TNF-deficient mice are as susceptible to CM as are
wildtype mice, whereas LT-α-deficient mice were resistant to CM pathology (72). The picture

seems to be complicated by the fact that many of the existing immunological agents do not
discriminate between LT-α and TNF, meaning that studies that intended to focus only on TNF
might unintendedly have measured TNF and LT-α (71). The up-regulation of ICAM-1 through
TNFR2 has been proposed to be a probable mechanism of action of LT-α in murine CM, but so
far there is very limited evidence for the involvement of LT-α in human CM (26). Interferon-γ
Interferon-γ (INF-γ) seems to be essential to the development of CM, as INF-γ-deficient mice
and INF-γ receptor-deficient mice are protected against CM (73-75). In humans, serum INF-γ is
elevated in both South East Asian and African malaria patients (76;77). The detrimental effects
of INF-γ are believed to be due to its ability to activate macrophages, which, in turn, produce
TNF, IL-1 and IL-6, leading to an inflammatory cascade (71;78). Additionally, the presence of
INF-γ enhanced the receptor expression on the surface of mouse brain endothelial cells, which
was induced by TNF and/or LT-α (79). On the other hand, INF-γ may be protective by inducing
parasite killing (80), as INF-γ knockout mice in a non-CM model had significantly higher
parasitaemia and mortality (81). Interleukin-1
Interleukin-1 (IL-1) - especially the β-form - is believed to be the most important molecule
capable of modulating cerebral functions during systemic and localised inflammatory insults
(82). IL-1 synergises with TNF and can enhance its effectiveness (83). In the brain, IL-1β is
rapidly induced in response to neuronal cell death, is in itself neurotoxic in vitro and is thought
to contribute to ischemic cell death in the brain (63). IL-1β mRNA is not expressed in a normal
brain, but was found to be induced during CNS disease, also fatal human CM cases (63).
Plasma concentrations of IL-1 were seen to be related to the severity of malaria, especially to
fatal cases of CM (60).
Findings of elevated plasma concentrations of the IL-1 receptor antagonist (IL-1RA) in patients
with various diseases suggest that IL-1RA is part of the natural response to illness (84).
Elevated levels of circulating IL-1RA levels have been reported in paediatric CM, but
corresponding levels of IL-1RA in children with SMA do not suggest that the molecule plays a
central role in the specific pathophysiology of CM (85).

8.7.2     Anti-inflammatory cytokines in cerebral malaria    Interleukin-10
Interleukin-10 (IL-10) is pleiotrophic, but its principal routine function appears to be to limit and
ultimately terminate inflammatory responses (86). There are varying views of its role in CM. IL-
10 seems to have a host-protective role in murine CM (26): From experiments in mice it is
known that in vivo neutralisation of endogenous IL-10 in otherwise CM-resistant mice can
induce CM, and susceptible mice were significantly protected against CM when injected with
recombinant IL-10 (87). In line, high TNF:IL-10 ratios have been identified as a risk factor for
both SMA and CM compared to uncomplicated malaria in African children (88). In Asian adults
with P. falciparum malaria, IL-10 was significantly higher in patients with severe disease, and as
recombinant human IL-10 completely abolished in vitro TNF production in response to malarial
antigens, the authors propose a protective effect of IL-10 in terms of a negative feedback on
pro-inflammatory cytokines (76). But the biological role of IL-10 in P. falciparum infection is still
not clear: Anstey et al. measured high concentrations of IL-10 and decreased NO production in
African children with CM compared to cases with uncomplicated malaria and uninfected
children. Quoting IL-10’s ability to reduce NO production in macrophages (89), the authors
propose that IL-10 is increasing disease severity in CM (90). In a separate study which was
carried through in Ghana and which compared children with SMA and CM with cases of
uncomplicated malaria, IL-10 levels were significantly lower in SMA, while IL-10 concentrations
were similar in children with and without cerebral malaria, excluding patients with severe
anaemia. Moreover, the TNF/IL-10 ratio did not differ significantly between patients with
cerebral and uncomplicated malaria (61).

8.7.3     Transforming growth factor-β
Transforming growth factor-β (TGF-β) is thought to play a role in modulating events at the BBB
in CM (91). TGF-β was up-regulated in the post-mortem brain of CM patients and showed
intravascular and perivascular distribution in African children (92). TGF-β might have a dual role
in CM, as in vitro studies indicate that platelet-derived TGF-β1 may synergise with TNF to
induce the microvascular lesion of CM (93).

8.7.4     Reactive oxygen species
Reactive oxygen species (ROS) are known to be secreted from actived phagocytes (94). In
pathologic situations, ROS can be generated in excess of the antioxidant capacity of a cell,
resulting in severe damage to cellular constituents, including proteins, DNA and lipids (95).
They can trigger apoptosis and necrosis of EC in vitro (96), have been shown to damage
cerebral EC in vivo (97), and were capable of parasite killing in vitro (98). Thus, ROS have the

potential to cause oxidative damage to the endothelium and breakdown of the blood-brain
ROS are produced during malaria infections in mice and appear also to be produced in human
CM, but their role in the pathogenesis is unclear (6). The fact that antioxidants could prevent
CM in mice (99) and can reduce endothelial cell apoptosis induced by serum from malaria
patients in vitro (44) indicates a possible role for ROS in CM.

8.7.5      Nitric oxide
Endogenous nitric oxide (NO) can be produced by three isoforms of NO synthase (NOS) from
L-arginine in the brain, is a major regulator of vascular haemodynamics, and is the primary
messenger molecule mediating blood vessel relaxation (100). NO could be induced in mice in
response to both, exogenous TNF, LT-α and IL-1, all of which are increased in malaria (101).
NO generated by human monocytes could kill P. falciparum in vitro (102), and it is proposed
that it could act in the same way in human malaria (103).
Regarding the role of NO and the different nitric oxide synthase (NOS) isoforms in focal
cerebral ischemia, neural NOS (nNOS) and inducible NOS (iNOS) are perceived to play key
roles in neurodegeneration, while eNOS plays a prominent role in maintaining cerebral blood
flow and preventing neuronal injury (104).
A study in African children found cerebrospinal fluid concentrations of NO to be significantly
higher in children who died from CM than in survivors and added to the notion that locally
produced NO may be detrimental in CM (105). Moreover, elevated levels of iNOS protein are
detected in the cerebral-blood-vessel walls of patients who have succumbed to CM (103).
Knowing that INF-γ, in combination with TNF and IL-1, can induce NO production in murine
brain endothelial cells (106), it was long believed that NO was a contributor to CM pathology
(103). On the contrary, there are higher systemic concentrations of NO in uncomplicated
malaria than in CM (90;107), and in the mouse model treatment with exogenous, NO did not
affect parasitaemia, but provided marked protection against CM (108). In line with these results,
NO production had been found to be inversely correlated with malaria severity, suggesting that
NO has a protective rather than a pathological effect in African children with malaria (90).
Low levels of NO in certain areas of the brain could be crucial for localised pathology in CM:
Hypoxia causes a reduction of endothelial NOS (eNOS) activity, and this effect is enhanced by
high levels of IL-1 and TNF (109). This effect could further increase hypoxia due to sequestered
iRBC by further impairing the cerebral blood flow locally. NO could thereby act as the common
effector influenced by both high levels of pro-inflammatory cytokines and localised hypoxia due
to sequestration of iRBC.

8.7.6   The cellular immune system in cerebral malaria
The observation that PbA infection in mice deficient for both B-cells and T-cells does not induce
murine CM indicates that CM is a lymphocyte-mediated disease (74). It has later been
established that B-cells are not a requisite for CM pathology in mice, as B-cell-deficient mice
develop CM to the same degree as do wildtype mice, indicating that neither B-cells, their
products, nor antigen presentation are key elements in the development of CM (74). In the
following, αβT-cells (i.e. CD4+ and CD8+ T-cells), but not γδT-cells, were found to be important
to the genesis of CM (110). The activation of CD4+ and CD8+ T-cells have been identified as
key mechanisms that contribute to the onset of CM in mice: Depletion with specific anti-CD4 or
anti CD8-antybodies prevents murine CM in otherwise susceptible mice (45;74;74;111).
The mechanisms by which CD4+ and CD8+ T-cells synergise to mediate murine CM are
unclear. It is possible that CD4+ T-cells (apparently of the Th1 phenotype) provide help to CD8+
T cells and/or release cytokines that help potentiate or augment CM induction. Similarly, the
role of CD8+ T-cells, which sequester in the brain at the time when neurological symptoms
appear (45), may be to exert direct cytotoxicity- or cytokine-mediated effects. Direct damage of
endothelial cells is seen in the human post-mortem brain (50), and subsequent impairment of
the BBB, oedema, and haemorrhage have been proposed to be direct consequences of CD8+
T-cell activity in CM pathology. Data from mice suggest that CD8+ T-cells induce apoptosis of
endothelial cells via a perforin-dependent process (47). Moreover, CD8+ or CD4+ T-cells could
be an additional source of IFN-y for macrophage recruitment, or they might be a direct source
of TNF at the site of injury (74).
A very recent study in a mouse model identified a subset of dendritic cells – conventional
dendritic cells, but not plasmacytoid dendritic cells – to be required for the induction of malaria-
specific CD4+ T-cells and subsequent clinical experimental malaria (112).
Neutrophils, recruited at the inflammation site, are a source of cytokines (IL-12, IL-12 p40, IL-
18, IFN-γ, TNF) and chemokines that participate in pathogenesis (113). Early neutrophil
depletion prevented the development of murine CM, down-regulated the expression of Th1
cytokines in the brain, and decreased the sequestration of monocytes and BBB impairment
(114-116). In vitro neutrophils were able to induce apoptosis of endothelial cells when incubated
with patient serum from CM cases.
The natural killer (NK) cell function seems to be depressed in CM patients compared to
uncomplicated malaria (117), and depletion of NK1.1+ cells did not prevent murine CM (74).
Thrombocytopenia is a common feature in P. falciparum infections (118). More recently a more
precise role for platelets in CM pathology has been established: In an in vitro model of
endothelial lesion, platelets could act as bridges between iRBC and ECs, allowing the binding of

iRBC to endothelium and potentiating the cytotoxicity of iRBC for brain EC by inducing an
alteration of the integrity of their monolayer and increasing apoptosis (119;120).

8.7.7   Malaria toxin Glycosylphosphatidylinositol

The idea that parasite-derived toxins are involved in malaria pathology dates more than 100
years back (reviewed in (121)). The current understanding includes a variety of bioactive
molecules that either up- or down-regulate pathogenic processes, largely through their effects
on the innate immune system (122). Many studies implicate glycosylphosphatidylinositol (GPI)
of P. falciparum as a malaria toxin (122). Purified GPI induces the expression genes that
encode pro-inflammatory cytokines — such as TNF, IL-1 and IL-12 (123). More importantly,
immunising against GPI in mice reverses the susceptibility to severe malaria, including CM in
PbA infections, establishing it as a crucial component of pathogenesis in this model (124). Haemozoin
During its intra-erythrocytic stage, the malaria parasite feeds on host haemoglobin. While this
process delivers essential amino acids, it also liberates cytotoxic haem. To protect itself from
oxidative damage, the malaria parasite has developed a detoxification process that converts
haem into an insoluble crystal called haemozoin (HZ) to be stored in its food vacuole. HZ is
shed into the blood stream during schizont rupture and subsequently ingested by macrophages.
Human and murine macrophages produce increased amounts of TNF, IL-1, IL-6, and
macrophage inflammatory proteins 1α and 1β during malaria infection (125). HZ has been
shown to have the potential to amplify the induction of pro-inflammatory cytokines and fever
caused by malarial DNA. This immunogenic activity of HZ is proposed to be dependent on its
ability to present malarial DNA to Toll-like receptor-9 (126;127). Final evidence as to whether or
not HZ does in fact contribute substantially to fever and disease in P. falciparum-infected
humans is yet lacking (126).

8.7.8   Possible involvement of antibodies in cerebral malaria pathology
Plasma samples from Gabonese children with CM showed a correlation between self-antibody
responses to a human brain protein with the manifestation of CM in Gabonese children.
Antibody response to brain antigens induced by P. falciparum infection may thereby be
associated with pathogenic mechanisms in patients developing CM (128). Similar results are

presented in a study from Senegal where self-reacting antibodies against ECs were measured
in P. falciparum-infected patients (129).
Total IgE or IgE anti-malaria antibody levels are significantly higher in patients with CM than in
those with uncomplicated malaria (130), and it has been suggested that exposure of monocytes
to IgE containing immune complexes or aggregates activates them to produce TNF. It has
further been suggested that such mechanisms may also be of importance in malaria
pathogenesis (130).

8.7.9   Apoptosis and cerebral malaria
In the adult human body, several hundred thousand cells are produced every second by
mitosis, and a similar number die by programmed cell death - apoptosis (131). The proposed
purpose of apoptosis is the safe and non-inflammatory removal of cell corpses by phagocytes
(132). On the contrary, during necrosis, the cellular contents are released uncontrolled into the
cell environment, which results in damage to surrounding cells and a strong inflammatory
response in the corresponding tissue (132).
The survival of cells is dependent on the homeostasis of pro- and antiapoptotic signals.
Apoptosis can be induced in response to various signals from inside and outside the cell, e.g.
by ligation of so-called death receptors or by cellular stress triggered by oncogenes, irradiation
or drugs. The cell membrane receptor Fas (CD95) is one of those death receptors, mediating its
effect by initiating a caspase cascade including the initiator caspase-8 and the effector
caspases-3, -6, and -7, which in turn cleave a number of protein substrates. Antiapoptotic
activity can be up-regulated in response to survival signals such as those coming from growth
factor receptors, e.g. by activation of the transcription factor NF-kB, therefore providing a means
to suppress apoptosis signalling (reviewed in (132;133)).
Ischaemic stroke and cancer are two examples of diseases associated with increased or
decreased apoptosis, respectively (133). As for human ischaemic stroke, apoptosis is
responsible for a substantial loss of central neurons especially in areas with less severe
ischaemic damage (134).
Apoptosis of host cells in CM is a target of research both with respect to brain parenchyma cells
and cerebral endothelial cells. The former as a possible morphological correlate of the
neurological impairment, and the latter as a step in the breakdown of BBB. Very recent results
from mouse models increase the body of evidence for apoptosis of neurons as well as
endothelial cells in CM (135;136).
This is discussed in more detail in the following.

8.7.10 Endothelial cells and the blood brain barrier in cerebral malaria
The BBB is a highly specialised structural and functional interface between the intravascular
lumen and the CNS. Three cellular elements of the brain microvasculature compose the BBB –
endothelial cells, astrocyte end-feet and pericytes. Tight junctions, present between the ECs in
the brain, form a diffusion barrier, which selectively excludes most blood-borne substances from
entering the brain (reviewed in (137)). At the capillary level, it consists of cerebral microvascular
endothelium, the basement membrane, and astrocyte foot processes (Figure B) and has a
decreased permeability compared to vascular beds in other organs (138).
In addition, postcapillary venules have a perivascular space (Virchow-Robin space) located
between the two basement membranes of the endothelial cells on the one side and the
astrocyte end-feet of the other side (138;139). The differences in the biochemical composition
of the vascular and the astrocyte basement membranes might explain how leukocytes can,
under normal conditions, pass the former but not the latter (139).

Figure B: Schematic cross-sectional representation of a cerebral capillary. The circumference of the capillary lumen
is completely surrounded by a single endothelial cell (EC), the apposing membranes of which are connected by tight
junctions (TJ). Pericytes (PC) are attached to the abluminal surface of the EC, and these two cell types are
surrounded by the basal lamina (BL), which is contiguous with the plasma membranes of astrocyte (AC) end-feet
and EC (from (138)).

The BBB represents a key interface between the intraerythrocytic stages of P. falciparum that
never enter the brain parenchyma and the CNS (91). Impairment of the BBB is regularly seen in
African children with CM (140), while data from Asian adult CM patients show conflicting results
(141-144). Intracranial pressure is increased in most African children with CM (145;146), most

probably as a result of an impaired BBB. In mouse models of CM, cerebral oedema and
breakdown of the BBB are regular features as seen in several studies using a variety of
techniques (including movement of Evan’s blue dye, immunohistochemistry, magnetic
resonance imaging) (147-150). The onset of neurological symptoms in the mouse model could
be linked to the breakdown of the BBB and the resulting oedema including compression of
cerebral arteries as seen using magnetic resonance (151).
Concerning the morphological background of BBB impairment, there is evidence of alterations
of cell junction proteins of the BBB (occludin, vinculin, ZO-1) in CM as well as activation and
damage of endothelial cells, including apoptosis. Damage of endothelial cells in human CM is
indicated by high plasma levels of thrombomodulin in patients with CM. Thrombomodulin is a
nonsecretable membrane protein of resting endothelial cells, which is normally not found in the
circulation (44).
Both sequestration of iRBCs and direct effects of cytotoxic CD8+ T-cells are possible causal
mechanisms related to the impairment of the BBB in CM (148). It is proposed that brain ECs
can phagocytose iRBCs and present parasite-derived proteins to the cell surface, enhancing
the binding of cytotoxic T-cells and thereby enhancing BBB destruction (148). Activated ECs
show an enhanced vesiculation, i.e. the release of circulating microparticles which are proposed
to have the ability to decrease plasma-clotting time and to induce the release of inflammatory
cytokines (119). In a study of African children, the number of circulating microparticles was
found to be markedly increased in severe malaria complicated with coma only as measured on
the day of hospital admission (152).

8.7.11 A proposed sequence of events leading to BBB impairment
Focusing on the BBB, Coltel et al. have recently proposed a sequence of events explaining
some aspects of the pathogenesis of CM (Figure C) (113). Their theory points to the interplay of
intravascular T-cells and the activation of microglia and astrocytes and their effect on brain
endothelial cells.
Recently Renia et al. have expanded this model and propose a role for the sequestration of
parasites (153). They suggested that brain EC, in response to IFN-γ and/or TNF, can
phagocytose cytoadherent iRBC or parasite-derived materials, process them, and present them
to T-cells. T-cells and in particular CD8+ T-cells binding to the endothelium and via production
and release of perforin induce apoptosis of endothelial cells via a perforin-dependent process,
locally contributing to the impairment of the BBB. This possible mechanism builds on data that
showed increased immunoreactivity for perforin and elevated numbers of CD8+ T-cells during
CM in mice (45;47) and protection against CM in mice unable to produce perforin (47). A role

for platelets might be to enhance the proposed lytic actions of certain cytokines on endothelial
cells and favour the interactions between iRBC and endothelial cells (120).

Figure C. A sequence of events at the BBB in CM as proposed by Coltel et al.
Malarial antigens are processed and presented by DC to CD4+ T lymphocytes (1), leading to their expansion and
activation (2). These activated CD4+ T cells produce large quantities of IL-2 (3), which drive the expansion and
activation of CD8+ T lymphocytes(4). The activated CD8+ T cells damage the endothelial cells (EC) in the small
vessels of the brain, thereby permeabilising the blood brain barrier (BBB) (5). CD4+ and CD8+ T cells secrete
copious amounts of IFNγ (6), which can penetrate the permeabilised blood brain barrier (7) and activate microglia
in the CNS to up-regulate their surface expression of FasL(8) and induce astrocytes to up-regulate cell surface Fas
(9). FasL on microglia interacts with Fas on astrocytes and this interaction damages the cells (10). Damage to
astrocytes further compromises the integrity of the BBB and affects the inter-neuronal milieu, disrupting CNS
function (11). Astrocytes and microglial cells produce mediators that can modulate the phenotype and functions of
brain EC and, potentially, of leucocytes within the cerebral microvasculature, given the enhanced permeability of
the BBB (12).

The weakness of models favouring a strong role for T-cells in CM pathology is the limited data
available describing the involvement of T-cells in the pathophysiology of human CM. Malaria
infection in humans has been associated with an increase in the frequency of circulating CD8+
T-cells (117;154), and CD4+ and CD8+ T-cells were involved in the immune response after
experimental P. falciparum infection in healthy volunteers (155). Nevertheless, evidence still
needs to be established.

8.7.12 What is the morphological correlate of the neurological sequels?
The pathological process by which P. falciparum malaria induces severe, but usually reversible,
neurological complications has not been elucidated (156), but the loss of neurons or at least a
substantial loss of function is likely. Disruption in axonal transport has been proposed as a final
common pathway leading to neurological dysfunction in cerebral malaria as seen in a post-
mortem study of Vietnamese CM patients (156).
Apoptosis has been proposed, and recent works are adding to the evidence of neuronal
apoptosis in murine CM: Neurons positive for cleaved caspase-3 showing ultrastructural
changes suggesting apoptotic cell degeneration were seen in one study (135) and Terminal
deoxynucleotidyl transferase (TdT)-mediated deoxyuridine triphosphate (dUTP)-digoxigenin
nick end labelling positive (TUNEL+) neurons have recently been seen (136), while up-
regulation of proteins involved in the apoptotic signalling cascade is seen in the brain of PbA-
infected mice (157). The picture in human CM is not conclusive: A single study detected DNA
fragmentation typical for apoptotic cells in central neurons of a patient with CM (158), a finding
that has not been confirmed by others so far.
And a post-mortem study of Vietnamese CM patients found caspase-3-reactivity – again as a
measure for apoptosis – in only few central neurons with similar findings made in non-malaria
controls (159).

8.8   Adjuvant treatment for cerebral malaria
Eliminating the parasite is an essential step in the treatment of CM. Parenteral treatment with
quinine, artesunate or artemether are the first choices (21). In addition, severe manifestations
and complications need to be treated. Management of hypovolaemia with intravenous
cristalloids, of hyperpyrexia, convulsions, hypoglycaemia, severe anaemia, acute pulmonary
oedema, spontaneous bleeding, metabolic acidosis, circulatory shock, and acute renal failure,
metabolic acidosis and aspiration pneumonia is important (21).
No effective treatment to protect the brain against damage is otherwise available; despite of the
many efforts that have been made. The list of treatments that did not prove to be beneficial and
which are not recommendable includes: Corticosteroids and other anti-inflammatory agents,
mannitol, adrenaline, heparin, prostecyclin, pentoxyphylline, hyperbaric oxygen, cyclosporine A,
hyperimmune serum, desferoxamine, and anti-TNF antibodies (6;21).

8.8.1    New approaches for adjuvant treatment in cerebral malaria
Following the hypothesis that the BBB and the endothelium of the small blood vessels in the
brain are crucial for CM pathology, a number of recent reports focus on the potential of
interventions targeted at the BBB: Pharmacological induction of Haeme oxigenase-1
Haeme oxigenase-1 (HO-1) has been identified as a central enzyme in the protection against
CM: Mice resistant to CM express significantly more HO-1 in the brain during infection than
susceptible mouse strains, but can be rendered susceptible by inhibition of HO-1. Furthermore,
HO-1 induction in mice susceptible for CM reduced the incidence of CM significantly. In these
mice parasitaemia was not affected, BBB disruption, brain microvasculature congestion and
neuroinflammation, including CD8+ T-cell brain sequestration, was prevented (160). The
authors describe the generation of free haeme during malaria infection, a molecule cytotoxic to
the host when released into the circulation (160;161). Interestingly, pharmacological induction of
HO-1 as well as exposure to carbon monoxide (CO), both reducing free heme, were protective
(160). Exogenous nitric oxide
The role of NO in the pathophysiology of CM is not quite clear (see above), but a recent study in
the mouse model adds significant evidence for a protective role. Treatment with the NO donor
dipropylenetriamine NONOate or NO gas resulted in a significant reduction of CM pathology
(108). Without effecting parasitaemia, the treatments were able to reduce inflammation in the
brain and markedly reduced vascular leak and petechial hemorrhage into the brain (108). A
limiting factor for this treatment is the strong, lowering effect on the systemic blood pressure. Antioxidants reducing endothelial cell apoptosis
In an in vitro model of the BBB, apoptosis of endothelial cells could be induced by serum from
patients with severe malaria in the presence of neutrophils. The antioxidants ascorbic acid and
tocopherol and the protease inhibitor ulinastin were able to reduce apoptosis (44).
N-acetylcysteine – another antioxidant – was safe when given to patients with severe malaria
and was able to reduce serum TNF levels and improve deformability of RBC’s (162;163), but
there are concerns about possible drug interactions with artesunate (164). Large clinical trials
have been announced, but no reports have been published so far.
An older report described increased survival in a mouse model for CM after both oral
application and injection of butylated hydroxyanisole, a substance known to have antioxidant
properties (99), but a follow-up could not be found in the literature.

                                                 28 Anti CD8+ antibodies
CD8+ T-cells are believed to play a crucial role in CM pathology (see above). In the mouse
model, depletion of CD8+ T-cells with specific antibodies prevented CM pathology even when
given late in the course of the disease (45). CD8+ T-cells are believed to act by binding to
endothelial cells in the brain, and the number of CD8+ T-cells in the brain of antibody-treated
mice was lower than in untreated mice which developed CM (45), suggesting interference with
T-cell binding. Specific antibodies targeting molecules that mediate the binding of CD8+ T-cells
to endothelial cells are used for the treatment of multiple sclerosis (165). Further studies are
needed to investigate this approach. Phosphatase inhibitors to decrease sequestration
This approach is based on the finding that CD36 mediated binding of iRBC to endothelial cells -
one of the major mechanisms responsible for sequestration - is stabilised through
dephosphorylation of the CD36 molecule. Moreover, it was shown in vitro that adhesion of iRBC
to endothelial cells enhanced their adhesion through a alkaline phosphatase-dependent
mechanism (166). Inhibition of phosphatase activity by the specific phosphatase inhibitor
levamisole could reduce iRBC adhesion significantly in vitro (167). In a subsequent proof-of-
concept trial in patients with uncomplicated malaria in Thailand, a single dose of levamisole
resulted in an increase of late-stage iRBC in the peripheral blood. The authors consider
reduced sequestration caused by the levamisole treatment rather than decreased clearance of
late-stage parasites in the spleen to be the mechanism responsible (167). Even though the
study was small, and a satisfactory explanation for the also observed prolonged parasite
clearing time was not given, the findings make levamisol a possible candidate for adjunctive
therapy in cerebral malaria.

8.9   Erythropoietin
Erythropoietin (Epo) was first identified as an haematopoietic growth factor made by the
kidneys, and recombinant human Epo (rhEpo) is widely used as an anti-anaemic drug (168).
Endogenous Epo is hypoxia-inducible and was found to maintain optimal tissue oxygenation by
regulating the number of erythrocytes through negative feedback control between the kidney
and the bone marrow (169). For this, Epo levels promote survival of erythroid precursor cells of
the bone marrow - that would otherwise undergo apoptosis - through binding to a specific Epo-
receptor (EpoR) and further downstream the induction of members of the Bcl family of
antiapoptotic proteins (170). Epo and its receptor (EpoR) are members of the type 1 cytokine

superfamily, which as a class is multifunctional. Epo is closely related to cytokines that
modulate growth and inflammation (168). Plasma levels of endogenous Epo may rise up to
10000 U/L in severely anaemic persons with normal kidney function (168), while the normal
concentrations of circulating Epo in healthy non-hypoxic humans are 6-32 U/L with normal Epo
concentrations in the CSF of about 1 U/L (168).

8.9.1   Epo, Epo receptor and the brain
The observation that Epo and its receptor are expressed in the brain expanded the biological
role of Epo beyond haematopoiesis. It is now known that practically all brain cells are capable of
producing and releasing Epo and its receptor (reviewed in (171;172)). Astrocytes have been
shown to be the main Epo-producing cell type in the CNS (173), while neurons produce Epo to
a lesser degree (174). Epo produced in the brain is smaller in size compared to renal Epo (33
kDa vs. 35 kDa), due to post-translational modification, but is functionally equivalent (175). The
EpoR in the brain is expressed on neurons and astrocytes as well as endothelial cells and
microglia/macrophages (176;177).
Both Epo and EpoR mRNA expression is induced upon hypoxia in rodent and human brains
(178). Epo gene expression is regulated by hypoxia-inducible-factor-1 (HIF-1) (179) in the
following way: The HIF-1α subunit is inactivated during normoxic conditions but is very rapidly
stabilised under hypoxic conditions, leading to increased expression of genes such as Epo and
- interesting in the context of CM - VEGF as well (180). It was earlier assumed that systemically
administered Epo would not be able to pass the BBB in order to enter the brain. However, in
experiments with biotinylated rhuEpo in healty rats it was observed that Epo can pass the BBB
and reach the parenchyma of the CNS (181).

8.9.2   Mechanisms of Epo action in the CNS
The physiological effect of Epo on erythoid precursor cells in the bone marrow has been shown
to be mediated by binding to the membrane bound EpoR, a homodimeric class 1 cytokine
receptor (182). Intracellular-signalling mechanisms involve the activation of janus tyrosine
kinase 2 (JAK-2), ras-mitogen-activated-kinase (rasMAPK) and signal transducers and
activators of transcription 5 (Stat-5). These pathways can alter the expression of target genes
such as the antiapoptotic factor bcl-2 and bcl-XL and the antioxidant enzyme superoxide
dismutase (178).
It has earlier been assumed that the protective effects of Epo in the CNS are at least partially
mediated through anti-apoptotic molecules using the same pathways as in the bone marrow
(183;184). However, the receptor complex mediating the neuroprotective effects of Epo differs
from the hematopoietic receptor with respect to apparent affinity for Epo (which is in fact lower

for neuroprotective effects), apparent molecular weight, and associated proteins (182;185). A
region of Epo not within the binding domains involved in the haematopoietic functions has been
associated with neuroprotective effects (186), and recent data suggest that tissue protection
signals through the interaction of Epo with an EpoR–βcR heteromer comlex (187). This is in line
with the observation that it is possible to construct erythropoietin derivates such as
carbamylated erythropoietin which do not bind to the EpoR and does not stimulate
erythropoiesis, yet it prevents tissue injury in a wide variety of models in vivo and in vitro (187).
Increasing evidence shows that rhEpo can prevent hypoxia-induced neuronal cell death in vitro
as well as in animal models (184;188;189) and induces up-regulation of apopotosis-inhibitor
genes in brain cells both in vitro and in vivo (190;191).
The cytoprotective effects are not restricted to neurons: Epo can also protect endothelial cells
(192) and glial cells (193) from ischemic damage. Potential cytoprotective mechanisms of Epo
besides anti-apoptosis include reduced astrocyte activation (194), direct neurotrophic effects
(186), antioxidation (195), and angiogenesis (196). In addition, Epo stimulated the proliferation
and differentiation of neuronal stem cells (197;198). The anti-inflammatory effect may be partly
indirect: Epo did not affect the production of proinflammatory cytokines in astrocyte culture or in
peripheral blood mononuclear cells (194;199) and did not affect serum-TNF levels in vivo
stimulated by injection of bacterial lipopolysacharides (194).
With regard to the functions of Epo on endothelial cells, it is noteworthy that endothelial cells
and haematopoietic cells are possibly derived from the same mesenchymal precursor, the so-
called haemangioblast (200). This could explain why endothelial cells carry the EpoR.
Several studies point towards a role for vascular endothelial growth factor (VEGF), even though
the picture given by the published results is incomplete. It has also been proposed that Epo can
act on endothelial cells through activation of the VEGF/VEGF receptor system in the hypoxic
brain (201). Others show an up-regulation of VEGF, induced by rhEpo which, at the same time,
is counteracted by a reduced expression of its VEGF receptor (Flt-1) (202). Incubation of
endothelial cells with rhEpo resulted in VEGF release, while in an in vitro model of the BBB,
rhEpo was able to reduce VEGF-induced permeability (203). In the human post-mortem brain of
CM patients, VEGF is increased as well as the VEGF receptor when compared to controls
(204). No results demonstrating the effect of rhEpo in CM have been published so far.

8.9.3   Neuroprotective effects of recombinant human erythropoietin in vivo

The first clear evidence for the neuroprotective effects of Epo in vivo was provided in 1998:
Infusion of rh-Epo prevented ischaemia-induced learning disability and rescued hippocampal
neurons from lethal ischaemic damage in a dose-dependent manner, while, also in a model of

experimental cerebral ischaemia, infusion of EpoR into the lateral ventricles resulted in
increased neuronal degeneration and impaired learning ability. These results indicate that brain-
derived Epo is crucial for neuronal survival (189). There is now substantial evidence that
treatment with rhEpo can reduce neuronal damage and neurological dysfunction in rodent
models of stroke (189;205;206).
Subsequent studies in various animal models reported the ability of Epo to reduce neuronal cell
death in subarachnoid haemorrhage (207), neonatal hypoxic brain injury (208), kainite-induced
seizures (181), spinal cord injury (209), and also revealed activity against cerebral inflammation
in a model for multiple sclerosis (181).
As in the case for erythroid precursor cells and neurones, Epo also seems to be a survival
factor for endothelial cells by preventing cell injury and apoptosis (192). In addition, a protective
effect on heart muscle cells in cardiac ischaemia has been shown, which is thought also to be
based on an antiapoptotic effect of Epo (210;211).
Promising results have come from a proof-of-concept trial in patients with acute ischemic
stroke: High-dose rhEpo treatment (33,000IU rhEpo given on three consecutive days) within 8
hours after the onset of symptoms was well tolerated and associated with an improvement in
clinical outcome after one month (212). The treatment also resulted in a strong trend toward
reduction of infarct size as assessed by magnetic resonance imaging in the rhEpo-treated
patients compared to untreated controls (212).More comprehensive studies are on their way,
examining this possible improvement in functional outcome in rhEpo-treated patients after
stroke (213).

8.9.4   Different recombinant human Epo molecules
Several brands of rh-Epo and Epo-like molecules are presently commercially available. These
include the established epoetins alfa and beta, which are produced in Chinese hamster ovary
cells, epoetin omega, which is produced in baby hamster kidney cells, and darbepoetin alfa,
which is manufactured in Chinese hamster ovary cells, but has an amino acid sequence that
differs from that of the endogenous hormone and the epoetins. Darbepoetin alfa has a longer
half-life (24-26 h) compared to that of the epoetins (4-8 h) on intravenous application (168).

8.9.5   Safety profile of recombinant human Epo
Safety data for the treatment with rhEpo in humans are available from its use as antianaemic
drug since 1989, from more recent experimental approaches using rhEpo as a tissue-protective
agent and from pharmacological studies in healthy individuals.
RhEpo is routinely used for long-term therapy in patients with chronic renal failure. Given in a
dose of 50-150IU/kg per week, it has been shown to be safe, both for short-term and long-term

use. Known rare adverse effects are aggravation of pre-existing hypertension and hypertensive
encephalopathy, seizures, a so-called “flu-like” syndrome and arteriovenous thrombosis. The
hypertension is partly based on increased packed cell volume (for review see (214) and (215)).
Additionally, neutralising anti-Epo antibodies and pure red-cell aplasia can develop in patients
with the anaemia of chronic renal failure during treatment with rhEpo, especially long-term
treatment (216). The safety of the antianaemic treatment in cancer patients is currently under
debate based on possible induction of cancer growth (217).
Antianaemic treatment with rhEpo in premature (birth weight <1500g) and infants with
extremely low birth weight (birth weight <800g) caused short changes in the production of
thrombocytes without clinical impact (218). Conflicting results concerning a possible induction of
retinopathy of prematurity and of necrotising enterocolitis in these children – two conditions
frequently seen in very small newborns - are reported (219-221).
In a small safety and proof-of-concept trial in stroke patients using rhEpo as a neuroprotective
drug, rhEpo was well tolerated and not associated with changes in blood pressure (212). Also in
a small pilot study of patients with acute myocardial infarction, treatment with a single dose of
darbepoetin alfa (300µg) caused no adverse effects during the 30-day follow-up (222).
Finally, in pharmacological studies, very high doses of rhEpo have been given to healthy
individuals as single subcutaneous doses of 300, 450, 600, 900, 1200, 1350, 1800, and 2400
IU/kg without adverse effects (223). In the light of these data, rhEpo appears to have the
potential to become a safe and efficacious neuroprotective agent (224).

8.10 Metallothionein in cerebral pathology
Metallothioneins (MTs) constitute a family of small proteins (molecular weight 6-7 kDa)
characterised by a high metal [Zn(II), Cu(I)] content and also by an unusual cysteine
abundance. Most cells in the mammalian body possess the ability to express MTs. Within the
mammalian brain, MTs occur in isoforms designated MT-I to MT-IV; MT-I+II are the best
characterised MT isoforms which are regulated and expressed coordinately and rapidly induced
by many pathogens. During CNS pathology levels of MT-I+II mRNA and proteins are
significantly increased in reactive astrocytes (225). To some extent, MT-I + II are also
increased in the vascular endothelium, choroid plexus, ependyma, activated microglia ⁄
macrophages, and meninges (226). MT-I + II have been considered to be strictly intracellular
proteins (227), but it is now found that astroglia may secrete MT-I + II to the extracellular space
in order for them to protect the surrounding neurons (228). MT-I+II are considered to have a
wide range of protective functions in the CNS. They are antioxidant and antiapoptotic factors
scavenging ROS and reducing oxidative stress and apoptosis during various CNS disorders

such as epilepsy, traumatic injury, meningoencephalitis, and neurodegenerative diseases (225).
Following traumatic brain injury, MT-I+II induce wound healing, glial scar formation and
angiogenesis (229-231). MT-I+II also reduce cerebral oedema and infarct volume and improve
the functional outcome after transient focal cerebral ischemia (232). Treatment with MT-II in
mice with experimental autoimmune encephalomyelitis can decrease the expression of pro-
inflammatory cytokines in the brain such as IL-6 and TNF-α as well as significantly reduce the
number of apoptotic neurons (233).

9 Neuronal Apoptosis, Metallothionein Expression and
Proinflammatory Responses During Cerebral Malaria in Mice

Lothar Wiese1,2,3, Jørgen A. L. Kurtzhals2,3, Milena Penkowa1

 Section of Neuroprotection, Department of Medical Anatomy, The Panum Institute, Faculty of Health
Sciences, University of Copenhagen; 2Department of Clinical Microbiology, Copenhagen University
Hospital (Rigshospitalet); 3Center for Medical Parasitology, University of Copenhagen

9.1   Abstract

Cerebral Malaria (CM) is an acute encephalopathy in humans due to the infection with
Plasmodium falciparum. Neuro-cognitive impairment following CM occurs in about 10% of the
treated survivors, while the precise pathophysiological mechanism remains unknown.
Metallothionein I+II (MT-I+II) are increased during CNS pathology and disorders. As previously
shown, MT-I+II are neuroprotective through anti-inflammatory, antioxidant and antiapoptotic
functions. We have analysed neuronal apoptosis and MT-I+II expression in brains of mice with
experimental CM.
Methods C57BL/6j mice, infected with P. berghei ANKA were studied on day 7, day 9, and
when presenting signs of CM on day 10-12. We investigated brain histopathology by
immunohistochemistry and TUNEL (Terminal deoxynucleotidyl transferase (TdT)-mediated
deoxyuridine triphosphate (dUTP)-digoxigenin nick end labelling). For statistics we used
quantitation (cellular counts) of the analyzed variables.
Results: During CM we observed significant inflammatory responses of F4/80+
microglia/macrophages and GFAP+ reactive astrocytes, and increased immunoreactivity of 8-
oxoguanine (marker of oxidative stress). As novel findings we show:
1. A localised CM-induced neuronal apoptosis (detected by TUNEL) indicating severe and
irreversible pathology.
2. A significant increase in MT-I+II expression in reactive astrocytes, macrophages/microglia
and vascular endothelium.
Interpretation: This is the first report showing apoptosis of neurons in CM by TUNEL, pointing
out a possible pathophysiological mechanism leading to persisting brain damage. The possible
neuroprotective role of MT-I+II during CM deserves further attention.

9.2    Introduction
Plasmodium falciparum malaria remains a massive burden of disease and death. The two major
complications, cerebral malaria (CM) and severe anaemia are responsible for more than 1
million deaths/year mainly in children under 5 years of age in sub Saharan Africa (234). At least
a similar number of children is likely to be left with serious neurological impairment following CM
Although CM has been studied extensively, many of the pathophysiological mechanisms remain
unclear and treatment other than antiparasitic medication is not available. CM is an acute
encephalopathy with increased levels of proinflammatory cytokines like tumor necrosis factor
(TNF), interferon-γ (IFN-γ), and lymphotoxin (LT) (26;235). Postmortem examination of the
brain reveals haemorrhages, oedema and adherence of parasitized red blood cells (pRBCs) to
the cerebral microvasculature. The CNS microvasculature may eventually become obstructed
by the pRBC adherence, and thus ischemia and metabolic dysfunctioning are seen, which
contribute to the CM pathophysiology (39). However, it is not known how pRBC, which do not
leave the vascular bed, influence parenchymal brain function to induce coma and in some
cases lead to persisting brain damage. The endothelial lining is likely to play a major role in this
A histopathological equivalent to the clinically observed persisting brain damage has not been
identified. Neuronal necrosis is not a feature of CM, but high levels of reactive oxygen species
(ROS) are generated leading to oxidative stress and neuroglial degeneration (236;237). ROS
and oxidative stress are important inducers of apoptosis, a genetically controlled, active form of
cell death characterized by surface blebbing, contraction of cells and their nuclei, proteolysis,
and DNA digestion (238). It differs from necrosis in its programmed manner, complex regulatory
mechanisms and distinctive morphological changes (239).

Metallothioneins (MTs) constitute a family of small proteins (molecular weight 6-7 kDa)
characterized by a high metal [Zn(II), Cu(I)] content and also by an unusual cysteine
abundance. Most cells of the mammalian body possess the ability to express MTs. Within the
mammalian brain, MTs occur in isoforms designated MT-I to MT-III; MT-I+II are are the best
characterized MT isoforms which are regulated and expressed coordinately and rapidly induced
by many pathogens. During CNS pathology levels of MT-I+II mRNA and proteins are
significantly increased in reactive astrocytes and microglia/macrophages (225).
MT-I+II have a wide range of protective functions in the CNS. They are antioxidant and
antiapoptotic factors scavenging ROS and reducing oxidative stress and apoptosis during
various CNS disorders such as epilepsy, traumatic injury, meningoencephalitis, and
neurodegenerative diseases (225). Following traumatic brain injury, MT-I+II induce wound

healing, glial scar formation and angiogenesis (229-231). MT-I+II also reduce cerebral oedema
and infarct volume and improves functional outcome after transient focal cerebral ischemia
(232). Treatment with MT-II in mice with experimental autoimmune encephalomyelitis can
decrease the expression of pro-inflammatory cytokines in the brain such as IL-6 and TNF-α as
well as significantly reduce the number of apoptotic neurons (233).

Experimental cerebral malaria (ECM) in mice is an animal model widely used for the study of
the CM pathophysiology (27). Infection with P. berghei ANKA (PbA) in specific murine strains
causes neurological symptoms such as seizures and coma followed by death. The
histopathological features are comparable to those of human CM, including petechial
haemorrhages, oedema and disruption of the blood-brain barrier (BBB). However, in contrast to
humans, the mice show vascular adhesion of monocytes rather than pRBCs (27;240). The
number of pRBCs used for inoculation effects the development of ECM in mice. A low infective
dose frequently leads to ECM whereas a high infective dose does not cause ECM but severe
anaemia and hyperparasitaemia (241).

In this report we have analysed cerebral apoptosis and MT-I+II expression in the brain of mice
during the course of ECM and compared to normal controls using immunohistochemistry and

9.3   Materials and methods
Female pathogen-free C57BL/6j mice 8-10 weeks old, weighing 18 –25g, were purchased form
Møllegaard, Denmark. All animals were pathogen free and were kept under standardized
conditions at the animal facilities at the Panum Institute, University of Copenhagen, with free
access to food and water ad libitum. All experiments adhered to Danish and European
guidelines for animal research and were approved by the national board for animal studies. All
efforts were made to minimize animal suffering and to reduce the number of animals used.

Induction of ECM
Of 18 mice, sixteen were infected with P. berghei ANKA parasites, while two remained
uninfected as healthy controls. Those infected were inoculated intraperitoneally with 104
parasitized red blood cells (pRBC) from mice of the same strain diluted in normal saline on day
0. Parasitaemia was determined by Giemsa stained thin blood films from tail blood. The animals
were under daily supervision for clinical signs of disease. Body temperature was measured

rectally (digital thermometer DM852 with rectal probe, Ellab, Denmark) on day 1 and twice daily
from day 7 on. Neurological symptoms were recorded every day. ECM was diagnosed by
clinical signs including ataxia, paralysis (mono, hemi, para, or tetraplegia), deviation of the
head, convulsions, and coma and in addition by a sudden drop of body temperature to values
under 34°C. This drop is seen in mice developing ECM shortly before death and is closely
related to the development of severe neurological symptoms (242).
The infected mice were divided into 3 groups to be killed on day 7 (5 mice), day 9 (6 mice) or
after drop of body temperature below 30°C or when showing severe signs of ECM (5 mice).
Healthy control mice were injected with normal saline only instead of pRBCs and killed on day

Tissue processing
The mice were deeply anaesthetized by i.p. injection of 0.1ml/10g body weight of a solution of
Hypnorm®/Dormicum® (Hypnorm®: fentanyl citrate 0.315 mg/ml and fluanisone 10mg/ml,
Janssen Pharma, Denmark; Dormicum®: midazolam 5 mg/ml, Roche, Switzerland).The mixture
for anasthaesia is made from mixing 1 part Hypnorm® with 1 part sterile water and adding this
to a mixture of 1 part Dormicum® plus 1 part sterile water. They were transcardially perfused
with 0.9% saline with 0.3% heparin (15,000 IU/L) for 2-3 min, followed by perfusion with
Zamboni's fixative (buffered 4% formaldehyde added to 15% picric acid solution from a 1.2%
saturated aqueous picric acid solution, pH 7.4) for 6-8 min. Afterwards the brain, liver and
spleen were fixed by immersion in Zamboni's fixative, pH 7.4, for 2-4 h at room temperature and
stored in 70% alcohol. The brains were dehydrated according to standard procedures,
embedded in paraffin, and cut in serial 3-μm-thick sections, to be used for hematoxylin-eosin
(H&E), immunohistochemistry and terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-
biotin nick end labeling (TUNEL).
The sections were rehydrated and, where necessary for heat-induced antigen retrieval, boiled in
citrate buffer or TBS buffer, pH 9.0 or pH 6.0, in a microwave oven for either 10 or 30 min. After
cool-down to room temperature, the sections were incubated in 1.5% H2O2 in Tris-buffered
saline (TBS)/Nonidet (TBS: 0.05 M Tris, pH 7.4, 0.15 M NaCl; with 0.01% Nonidet P-40)
(Sigma-Aldrich, USA, code N-6507) for 15 min at room temperature, to quench endogenous
peroxidase. The sections were then incubated with either 10% goat serum (In Vitro, DK, code
04009-1B) in TBS/Nonidet for 30 min at room temperature in order to block nonspecific binding.
Where monoclonal mouse-derived antibodies were used, incubation with blocking solutions A
and B from HistoMouse-SP Kit (Zymed Lab, USA, code 95-9544) were applied, to quench
endogenous mouse IgG.

Hematoxylin and eosin (HE) staining of brain sections was carried out according to standard

The sections were incubated overnight at 4°C with one of the following antibodies: Rabbit anti-
cow GFAP, 1:250 (Dako, DK, code Z 334) (as a marker for astrocytes). Mouse anti-8-
oxoguanine, 1:100 (Chemicon, USA, code MAB-3560) (as a marker for oxidative stress). Rabbit
anti-human Albumin, 1:5000 (Dako, DK, code A0001) (as a marker for impaired blood-brain-
barrier function). Rat anti-mouse F4/80, 1:15 (Serotec, USA, code MCA 497) (as marker for
macrophages/microglia). Rabbit anti-rat MT-I+II, 1:500 (kindly provided of Dr. J. Hidalgo,
Barcelona (243;244)). Mouse anti-human P53, 1:50 (Dako, DK, code M7001). Rabbit anti-
human Caspase 3, 1:50 (Cell Signaling, USA, code 9661). These primary antibodies were
detected using biotinylated goat anti-rat IgG,1:1,500 (Amersham, UK, code 1005), or
biotinylated mouse anti-rabbit IgG, 1:400 (Sigma-Aldrich, USA, code B3275), or biotinylated
goat anti-mouse IgG, 1:200 (Sigma-Aldrich, USA, code B8774), or biotinylated goat anti-mouse
IgM ( -chain specific), 1:10 (Jackson ImmunoResearch Lab, code 115-065-020), or biotinylated
donkey anti-sheep/goat IgG, 1:20 (Amersham, UK, code RPN 1025), by incubating the sections
for 30 min at room temperature followed by StreptABComplex/HRP (as above). Where needed
sections were incubated with biotinylated tyramide and streptavidin-peroxidase complex (NEN,
Life Science Products, USA, code NEL700A) prepared following the manufacturer's
recommendations. The immunoreaction was visualized using DAB as a chromogen (as above).

In Situ Detection of DNA Fragmentation (TUNEL)
Terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine triphosphate (dUTP)-
digoxigenin nick end labelling (TUNEL) staining was performed according to manufacturers
protocol and after tissue processing as mentioned above. Sections were deparaffinized and
TUNEL was accomplished using the TdT-FragEL DNA Fragmentation Detection Kit (Oncogene
Research Products, Cambridge, UK; code Cat# QIA 33). After immersion in equilibration buffer
for 20 min, sections were incubated with TdT and dUTP-digoxigenin in a humidified chamber at
37°C for 1,5 hrs and then incubated in the stop-wash buffer at 37°C for 5 min to stop the
reaction. DAB was used as chromogen, and the sections were counterstained with methyl

Immunofluorescence and fluorescein-linked TUNEL
In order to determine which cells contain MT-I+II during CM, double- and triple
immunofluorescence stainings were performed. For this, we used monoclonal mouse anti-horse
MT-I+II 1:50 (Dakopatts, DK, code M0639) simultaneous with polyclonal rabbit anti-cow GFAP
(like mentioned above) or simultaneous with rat anti-mouse F4/80 (like mentioned above).
These primary antibodies were visualised by using goat anti-rabbit IgG linked with fluorescein
(FITC) 1:50 (Southern Biotechnology Ass., USA, code 4050-02); goat anti-mouse IgG linked
with aminomethylcoumarin (AMCA) 1:30 (Jackson ImmunoResearch Lab, USA, code 115-155-
146); ); goat anti-rat IgG linked with FITC 1:30 (Jackson ImmunoResearch Lab, USA, code 112-
095-102). To determine the cells suffering oxidative stress we used Mouse anti-8-oxoguanine
(as mentioned above) simultaneous with rabbit anti-VEGF 1:50 (Neomarkers, USA, code RB-
222P0). These were visualised by using goat anti-rabbit IgG linked with Texas Red (TXRD)
1:50 (Jackson ImmunoResearch Lab, USA, code 111-075-144) and goat anti-mouse IgG 1:30
(Jackson ImmunoResearch Lab, USA, code 115-095-075). To determine the cells suffering
apoptosis and their MT-I+II levels, we used fluorescein-linked TUNEL (FragEL DNA
fragmentation kit, Calbiochem, USA, code QIA39) simultaneous with mouse anti-horse MT-I+II
(as mentioned above) and Rabbit anti-NSE (neuron specific enolase) 1:1000 (Calbiochem,
USA, code D05059). These were visualised by using
goat anti-rabbit IgG linked with Texas Red (TXRD) and goat anti-mouse IgG linked with AMCA
(both as mentioned above). The sections were embedded in Fluorescent mounting (Dakopatts,
DK, code S3023) and kept in darkness at 4oC. In order to evaluate the extent of non-specific
binding of the antisera in the fluorescence stainings, control sections were incubated in the
absence of primary antibody. Results were considered only if these controls were negative.

Sections were examined on an Axioplan 2 light microscope (Zeiss, Germany) equipped with a
tripleband (FITC/TXRD/AMCA) filter. Images were recorded using a digital camera (Coolsnap
1.2; RS Photometrics, USA) and processed using Adobe Photoshop 5.5 (Adobe, USA).

In addition to morphological analysis, cellular counting was carried out in a blinded and
randomized manner by the same person. One section per animal was used. Bleeding spots and
TUNEL+ cells defined were counted from sagittal sections from both hemispheres, F4/80
positive and 8-oxoguanine positive cells from sagittal sections from one hemisphere.
Oxoguanin+ cells were counted in groups, because mainly endothelial cells showed positive
reaction. Due to the anatomy of the small vessels a quantification of single 8-oxoguanin+ cells

could not be performed. Instead of we counted groups of positive cells. Cells connected to
each other were seen as one count. We accepted the probability to count a single vessel more
than once, due to the fact that a one vessel can be cut more than one time in a two-dimensional
section. Caspase-3+ and P53+ cells were counted from corresponding 0,32-mm2 big areas in
the pons. In the case of TUNEL, nuclear staining of round shaped, parenchymal nuclei were
counted thereby focussing on neurons.

Statistical analysis
Results were evaluated by student-t test when comparing 2 groups and by one-way analysis of
variance (ANOVA) when comparing differences in the mean values among the four groups:
uninfected controls, infected animals killed on day 7 and day 9 respectively and terminal ill
animals. Raw data were processed by log- or ln-transformation to improve the homogeneity of
variances where necessary. A difference was considered significant when p<0.01.

9.4   Results
Mice killed on day 7 and 9 did not show clinical signs of ECM, nor did the body temperature
drop. The remaining 5 infected mice presented signs of severe ECM like ataxia, convulsions or
coma. Four of these mice were killed and perfused when showing signs of terminal illness i.e.
coma and hypothermia, one mouse was found dead on day 11 and could not be perfused. No
significant difference in the severity of neurological symptoms was observed in this group. No
clinical signs of ECM were seen prior to day 10. Our clinical observations correspond well to
previous descriptions (30). Brain sections were obtained from all three groups as well as
healthy controls. Mice killed on day 7 showed parasitaemia of 1.9% (range: 0.5-4.6%), those
killed on day 9 of 6,6% (range: 0.6-11.4%) and those killed with severe CM of 7.5% (range: 1-

Histopathological features
The CM Histopathological features were studied in H&E stained brain sections. All terminally ill
animals had small haemorrhages (mean: 11.8, range: 3-30, standard error: 6.2) like shown in
figure 1A. In 3 out of the 6 brains of animals killed on day 9, brain haemorrhages were seen.
Mice examined on day 7 post inoculation and controls did not show this feature. Figure 1B
illustrates the time course as seen in our experiment showing haemorrhages as a late feature (p
= 0,007). The bleeding sites were distributed throughout the brain and topographical patterns
could not be detected, though the olfactory bulb and the brain stem were frequently involved.
Morphological differences of the haemorrhages other than size could not be detected.

Immunoreactivity for F4/80 revealed an increasing number of macrophages/microglia (Mφ) in
the brain parenchyma during development of CM (p=0.002). Hence, a 2-fold increase in brain
Mφ was seen by day 7 (mean: 180, range: 110-296, standard error: 38.8) and 9 (mean: 197,
range: 118-427, standard error: 58.6) relative to healthy controls (mean: 75.5, range: 65-86,
standard error: 10.5; p= 0.033), while a 10-fold increase in terminally ill mice (mean: 767.8,
range: 341-1386, standard error: 232.8; p= 0.01) was seen (Fig. 1F). In terminal ill mice Mφ
showed a more activated phenotype (round instead of ramified) as CM deteriorates, when
compared to early CM stages and controls, where mostly resting (ramified) microglia was seen
(Fig. 1E). Activated Mφ were seen in multifocal infiltrates. The olfactory bulb is frequently
involved, but no clear pattern was visible.

Reactive Astrogliosis
The reactive astrogliosis determined by GFAP immunoreactivity and judged by cell morphology
was significantly increased in infected animals compared to uninfected controls. In control mice
GFAP+ astrocytes were unaffected, whereas they were clearly reactive with hypertrophy and
hyperplasia in infected animals (Fig. 1C and 1D). A successive increase in the size of
astrocytes indicating activation was seen from day 7 and onwards (data not shown). All parts of
the brain including cortex, midbrain, brain stem, cerebellum and olfactory bulb showed patches
of reactive astrocytes.

Blood brain barrier (BBB) properties
For evaluation of the BBB properties in control and CM infected mice, we used immunostaining
for serum albumin, which showed an intact BBB in control mice, while the BBB of severely CM
diseased or terminal mice was impaired and albumin was observed in the extracellular space as
well as intracellularly in perivascular phagocytotic cells. (Fig. 1G and 1 H). The BBB impairment
was most pronounced in the severely diseased or terminal animals, while BBB leakage was
marginal on day 7 and 9 after inoculation (data not shown). A clustered distribution of the
damage of the BBB was seen without clear sites of predilection. Albumin immunoreactivity
could not be quantified exactly by cellular countings since albumin was mainly observed in the
extracellular space.

Oxidative stress
Oxidative stress levels were determined by immunoreactivity for 8-oxoguanine (8-oxo). The
mean number of 8-oxo+ areas was increasing from 50.5 in the controls (range: 18-83, standard

error: 32.5) and 36.2 on day 7 (range: 19-57, standard error: 6.6) to 159 on day 9 (range 33-
338, standard error: 52.9) and 374.8 in moribund mice (range: 207-448, standard error: 56.4) as
shown in Fig. 2B (p<0.001). The predominant morphology of 8-oxo+ cells was that of
endothelial cells (Fig. 2A) with single perivascular cells (data not shown). This was confirmed by
using double immunofluorescence, which showed that 8-oxo+ and VEGF+ cells co-localized to
the vascular endothelium (Fig. 2C). In healthy control mice 8-oxo immunoreactivity was low or
practically absent (data not shown). Endothelial cells immunoreactive for 8-oxo were found to
be more evenly distributed throughout the brain when compared to the features described
above like activation of microglia/macrophages, reactive astrogliosis and impairment of the BBB
where a clustered distribution dominates.

MT-I+II expression during CM
In brain sections of control mice, MT-I+II expression was only observed in meningeal,
ependymal and choroid plexus cells and in a few scattered astroglial cells as previously shown
(245). After induction of CM, a significant increase in MT-I+II expression was seen as shown in
Fig. 3B (p=0.026). The mean number of MT-I+II positive cells increased from 79 in controls
(range: 76-82; standard error: 3.0) and 114.2 on day 7 (range: 55-179, standard error: 23.4) to
286.7 on day 9 (range: 181-414; standard error: 32.1) and 276.2 in moribund mice (range: 34-
485, standard error: 74.6). MT-I+II were mainly expressed in reactive astrocytes (Fig. 3C) and
to a lesser extend in Mφ (macrophages/microglia) (Fig. 3D), vascular endothelium and choroid
plexus, meninges and ependymal cells. Generally MT+ cells were seen in the gray matter and
only occasionally in the white matter.
There was marked heterogeneity within and between cases in the extent of MT-I+II expression
in different brain regions. The paraseptal nuclei were always involved, while the majority of
terminally ill mice showed clusters of MT-I+II immunoreactivity in the olfactory bulb, cerebellum
and the thalamus. A distinct accumulation was seen in submeningeal regions of the cerebral
cortex (data not shown).
Virtually all clusters of MT-I+II immunoreactivity where paralleled by activated GFAP+
astrocytes and 8-oxo+ cells indicating that oxidative stress was almost always present. In
contrast there was no clear topographical relationship to cerebral haemorrhages, apoptotic cells
and impaired BBB (data not shown).

Apoptosis was determined by TUNEL, a method labeling the nick end of fragmentated DNA as
seen during apoptosis. Two types of TUNEL+ nuclei were identified: Flat shaped ones
morphologically identified as endothelial cells and round shaped nuclei (Figure 4A+B). Double

staining for TUNEL and neuron specific enolase (NSE), identified round shaped TUNEL+ nuclei
as neurons (Figure 3C). Terminally ill mice showed clusters of apoptotic neurons while healthy
controls and animals on day 7 and day 9 only showed single TUNEL+ cells. Apoptotic
endothelial cells and apoptotic neurons were only occasionally seen in close topographical
relation like shown in (Fig. 4B) and TUNEL+ endothelial cells occurred without topographical
relation to apoptotic neurons (Fig. 2D). The mean number of round shaped, parenchymal
TUNEL+ cells was significantly increased during CM from 8.5 (range: 5-12, standard error: 3.5)
in healthy controls to 322.5 (range: 36-513, standard error: 108.3) in the brains of terminally
mice (Fig. 4D; p=0.001). The day 7 group included a single animal with a more than 100-fold
higher number of round shaped TUNEL+ compared all other 5 mice while the parasitaemia was
4 times higher than the mean of the group. It was regarded as an outlier due to variance of the
time course of the model and excluded from statistical analysis.
In order to support the finding of apoptosis we examined the tissue for immunoreacitvity for
cleaved caspase-3 and p53 both involved in the signalling cascade of apoptosis (246). We
found that both were increasingly induced over the time course of murine CM (Fig. 4F and 4H).
Immunoreactivity to cleaved caspse-3 and P53 was seen in two different patterns: Regularily in
the brain stem and patch-like distributed with marked interindividual differences in other parts of
the brain. In some cases we were able to indentify caspase-3+ endothelial cells (Fig. 4E, arrow)
A clear topographical correlation between the immunoreactivity for p53 and Caspase 3 on the
one hand and TUNEL+ neurons could not been established.

9.5   Discussion
In this study, we have demonstrated apoptotic cell death and the expression of MT-I+II protein
in the brains of mice with cerebral malaria. In addition, we describe a number of
histopathological findings in the brains of mice with CM that parallel the human disease. As the
availability of tissues from post-mortems is restricted by religious and cultural objections to
autopsy only a limited number of histopathological studies in patients that died from CM have
been carried out. An animal model is therefore an appropriate approach to further elucidate the
pathophysiological mechanisms. The model used for this study (C57B/6j mice infected with PbA
parasites) mimics principal features of the human disease and is therefore a valuable tool to
study CM. It has been widely used in previous studies, representing a substantial amount of
data for comparison (27).
It is widely agreed that the histopathological features of human cerebral malaria include
sequestration of pRBCs in the brain and other organs, cerebral haemorrhages, cerebral
oedema and activation and degeneration of cerebral endothelial cells. Activated astrocytes,

oxidative stress of cerebral endothelial cells, microglial activation and a patchy distribution of
diseased areas in the brain are commenly found in autopsy material of patients who died of CM
(159;247;248). Several hypotheses try to explain the pathogenesis of CM: Mechanical
obstruction caused mainly by cytoadherence of pRBC to the endothelial lining of cerebral blood
vessels; localized hypoxia; breakdown of the BBB; receptor mediated signalling across the BBB
and production of neuroactive mediators in cerebral vascular cells. There is evidence for all of
these mechanisms in different studies and the disease is more and more seen as multifactorial
(reviewed by(240). A recent in vivo study applied magnet resonance techniques in a murine
model of CM and was able to show inflammatory and ischemic lesions as well as impairment of
the BBB and cerebral edema (249).
In this study extravascular immunoreactivity for albumin occurred mainly in terminally ill
animals. This feature can be due to slow low-grade leakage of the BBB and accumulation of
extra vascular albumin as well as more sudden breakdown of the BBB in severely diseased
animals. The pronounced co-localisation of immunoreactivity of VEGF and oxidative stress
could be causally related as hypoxia has been shown to trigger the expression of VEGF in
endothelial cells (250). Numerous activities of VEGF have been discovered not least its
involvement as an angiogenetic factor in tumor growth. A possible role in the pathology of CM
could be based on its ability to induce antiapoptotic proteins like Bcl2 in endothelial cells and its
well documented ability to induce vascular leakage as reviewed by Ferrara (251).
The mechanisms that cause irreversible neurological dysfunctions have not been satisfactorily
resolved. Several clinical studies in Africa have shown residual neurological deficits in a
significant number of children surviving CM. Most children recover entirely, but persisting
deficits are seen in 4.4-9.7% after hospital treatment for CM. The clinical picture of these
persisting deficits is widespread and ranges from mono- and hemiplegia, ataxia, propensity for
epilepsy, cortical blindness and hearing-, speech- and cognitive deficits (8;64;252-254). It is
likely that these dysfunctions are due to structural damage in various parts of the brain.
Here we propose neuronal apoptosis as a possible morphological correlate for the observed
irreversible brain damage following CM. We also show expression of MT-I+II proteins in murine
ECM as these molecules have neuroprotective and immunoregulating properties in other
cerebral disorders and potentially could play a role in CM.
To our knowledge this is the first report showing clustered neuronal apoptosis in ECM. Using
the TUNEL technique we found apoptosis of neurons as well as apoptosis of endothelial cells in
the brains of mice with ECM.
In general apoptosis of neurons is seen in various cerebral pathologies including cerebral
ischaemia, traumatic brain injury and epilepsy (255) and so far apoptosis of host tissue cells

due to P. falciparum infection has been described in lymphocytes (256), hepatocytes (257),
endothelial cells (44;237) and placental cells (258).
By inducing ECM in Swiss Albino mice infected with PbA Kumar and Babu were able to show
upregulation of apoptosis related genes in the cortex of terminally ill animals (157). Their
findings include a high Bax/Bcl2 ratio as well as induction of P53, while a post mortem study of
brainstems from Vietnamese patients that had died of CM was able to show immunoreactivity
for caspase-3 related to neurons in four out of 10 cases (159). As neurons with typical nuclear
morphology of apoptosis were not found, the authors conclude that irreversible damage of
neurons in the brainstems of these malaria patients is unlikely to occur via apoptosis. Another
study on human post mortem brains from patients with CM demonstrated single TUNEL+
neurons (158).
In our study neuronal and endothelial apoptosis (TUNEL+ cells) and cerebral haemorrhages
were only observed in terminally ill animals with clear clinical signs of cerebral involvement on
day 10-13 after inoculation. Assuming a role for neuronal apoptosis in human CM, these results
correspond well to the clinical observation that irreversible brain damage is associated with a
prolonged clinical course of human CM (259).
The topographical distribution of apoptotic neurons did not follow a clear pattern though they
were mainly seen in clusters. Clusters of apoptotic neurons were seen in virtually all parts of the
brain with marked inter-individual differences. Endothelial apoptosis likewise showed no clear
topographical pattern. Surprisingly, a close topographical correlation between endothelial
apoptosis and apoptotic neurons was only occasionally seen. This might be explained by
limitations of the 2-dimensional technique used. Recently, Medena et al. proposed axonal injury
as a possible mechanism of brain damage due to CM by studying post mortem brain tissue of
patients who died of P. falciparum malaria (260). Their findings include focal patches of axonal
injuries with a noticeable heterogeneity within and between cases. The distribution of areas with
axonal injury seemed to vary in a similar way as the distribution of apoptotic neurons seen in
our experiments. Also, cerebral hemorrhages were distributed throughout the brain with
marked interindividual differences, though the olfactory bulb and the brain stem were frequently
involved, which corresponds well to previous findings (261).The morphological distributions is a
possible explanation for the widespread range of neurological deficits that are observed after
successful treatment of CM in children.
We compared TUNEL with other markers of apoptosis. Immunoreactivity for cleaved caspase-3
and p53 in our material showed an increase over time and was already seen on day 7 prior to
the detection of neuronal apoptosis by TUNEL. This matches the fact that both molecules are
part of signalling cascades that lead to apoptosis (255). Surprisingly, our results fail to show a

close topographical relationship between the expression of caspase-3 and p53 on the one hand
and TUNEL+ neurons on the other hand as would be expected.
This could have importance for the interpretation of studies that use caspase-3 as a marker for
apoptosis. The Study of Vietnamese adult patients who died of CM mentioned above used
caspase-3 as a marker for apoptosis but no clear evidence for neuronal apoptosis due to CM
was detected (159). An explanation could be found in caspase-3 independent apoptosis-
pathways (reviewed in (255)) and further examination is needed.
Apoptosis of neurons as shown here in an animal model could be the pathophysiological
mechanism causing neuro-cognitive impairment in patients surviving cerebral malaria.
Neuronal apoptosis is as well a potential target for neuroprotective adjuvant treatment in CM
and the time course of our findings suggest a window for possible therapeutic intervention with
anti-apoptotic therapy.

We were also able to show that MT-I+II are up regulated during ECM in mice. This MT-I+II
expression reached maximum levels in mice with severe clinical signs of cerebral disease 10-12
days after inoculation of parasites. The physiological role of MTs is a subject of active
investigation and MT-I+II are implicated in diverse physiological an pathophysiological functions
in the CNS such as ion metabolism, regulation of inflammatory responses, protection against
reactive oxygen species and oxidative stress and reduction of apoptotic cell death. They have
been found to be significantly upregulated in the human brain in multiple sclerosis and other
autoimmune diseases (225;262). MT-I+II expression is also increased after traumatic brain
injury (229), kainic induced seizures (263), administration of gliotoxic 6-aminonicotinamide (231)
and experimental autoimmune encephalitis (EAE – a model for multiple sclerosis) (264) as well
as TNF induced CNS damage (265) in animal models. MT-I+II protein expression has also
been revealed during neurodegenerative diseases such as Alzheimer’s disease (266;267),
amyotrophic lateral sclerosis (268), multiple sclerosis (269), in all of which oxidative stress has
been involved as a significant factor contributing to tissue injury.
MT-I+II upregulation has been observed at the mRNA and the protein level, primarily expressed
in astrocytes and to a lesser extend also in microglia (270). Several studies using MT-
overexpressing and MT-knockout mice have demonstrated a neuroprotective and
immunomodulative capacity of MT-I+II in several conditions including focal cerebral ischaemia
(232), focal cryolesions (229) and administration of gliotoxic 6-aminonicotinamide (231). Mice
with genetic MT-I+II deficiency display reduced reactive astrogliosis and increased macrophage
activation after brain injury (230) suggesting a role for endogenous MT-I+II production in
regulating the brain inflammatory response.

This is the first report showing MT-I+II expression in a pathological condition of the brain
caused by parasitic infection. In parallel to other pathological conditions of the brain, astrocytes
were the main cell type producing MT-I+II. Further studies are needed to assess whether MT-
I+II have functions in CM similar to those in other diseases of the brain.
Interestingly, intraperitoneal injection of MT-II in rats, which had undergone cryogenic cerebral
lesions or had EAE led, to a reduction in inflammation and possibly even more important to a
decrease in the number of apoptotic neurons (229;233). Further work will be needed to examine
the possible role of MT-I+II during CM and their potential as future treatments.
This is of special interest on the background that many of the immunopathological features of
ECM and CM are found in EAE and multiple sclerosis in a similar way (113;271): Both
conditions share the activation of astrocytes and microglia, the production of TNF-α, impairment
of the blood brain barrier, axonal injury, apoptosis of neurons and activation of the vascular
endothelium even though ECM is an acute condition in contrast to the chronic time course of
The importance of the immunopathological component of CM is further illustrated by the fact
that the pathogen (i.e. the parasite) is inducing pathology in the brain parenchyma without close
contact to the effected neurons, unlike bacteria in the case of bacterial meningitis. Except of in
the case of hemorrhages the parasite is not leaving the vascular bed, unlike in other infectious
conditions of the brain like for example bacterial meningitis.
Focus on the immunopathology of CM and its similarities to autoimmune diseases could lead to
new strategies in the development of neuroprotective drug targets and further experiments are

9.6    Acknowledgements
Thanks to Lars Hviid for critically reviewing the manuscript and to Ming Chen for valuable
practical advice and help. The excellent technical assistance of Grethe Gomme, Hanne
Hadberg, Pernille S. Froh and Ha Nguyen is gratefully acknowledged. These studies were
supported by The Lundbeck Foundation, The Danish Medical Research Council, IMK Almene
Fond, Kathrine og Vigo Skovgaards Fond, The Danish Medical Association Research Fund,
Toyota Fonden, Fonden til Lægevidenskabens Fremme, Eva & Henry Frænkels Mindefond, Dir.
Leo Nielsens Legat, Karen A Tolstrup, Hørslevfonden.

9.7    Figures

Fig. 1. Typical histological and immunohistochemical features in the brains of mice with CM. Petechial cerebral
hemorrhages (Hematoxilin – Eosin staining) (A). Number of hemorrhages counted from both hemispheres (B). Glial
fibrillary acidic protein (GFAP) immunoreactivity in control mice showing slim unactivated GFAP+ astrocytes (C).
Enlarged, activated GFAP+ astrocytes in terminal ill animals (D). Enlarged, activated F4/80+
microglia/macrophages in terminal ill animals (E). Number of F4/80+ cells counted from one hemisphere (F).
Immunoreactivity for albumin in terminal ill mice (G and H). Two different patterns are seen: A diffuse impairment
of the blood brain barrier (BBB) (G) and localized impairment here shown here shown with additional intracellular
staining of phagocytotic cells containing albumin (H). All counts from sagittal sections near the mid-line. Statistical
analyses by one way ANOVA. Graph: Bars: group means; error bars: standard error of the mean. Scale bars = 50

Fig. 2. Oxidative stress and apoptosis of endothelial cells in the brains of mice with CM. Immunoreactivity for 8-
oxoguanin in a terminal ill mouse indicating oxidative stress (A). Number of 8-oxoguanin+ areas counted in a
sagittal section near the midline from one hemisphere (see text for details) (B). Double immunofluorescence for
vascular endothelial growth factor (VEGF, red) and 8-oxoguanin (green) shows nearly exclusively double staining
(yellow) localizing the oxidative stress in endothelial cells (C). TUNEL+ endothelial cells indicating apoptosis in
cortex of a terminally ill animal (D). All counts from sagittal sections near the mid-line. Statistical analyses by one
way ANOVA. Graph: Bars: group means; error bars: standard error of the mean. Scale bars = 50 µm.

Fig. 3. Expression of metallothionein I+II (MT-I+II) in the brains of mice. MT-I+II reactivity in the cerebral
cortex of a terminally ill mouse showing MT-I+II positive cells (A). Number of MT-I+II positive cells in the four
different groups counted from both hemispheres showing a significant increase over time (p=0.03) (B). Double
immunofluorescence for MTI+II (blue) and glial fibrillary acidic protein (GFAP, green). Double staining
(turquoise) is showing the presence of MT-I+II in astrocytes (C). Double immunofluorescence for MTI+II (blue)
and F4/80 as a specific marker for microglia/macrophages (green). Double staining is showing expression of MT-
I+II in F4/80+ microglia/macrophages (D). All counts from sagittal sections near the mid-line. Statistical analyses
by one way ANOVA. Scale bars = 50 µm.

Fig. 4. Apoptosis in the brains of mice with CM. TUNEL+ cells in the cortex of a terminal ill mouse demonstrating
apoptosis and showing the typical clustering. Most TUNEL+ cells are neurons with big round shaped nuclei. In
addition few endothelial cells were TUNEL+ (arrows) (A and B). Triple immunofluorescence for TUNEL (green),
neuron specific enolase (NSE; red) and MT-I+II (blue). TUNEL + nuclei are mostly seen in close relation to the red
cell bodies of neurons. No double staining of MT-I+II and NSE (turquoise) (C). Numbers of TUNEL+ cells in the
four different groups counted from both hemispheres showing that apoptosis in the brain of mice with CM is a
feature of the terminal disease (p=0.001) (D). Caspase-3+ and p53+ cells in the pons of terminally ill mice,
resembling mainly neurons plus few endothelial cells (arrow) (E and G). Number of caspase-3 and p53+ cells
respectively in the pons counted from a corresponding 0,32-mm2 big area. The increase in the number of positive
cells over time not statistically significant (F and H). All counts from sagittal sections near the mid-line. Statistical
analyses by one way ANOVA. Graph: Bars: group means; error bars: standard error of the mean. Scale bars = 50

10 Recombinant human erythropoietin increases survival in a dose-
and time-dependent manner and reduces neuronal apoptosis in a
murine model of cerebral malaria

Lothar Wiese1,2,3,, Casper Hempel1,2,4, Milena Penkowa3, Nikolai Kirkby2, Jørgen A. L. Kurtzhals1,2

 Center for Medical Parasitology, University of Copenhagen; 2Department of Clinical Microbiology,
University Hospital Copenhagen, Rigshospitalet; 3Section of Neuroprotection, The Panum Institute,
Faculty of Health Sciences, University of Copenhagen; 4Department of Zoology, Faculty of Life Sciences,
University of Copenhagen

10.1 Abstract
Background: Cerebral malaria (CM) is an acute encephalopathy with increased pro-
inflammatory cytokines, sequestration of parasitised erythrocytes and localised ischaemia. In
children, CM induces cognitive impairment in about 10% of the survivors. Erythropoietin (Epo)
has – besides its wellknown haematopoietic properties – significant anti-inflammatory,
antioxidant and anti-apoptotic effects in various brain disorders. We therefore examined the
neurobiological responses to exogenously injected Epo during murine CM.
Methods: Female C57BL/6j mice (4-6 weeks), infected with Plasmodium berghei ANKA, were
treated with recombinant human Epo (rhEpo; 50-5000U/kg/OD, i.p.) at different points in time.
The effect on survival was measured. Brain pathology was investigated by TUNEL (Terminal
deoxynucleotidyl transferase (TdT)-mediated deoxyuridine triphosphate (dUTP)-digoxigenin
nick end labelling) as a marker of apoptosis. Gene expression in brain tissue was measured by
real-time PCR.
Results: Treatment with rhEpo increased survival in mice with CM in a dose- and time-
dependent manner and reduced apoptotic cell death of neurons as well as the expression of
pro-inflammatory cytokines in the brain. This neuroprotective effect appeared to be independent
of the haematopoietic effect.
Conclusions: These results and its excellent safety profile in humans make rhEpo a potential
candidate for adjunctive treatment of CM.

10.2 Introduction
Plasmodium falciparum malaria remains a massive burden of disease with an estimated two
million deaths worldwide each year (1), mainly due to the two major complications, cerebral

malaria (CM) and severe anaemia. Moreover, long-term neurological sequelae in 2-11% of the
survivors of CM are increasingly recognised (4;6;7). CM is an acute – and potentially reversible
– encephalopathy with increased serum-levels of pro-inflammatory cytokines such as tumor
necrosis factor α (TNFα), interferon-γ (IFN-γ), and lymphotoxin (LT) (26;272). Postmortem
examination of the brain reveals haemorrhages, oedema and adherence of parasitised red
blood cells (pRBC) to the cerebral microvasculature. Although CM has been studied
extensively, many of the pathophysiological mechanisms remain unclear, and highly effective
treatment other than antiparasitic medication is not available. A number of adjunctive therapies
have been proposed for CM, including dexametasone, hyperimmune sera, anti-TNF antibody,
pentoxyfylline, osmotic diuretics, heparin, anti-convulsants and red blood cell (RBC) exchange
transfusion. Only anti-convulsants and, to a lesser extent, RBC exchange transfusion have
shown to be partially effective (6).

Erythropoietin (Epo) was first identified as an haematopoietic growth factor produced in the
kidneys, and recombinant human Epo (rhEpo) is widely used to treat anaemia (168). Increased
serum Epo levels promote survival of erythroid precursor cells of the bone marrow that would
otherwise undergo apoptosis (170). The observation that Epo and its receptor are expressed in
practically all brain cells expanded the biological role of Epo beyond haematopoiesis (reviewed
in (171;172)).
In vitro studies have shown that Epo protects neurons against ischemic damage by reducing
apoptosis (184), but the cytoprotective effects are not restricted to neurons: Epo can also
protect endothelial cells (192) and glial cells (193). Apart from its anti-apoptotic effect, Epo may
reduce astrocyte activation (194), cause direct neurotrophic effects (186), act as antioxidant
(195) and stimulate angiogenesis (196).
Subsequent in vivo studies of systemic treatment with rhEpo reported a reduction of neuronal
damage and neurological dysfunction in rodent models after stroke (189;205;206), mechanical
trauma, excitotoxic injury and experimental autoimmune encephalitis (EAE) (181). And,
strikingly, a recent study in a mouse model of CM showed an increase in survival in rhEpo-
treated animals (136).
Moreover, a proof-of-concept trial in human patients with ischaemic stroke has shown a
neuroprotective effect of high-dose rhEpo within 6 hours after the onset of symptoms (212).
Considering these preclinical and clinical data, rhEPO is an attractive, potential candidate for
the therapy of CM. The aim of the present study was to explore the effect of systemicly
administered rhEpo in an animal model of CM, and thereby to contribute to the body of
evidence that can form the basis for a treatment of human patients with CM.

10.3 Materials and methods
Female pathogen-free C57BL/6j mice, 4-6 weeks old and weighing 18 –22g, were purchased
from Taconic, Denmark. We used a total of 199 mice, 149 for the survival studies, 30 for gene
expression analysis and 20 for embedding of the brain and staining. The results were obtained
from three independent experiments. All animals were pathogen-free and were kept under
standardised conditions with ad libitum access to food and water. All experiments were
conducted in accordance with Danish and European guidelines for animal research and were
approved by the national board for animal studies. All efforts were made to minimise animal
suffering and to reduce the number of animals used. The development of ECM is accompanied
by a severe drop of body temperature. Mice with a body temperature below 32°C do not
recover. We therefore use a body temperature below 32°C as a proxy for death.

Induction of ECM
Mice were infected on day 0 with P. berghei ANKA parasites by intraperitoneal inoculation with
104 parasitised red blood cells (pRBC) from mice of the same strain, diluted in normal saline.
Parasitaemia was determined in Giemsa stained thin blood films. The animals were under daily
supervision for clinical signs of disease, body temperature, and neurological symptoms. ECM
was diagnosed by clinical signs, including ataxia, paralysis (mono, hemi, para or tetraplegia),
deviation of the head, convulsions and coma. The body temperature was measured rectally
once or twice a day (Digital thermometer DM852 with rectal probe, Ellab, Denmark).
The mice were treated with erythropoietin alpha, i.p. (Eprex, Janssen-Cilag, Schaffhausen,
Switzerland) diluted in NaCl (0.9%) in various dosages (1-200U daily i.p. diluted in 0,2 ml
normal saline per injection) and different treatment schemes. Infected control mice were treated
with NaCl (0.9%) only. For immunohistochemistry and rt-PCR analysis of the brain, animals
were killed on day 8. Survival studies were terminated by killing animals with a body
temperature below 32°C. Surviving animals were killed on day 15 after inoculation.

Tissue processing and in situ detection of DNA fragmentation
Tissue processing was performed according to standard procedures as described elsewhere
(273), embedded in paraffin, and cut in serial 5μm thick sections. Terminal deoxynucleotidyl
transferase (TdT)-mediated deoxyuridine triphosphate (dUTP)-digoxigenin nick end labelling
(TUNEL) staining was performed according to the manufacturers’ protocol using the TdT-
FragEL DNA Fragmentation Detection Kit (Oncogene Research Products, Cambridge, UK;
code Cat# QIA 33). DAB was used as chromogen, and the sections were counterstained with
methyl green.

Visualisation and counts
Sections were examined on an Imager.Z1 microscope with an AxioCam MRc5 camera (Carl
Zeiss, Germany). Images were processed using Adobe Photoshop CS (Adobe, USA). In
addition to morphological analysis, cellular counting was carried out in a blinded and
randomised manner by a single investigator. TUNEL+ cells were counted in sagittal sections
from close to the midline of the brain, using one section from each hemisphere.

Measuring of packed cell volume (PCV)
Tail blood or blood from the retro orbital venous plexus was sampled in EDTA-coated capillary
tubes (20µl, Bie and Berntsen, Denmark) and analysed using an automated analyser (KX-21N,
Sysmex, USA).

Gene expression
For quantitative PCR, separate groups of mice (n=27) were either infected as described above
and treated with rhEpo (100U i.p. on day 1,4 and 7 post inoculation), vehicle-treated or left
uninfected. Mice were trans-cardially perfused with NaCl (0.9%) with 0.3% heparin (15,000
IU/L) for 30 – 60 seconds. Brains were removed, parted into right and left cerebrum and
cerebellum and immediately frozen in liquid nitrogen.
RNA was isolated from homogenised tissue using a standard RNeasy lipid tissue kit (Qiagen,
USA) according to the manufacturers’ protocol followed by dilution first in Qiazol lysis reagent
(Qiagen, USA) and then in chloroform 99.9% (Sigma, Denmark) to separate phases.
Quantification of RNA content was done using the Rediplate Ribogreen kit (Invitrogen, USA),
and cDNA was synthesised using the QuantiTect Reveverse Transcription Kit (Qiagen, USA).
All real-time PCR reactions were carried out on a MX-3000p (Stratagene, USA) and run for 50
cycles using the QuantiTect Probe PCR Kit (Qiagen, USA) according to the manufacturers’
protocol and the following protein specific kits: β-actin: cat. no. 241014, caspase 3: cat. no.
241124, no. 241054, caspase 1: cat. no. 241151, TNF-α: cat. no. 241034 IFN-γ: cat. no.
241036, IL-1β: cat. and LT-α: cat no. 241174. Cycle-threshold (Ct) values were defined as the
PCR cycle in which the amplified product was first detected. Gene expression levels were
assessed by way of the comparative threshold method, using expression of the β-actin gene as

Statistical analysis
Results for multiple groups were evaluated by ANOVA using the Holm-Sidak method for
multiple comparisons. Student’s-t test was used when comparing two groups and by log-rank

statistic for the survival curves with pairwise comparison using the Holm-Sidak method. For rt-
PCR data, student’s-t-test was applied on Ct-values. Cox regression for survival analysis was
performed using survival as dependent variable and treatment, parasite level on day 8 and PCV
on day 8 as covariates. The heights of the columns represent mean values. Raw data were
processed by log- or in-transformation to improve the homogeneity of variances, where
necessary. A difference was considered significant for P-values <0.05. Box plot graphs show
the percentiles and the median. Cox regression analysis for survival was performed using the
SPSS 11.5 software package (SPSS Inc., USA). Data were otherwise analysed using the
SigmaPlot 9.01 software package (Systat Software, Inc, USA).

10.4 Results
Treatment with rhEpo resulted in increased survival of mice with cerebral malaria in a dose-
dependent manner. The highest survival increase was achieved when mice were treated from
day 4 to day 7 after one daily inoculation with either 200U, 100U or 50U (Figure 1). 56%, 48%
and 45%, respectively, survived until day 14, compared to the saline-treated group in which no
animal survived longer than until day 11 (p <0.001 for the two former groups and p= 0.001 for
the latter group, all vs. saline-treated controls). No significant difference was seen between the
three groups. A lower dose of 25U rhEpo given daily from day 4-7 resulted in a 30% survival on
day 14 (p= 0.001 vs. saline-treated controls), while daily inoculation with 1U and 10U rhEpo did
not increase the survival rate significantly.
RhEpo treatment on day 4-7 increased packed cell volume (PCV) levels measured on day 8
(before killing), compared to saline-treated controls: The lowest dose of 1U/day of rhEpo
increased the PCV from 44.2% (saline-treated, infected controls) to 51.0% and 10U/day
resulted in a PCV level of 58.4%, while higher daily doses of 25-200U did not further increase
PCV levels (Figure 2).
The effect of the treatment on the time and duration of rhEpo administration: 100U rhEpo daily
was able to increase survival only when administered on day 4-7 (as described above). Early
treatment on day 1-4 and late treatment on day 7-10 did not increase survival. Surprisingly, also
animals treated for a longer period from day 1-7 with 100U daily did not show increased
survival. Those animals reached a mean packed cell volume (PCV) of 73% on day 8. In order to
eliminate a possible negative effect of the high PCV, a group of animals likewise treated on day
1-7 with 100U rhEpo daily were bled for about 10% of their total blood volume on day 5,
resulting in a mean PCV level of 58.6%, which is similar to animals that received treatment only
on day 4-7. Despite that, survival was not increased significantly (Figure 3).

Mean parasitaemia levels on day 8 ranged from 6.3 – 12.6% pRBC (Figure 4). The difference
was significant only between animals that had received 25U/day on day 4-7 and saline-treated
controls (P<0.05). The differences between all other groups were not significant. Cox
regression analysis to explore the effects of different variables on survival and testing for the
covariates treatment, parasitaemia on day 8 and PCV on day 8 showed that only treatment was
correlated with increased survival (P<0.001), while parasitaemia and PCV were not (P=0.66
and P=0.74, respectively).

Body temperature

C57BL/6 mice infected with PbA show a characteristic drop in body temperature when entering
the terminal phase of disease. Data obtained in a typical experiment are shown in figure D. The
body temperature in the vehicle-treated group dropped suddenly, corresponding to the onset of
severe clinical disease (coma; paraplegia). The terminal rise of body temperature in two of the
animals coincided with convulsions. Surviving animals in the group treated with rhEpo 200U/day
on day 4-7 experienced a similar drop in body temperature, but they recovered and reached a
stable temperature at the end of the second week (Figure 5). Those animals would die later
from anaemia and hyperparasitaemia (data not shown). The development in body temperature
was similar in all groups of mice treated with 50-200U/day on day 4-7 (data not shown).

Gene expression

Gene expression in infected mice was elevated for IL-1β, TNFα, INF-γ (P<0.001) and Caspase
1 (P<0.05), but not for Caspase 3 and LT-α, compared to uninfected controls. Treatment with
rhEpo was able to decrease this elevation significantly for IL-1β and TNFα (P<0.001) and for
IFN-γ (P = 0.001). The treatment does not alter the expression of LT-α or Caspases 1 and 3
(Figure 6).

Apoptotic neurons

We had previously shown that neuronal apoptosis detected by TUNEL-staining is a feature in
mice terminal ill with ECM (273). Comparing the number of apoptotic neurons in the brain of
rhEpo-treated mice and vehicle-treated controls with ECM, we found a significant decrease in
the rhEpo-treated group (Figure 7). Furthermore, TUNEL+ showed a tendency to cluster (Figure
F). RhEpo treatment reduced the mean number of TUNEL+ neurons counted in sagittal
sections from both brain hemispheres from 197.1 (C.I. of the mean: 111.7) in vehicle-treated
animals to 53.1 (C.I. of the mean: 21.7; P=0.003).

10.5 Discussion
Our data demonstrate a protective effect of rhEpo in ECM in mice with a clear dose and time
dependency: Animals treated with 50U to 200U of rhEpo daily survive in about half of the cases.
Treatment with a low dose of 1 or 10 IU daily for the same period of time did not increase
survival significantly, while a dose of 25 IU daily increases survival by about 25%. The dose
dependency was seen in mice treated on day 4-7, while early (day 1-4) and late (day 7-10)
treatments were unsuccessful. Surprisingly, treatment from day 1-7 failed to increase survival.
We attributed this failure in the first place to the very high packed cell volume (PCV) observed in
this group (mean: 73% on day 8), as a side effect to the treatment with rhEpo, which we
suspected to outweigh a possible positive effect of the treatment. We therefore repeated the
experiment, this time bleeding the mice for 150µl of their blood (about 10% of their total blood
volume) on day 5 to prevent excessively high PCV values. This intervention lowered the PCV
on day 8 to mean levels equal to those seen in groups responding to treatment, but not with
increased survival. A blood loss of 10% of the total volume is substantial and reduces the
number of RBCs, as well as parasites, thrombocytes and leukocytes, while removing plasma
and its contents. However, none of these alterations offer an obvious explanation for the failure
of the treatment from day 1-7. An early harmful effect of the treatment cannot be ruled out, in
line with a study showing that pre-treatment with Epo of C57BL/6 infected with P. chabaudi can
induce 100% mortality in mice with an otherwise self-limiting course of the disease (274).
The histo-pathological correlate of the neurological dysfunction and sequelae in CM is not
known. Axonal injury has been proposed to play a role (156), and in a previous work we
described apoptotic cell death of neurons in ECM in a clustered manner (273). A recent study
by Kaiser et al. described this pattern only in rhEpo-treated mice, but not in untreated animals
with ECM (136). In our hands, both rhEpo-treated mice and untreated mice with ECM develop
neuronal apoptosis in this typical, clustered manner, while the numbers of TUNEL+ cells as well
as the number of clusters in treated animals are reduced (Figure 7). As TUNEL+ neurons are a
feature of the very late stage of ECM (273), a possible explanation for this discrepancy between
our results and those of Kaiser et al. could therefore be found in the different killing points of
time due to different survival times of treated and untreated animals in their study.
Systemic administration of rhEpo has been shown to prevent apoptotic cell death in an animal
model of stroke (184), an effect which is likely to be the basis of the reduced infarct size and
improved clinical outcome in human stroke patients treated with high-dose rhEpo (212).
Accordingly, the present study demonstrated that rhEpo treatment, in addition to increasing
survival in mice with ECM, prevents neuronal apoptosis in the CNS. As the Epo-R has been
detected on neurons (177), the neuroprotective effect could be due to a direct effect of rhEpo.
Our data also showed a reduced inflammatory response in the CNS of rhEpo-treated mice, as

indicated by reduced gene expression of proinflammatory cytokines in the brain tissue;
therefore, an ameliorated inflammatory response in the brain could contribute to
neuroprotection. Further studies are required to determine the sequence of events.
To identify possible pathways leading to apoptosis in ECM, we studied the expression of
caspases 3 and 8 in the brain, both involved in signalling pathways leading to apoptosis (255).
We have previously shown that immunoreactivity for cleaved (i.e. activated) caspase 3 is
increased in the brain of mice with ECM (273), without a topographical association with
TUNEL+ neurons. The gene expression of caspases 3 and 8 was not increased in ECM and
was independent of rhEpo treatment, indicating that they are possibly not involved in the
induction of apoptosis in ECM. However, the induction of apoptosis in ECM might as well have
been induced by caspase-independent pathways (reviewed in (132;255)). As caspases 3 and 8
are activated by enzymatic cleavage, increased gene expression might not be needed.
The expression of IL-1β-, TNFα- and INF-γ–mRNA in the brain increased during ECM, and this
increase was reduced by rhEpo treatment (Figure E). The data for TNFα- and INF-γ are in line
with previous findings (136), which, combined with our findings for IL-1β, clearly indicate that
rhEpo treatment can ameliorate the inflammatory process in ECM. However, our findings for
LT-α are surprising: Engwerda et al. had pointed to the importance of LT-α, as C57BL/6 mice
deficient for LT-α are resistant to CM (72). In an elegant study they showed that brain cells are
crucial producers of LT-α and proposed endothelial cells, astrocytes or microglia as the possible
source (72), while Rae et al. reported a 3-4-fold up-regulation of LT-α mRNA in the brain of
mice with CM (275). Nevertheless, in our hands, LT-α mRNA levels were not increased in ECM
compared to uninfected controls. Adding these findings to the findings of Engwerda et al., LT-α
appears to be necessary for the patho-mechanism of ECM, but is probably not a factor
triggering cerebral pathology. Treatment with rhEpo had no significant influence on the levels of
LT-α expression (P=0.08), but the power of the test was too low to reject the possibility
The means by which rhEpo treatment may influence the inflammatory response in the CNS
remain unclear. It is unlikely that increased levels of PCV are essential, if involved at all, as the
induction of erythopoiesis in the different treatment groups did not correlate with increased
survival: Even the lowest dose, 1U daily, had significant erythropoietic effect, and 10U daily
resulted in a maximal induction, not further amplified by higher doses (Figure 1 and 3). By
contrast, a dose of 25U/day rhEpo was necessary to increase survival, with maximal effect
when the mice were treated with 50U to 200U. The results of the statistical modelling with the
Cox regression model clearly support this interpretation, as it detects treatment, but neither
PCV nor parasitaemia as significantly associated with increased survival. The protective effect
of rhEpo in murine CM appears therefore to be independent from its haematopoietic effect.

Our data on body temperature indicate that the effect of the treatment was via reduced severity
as well as delayed course of disease: Surviving animals in the treatment group showed a later,
but equal, drop in body temperature as did vehicle-treated mice before returning to values close
to normal (Figure 5). RhEpo treatment in ECM therefore seems to work both before and during
severe disease. This is of importance because the main target group for a proposed adjunctive
treatment in CM is patients already having developed signs of severe disease. Marsh et al.
have previously reported a distinct age distribution of CM and severe malaria anaemia (SMA)
as regards children being hospitalised in Kilifi, Kenia (17). Most cases of SMA appear within the
first two years of age, while the peak of CM is in the age group 3-4 years. Our current findings
of the neuro-protective effect of Epo and the fact that endogenous Epo levels are high in
children with SMA lead us to propose that these increased levels of endogenous Epo in very
young children could contribute to this apparent dichotomy in that they offer protection against

10.6 Conclusions
RhEpo increased survival in ECM in a dose-dependent manner, reduced the inflammatory
response in the brain and protected functional brain tissue by reducing neuronal apoptosis. It
has therefore the potential to become a candidate for adjunctive treatment of CM in humans, in
particular when seen in the light of its excellent safety profile: For decades, rhEpo has been
used for long-term therapy of chronic renal failure. Given in weekly doses of 50-150U/kg, it has
been shown to be safe for extended periods with only rare adverse effects (for review see
(214)), and experience gained from studies of pre-term infants shows that doses of 2100U/week
are safe in very young children (219-221). Furthermore, rhEpo given in a neuroprotective dose
of 33,000IU to stroke patients recently proved to be safe and well-tolerated (212). Peak plasma
concentrations of rhEpo in these patients as well as in healthy volunteers receiving 2400 IU/kg
rhEpo did not exceed levels of endogenous Epo that can be measured in severely anaemic
patients with normal kidney function (168;212;223). Carefully controlled studies are now
required to prove the concept in humans, including on the timing and the duration of Epo

10.7 Acknowledgments
Thanks to Ib Bygbjerg and Lars Hviid for having critically reviewed the manuscript. The
excellent technical assistance offered by Grethe Gomme, and the assistance with statistical

analysis rendered by the Department of Biostatistics, University of Copenhagen are gratefully

10.8 Figures

Figure 1 – Dose-dependent increase in survival in recombinant human Erythropoietin (rhEpo)-treated mice with
cerebral malaria.
Dose-dependent increase in survival in recombinant human Erythropoietin (rhEpo)-treated mice with CM:
Cumulative survival analysis of mice with ECM treated with rhEpo from day 4-7 in different doses from 1-200U
daily versus vehicle-treated controls. Mice treated with a high dose (50-200U daily, red lines) show increased
survival by the end of the second week, 44.4%, 42.9% and 55.6%, respectively, while all saline-treated controls
(black dotted line) survived no longer than until day 11 (p<0.001). Mice treated with 25U daily survived in 30%
(p<0.001 vs. vehicle). Survival of mice treated with a low dose (1-10U daily) did not show a statistically significant
increase in survival. Statistical test: log-rank statistics for the survival curves with pairwise comparison using the
Holm-Sidak method.

Figure 2 - Packed cell volume in recombinant human Erythropoietin (rhEpo)-treated and untreated mice with
cerebral malaria.
Packed cell volume (PCV) measured on day 8 in animals treated on day 4-7 or day 1-7, respectively, and vehicle-
treated controls: Treatment with recombinant human Erythropoietin (rhEpo) increases PCV levels in mice with
cerebral malaria. 1U daily from day 4-7 increases the PCV to 51%, while 10U or more daily given on day 4-7 leads
to mean PCV levels between 58.4% and 59.7%. PCV levels in the group treated on day 1-7 with 100U daily reached
very high values of 73.0%, unless these mice were bled for 10% of their total blood volume on day 5. This
intervention reduced the mean PCV level to 58.6% (right plot). P<0.001. Statistical test: One-way ANOVA. The
ends of the boxes define the 25th and 75th percentiles, with a line at the median and error bars defining the 10th and
90th percentiles.

Figure 3 - Time-dependent increase in survival in recombinant human Erythropoietin (rhEpo)-treated mice with
cerebral malaria.
Time-dependent increase in survival in mice with CM treated with recombinant human Erythropoietin (rhEpo): The
cumulative survival analysis of mice ECM treated with 100U of rhEpo daily shows increased survival at the end of
the second week only in mice treated from day 4-7 (42.9%, p<0.001 vs. vehicle). The other treatment schemes used
(day 1-4; day7-10; day1-7) and mice that received treatment from day 1-7 and were bled for 10% of their total blood
volume to prevent excessively high levels of packed cell volume did not show a statistically significant increase in
survival. Statistical test: log-rank statistic for the survival curves with pairwise comparison using the Holm-Sidak

Figure 4 - Mean parasitaemia levels on day 8.
Mean parasiteaemia levels on day 8 as a percentage of parstitezed red blood cells in giemsa-stained smears: The
mean values for the different groups range from 6.3 – 12.6%. Only the treatment group that received recombinant
human Erythropoietin (rhEpo) 25U/day from day 4-7 (**) was significantly different from the control group that
had received normal saline (NaCl) (P<0.05). All other differences between the groups were not significant.
Statistical test: One way ANOVA. Data are presented as mean values +/- standard error.

Figure 5 - Body temperature of mice with cerebral malaria.
Vehicle-treated mice (left graph) show a characteristic drop of body temperature on day 9. Surviving mice in the
group treated with recombinant human Erythropoietin (rhEpo) 200U daily on day 4-7 show a similar drop on day 9-
10, but recover afterwards (right graph). The data shown are representative for repeated experiments. The data
presented derive from one single experiment. Data for the treatment group are representative for mice that received
rhEpo 50-200U daily on day 4-5.

Figure 6 - Gene expression in the brain of recombinant human Erythropoietin (rhEpo)-treated and untreated mice
with cerebral malaria.
Gene expression in the brain of recombinant human Erythropoietin (rhEpo)-treated and untreated mice with cerebral
malaria on day 8 relative to uninfected control mice (black columns). Increased gene expression in infected mice
was seen for IL-1β, TNF, INF-γ (P<0.001) and Caspase 1 (P<0.05), but not for LT-α, and Caspase 3. Gene
expression for the IL-β, TNF and INF-γ was significantly reduced in rhEpo-treated mice (P<0.001). The treatment
does not alter the expression of LT-α and Caspases 1 and 3. Statistical test: Two-way ANOVA. Values are
presented as mean values relative to uninfected controls. Error bars: Standard deviation (** = P<0.001).

Figure 7 - Treatment with recombinant human Erythropoietin (rhEpo) in mice with cerebral malaria reduces
neuronal apoptosis in the brain.
Apoptotic neurons in the brain of mice with ECM. The micrograph shows TUNEL (Terminal deoxynucleotidyl
transferase (TdT)-mediated deoxyuridine triphosphate (dUTP)-digoxigenin nick end labelling) positive nuclei of
cells in the cortex of a terminally ill mouse with ECM. We have earlier shown that the round-shaped nuclei belong
to neurons (273). The mean number of TUNEL+ neurons in the brain of recombinant human Erythropoietin
(rhEpo)-treated mice with ECM is significantly lower than in vehicle-treated controls (P=0.006). Countings from
TUNEL-stained sagittal sections of both hemispheres. Scale bar: 50µm. Statistical analysis: Students t-test. The
ends of the boxes define the 25th and 75th percentiles, with a line at the median and error bars defining the 5th and
95th percentiles.

11 Unpublished results

11.1 Treatment with a newly constructed erythropoietin fragment could not increase
             survival in mice with cerebral malaria

                              Survival Analysis

                                                                    EP3 16 1-10
                                                                 EP3 16 4-7
                                                                     EP3 32 1-10

           0,8                                                   NaCl
                                                                 EP3 8 1-10




                 4      6          8           10           12           14


Figure D: Treatment with the erythropoietin fragment “EP3” failed to increase survival in the same manner as
recombinant human erythropoietin.
Kaplan-Meier survival plot for mice infected with Plasmdium berghei ANKA and treated with the newly engineered
peptide “EP3”, a fragment of the erythropoietin protein. The peptide was administered subcutaneously in doses
from 8 – 32 mg/kg in two different schedules (day 1-10 or day 4-7) in a model for cerebral malaria (as described
previously (276)). None of the treatment groups (blue lines) differed significantly from untreated controls (black
dotted line). Statistical test: Log rank statistics for the survival curves with pairwise comparison using the Holm-
Sidak method.

11.2 Cytokines and VEGF in murine CM
                                                                TNF                                                                                                           IL-10

          100                                                                                                      800

                                                          n/s                                                      600






                                                                            Cl                                                             o                                               l              aCl
                              Epo                      - Ep
                                                                       - Na                        aCl                                 - Ep                    - Ep
                          -                      d
                                                                PbA                      cted
                                                                                                                                 PbA                    d
                                                                                                                                                                                   -                   d-N
                                           fecte                                    fe                                                             fecte                                          fecte
                                       unin                                    unin                                                            unin                                           unin



          50                                          P = 0.013
                                                                                                                           80                              P = 0.013






                           - Ep
                                  o                     po                    Cl                    aC l                                        o                        po                   Cl                    a Cl
                                                     d-E               - Na                   d-N                                        - Ep                      d-E                 - Na                   d-N
                    P bA                     fecte              P bA                  fecte                                       Pb A                     fecte                Pb A
                                      unin                                                                                                          unin                                              fecte
                                                                               unin                                                                                                            unin

                                                                                                                   Figure E: Cytokine and VEGF levels in the serum of
                                                     TNF / IL-10 ratio
                                                                                                                   infected/uninfected and erythropoietin-treated/untreated mice
   0,8                                                                                                             on day 8. TNF, IL-10, TNF/IL-10 ratio VEGF and INF-γ are
                                                                                                                   significantly elevated in infected mice compared to uninfected
                                                      n/s                                                          controls (P<0.05). Erythropoietin treatment from day 4 to 7
   0,4                                                                                                             reduced VEGF and INF-γ but not TNF and IL-10 levels
                                                                                                                   significantly (one way ANOVA or one way ANOVA on ranks
                                                                                                                   where necessary). Experimental setup as described in (276).

                                                                                                                   Nine mice per group. Analysis by laser-based fluorescent
                                                                                                                   analytical testing on a Luminex 100 analyser (Luminex, USA)

                       -   Epo
                                             po                        NaC
                                                                          l               aCl                      using Biosource reagents (Biosource Europe, Belgium).
                   PbA                                         PbA
                                                                   -                   d-N
                                     fecte                                        fecte
                                 unin                                         unin

11.3 Treatment with exogenous Metallothionein in mice with CM
                                      TUNEL+ cells

number of TUNEL+ cells



                               NaCl       Epo        MT


Figure F: Apoptotic neurons in the brain of mice with CM. Countings of TUNEL (Terminal deoxynucleotidyl
transferase (TdT)-mediated deoxyuridine triphosphate (dUTP)-digoxigenin nick end labeling) positive nuclei of
cells in the cortex of a terminally ill mouse with CM. We have earlier shown that the round-shaped nuclei belong to
neurons (273). Method as described above (276). The number of TUNEL+ cells in the brains of mice treated with
exogenous metallothionein-II (MT-II; 750μg/kg, twice daily) was not significantly different compared to vehicle-
treated controls (P = 0.4 by one way ANOVA followed by pairwise comparison using the Holm-Sidak method).
One single animal showed an area of the cortex with necrosis, surrounded by an edge containing numerous
TUNEL+ cells. This feature was not seen in neither vehicle nor in recombinant human Erythropoietin (rhEpo)
treated animals. In contrast, the mean number of TUNEL+ neurons in the brains of (rhEpo)-treated mice with CM is
significantly lower than in vehicle-treated controls (P=0.006). Data from vehicle- and rhEpo-treated animals in this
experiment are part of the results shown above (article 2 (276)). Countings from two TUNEL-stained sagittal
sections, one from each hemisphere. Statistical analysis: Students’ t-test. The ends of the boxes define the 25th and
75th percentiles, with a line at the median and error bars defining the 5th and 95th percentiles.

12 Perspectives
The results presented have given rise to new perspectives. The major topics are outlined in the
following chapter.

12.1 Erythropoietin fragments in the treatment of cerebral malaria
The promising results, including our own, as to the role of rhEpo as a neuroprotective agent are
encouraging. Translation of these research findings into therapeutic applications looks
promising and is probably most advanced in the treatment of ischaemic stroke (212;277). As
regards the treatment of stroke patients, which is likely to require multiple doses of rhEpo, it is a
major limitation of the compound that it would trigger undesirable overstimulation of the bone
marrow, raise the hematocrit, and induce a procoagulant state (278). This problem has led to
several attempts to design compounds which include the neuropotective, but exclude the
haematopoietic, properties of rhEpo. However, on the background of bone marrow depression
in CM and the anaemia found in many CM patients, stimulation of haematopoiesis in CM can be
seen more like a positive “side effect”. This reduces the need for new Epo-like compounds in
this respect.
Nevertheless, these Epo derivates can be used to dissect the mode of action of rhEpo in CM. In
a small study we used an Epo fragment design at the Protein Laboratory, Institute of Molecular
Pathology, University of Copenhagen, Panum Institute, Copenhagen, Denmark, a friendly gift of
Professor Elisabeth Bock and Professor Vladimir Berezin) in our mouse model of CM. The
fragment binds only to one binding site of the EpoR for Epo, the low affinity-binding site.
Treatment with the erythropoietin fragment “Ep3” failed to increase survival as seen in rhEpo-
treated mice (Figure D, chapter 11.1). Ep3 had earlier shown neuroprotective properties in vitro
and in vivo. In vivo it had been able to reduce the latency and severity of kainic acid-induced
seizures (Stanislava Pankratova, Elisabeth Bock, personal communication) in a similar way as
previously shown for the treatment with rhEpo (181), but without activating erythropoiesis. In
future work, this and other Epo derivates can be used as a tool to further examine the pathways
by which rhEpo is acting in CM.

12.2 The role of the blood brain barrier in cerebral malaria
Our results clearly showed a protective effect of rhEpo in murine CM, while Epo’s mode of
action in CM is left unexplained. The protective effect on the brain in terms of a reduced number
of lost neurons and a reduced expression of proinflammatory cytokines could be caused by a
direct effect on the brain. We observed that brain endothelial cells are positive for markers of

oxidative stress in the brain of mice with CM and that those cells are highly immunoreactive for
anti-VEGF (273). Measurements of serum levels of VEGF revealed an increase in diseased
animals compared to uninfected ones and – importantly – reduced levels in rhEpo-treated mice
with CM compared to saline-treated animals (Figure E, chapter 11.2). TNF and IL-10 levels and
TNF / IL-10 ratios were at the same time not significantly influenced by the treatment - an
observation which makes a major role of a systemic anti-inflammatory effect of rhEpo in CM
We therefore hypothesise that the BBB is a major point of action for rhEpo in CM. To test this
hypothesis, we plan to examine if rhEpo can reduce impairment of the BBB, followed by a
thorough investigation of changes relating to the BBB targeting at endothelial activation, BBB
integrity, tight junctions, astrocyte activation and pathways that are influenced by the
In addition to these ex vivo studies, in vivo models of the BBB are available and several of
those models have been applied in cerebral malaria research (43;44;55;237). Some of these
models indicate a role of endothelial apoptosis in CM - a feature demonstrated in vivo in murine
CM as well (273). Studying the effect of rhEpo in these models is therefore a matter of priority in
future research.
Also the possible involvement of VEGF in CM pathology calls for further examination, and it
seems worthwhile to draw inspiration from studies on cerebral ischaemia. There, the BBB is
likewise impaired and VEGF up-regulated (reviewed in (137)). Treatment with rhEpo in an
animal model for ischaemic stroke intervened with VEGF pathways as follows: Even though
VEGF expression was up-regulated in rhEpo-treated animals compared to untreated controls,
rhEpo treatment caused a significant reduction of the expression of the corresponding VEGF
receptor (Flk-1) (202). In the same study BBB impairment was reduced by the rhEpo treatment,
and the authors conclude that rhEPO protected against ischemia-induced BBB damage at least
partly by down-regulating the response to VEGF-signalling in the acute phase after stroke
(202). In the human post-mortem brain of CM patients, immunoreactivity for both VEGF and the
VEGF receptor (Flk-1) is increased compared to controls (204). A systematical assessment of
the role of VEGF in CM is highly indicated on the background of these findings, as the VEGF /
VEGF receptor system could play a major role in the effect of rhEpo in CM as well as it could
resemble a possible new target point for interventions in CM.

12.3 Screening for changes in mRNA expression in erythropoietin-treated mice with CM

A broader approach to locate the mechanism by which rhEpo treatment is acting in the brain of
mice with CM is currently being taken: The positive results for increased survival in rhEpo-
treated mice with CM and, in addition, the changes detected for mRNA expression in the brain

led us to perform microarray analysis of brain tissue from rhEpo-treated / untreated and P.
berghei ANKA-infected / uninfected mice. The raw data are currently being analysed.

12.4 Metallothionein in the treatment of cerebral malaria
MT I+II have been shown to have a wide range of protective functions in the CNS (225;228).
The results of our own studies show an increase of MT I+II expression in mice with CM
compared to uninfected controls (273). We therefore attempted to use exogenous MT-II as
treatment in murine CM in the same mouse model as used for the treatment with rhEpo (276).
In this study we used exogenous MT-II 750μg/kg (given twice daily subcutaneously from day 1
after inoculation of parasites). The dose, schedule and mode of application had previously
shown to significantly decrease expression of IL-6 and TNF-α in the CNS and to reduce
apoptotic cell death of neurons and oligodendrocytes during experimental autoimmune
encephalitis – a model for multiple sclerosis in mice. However, MT-II was not able to reduce the
number of apoptotic neurons in murine CM in this set-up (Figure F, chapter 11.3). During our
work with rhEpo in murine CM, we learned that the timing of the treatment is crucial and that
early treatment is not neuroprotective. Accordingly, and fuelled by observations that indicate
possible overlapping mechanisms of multiple sclerosis and cerebral malaria (reviewed in (279)),
future work will further explore the effect of MT in murine CM.

13 References

  (1)   Breman JG. The ears of the hippopotamus: manifestations, determinants, and estimates of the malaria burden. Am J
        Trop Med Hyg 2001 Jan;64(1-2 Suppl):1-11.

  (2)   Snow RW, Guerra CA, Noor AM, Myint HY, Hay SI. The global distribution of clinical episodes of Plasmodium falciparum
        malaria. Nature 2005 Mar 10;434(7030):214-7.

  (3)   Mung'Ala-Odera V, Snow RW, Newton CR. The burden of the neurocognitive impairment associated with Plasmodium
        falciparum malaria in sub-saharan Africa. Am J Trop Med Hyg 2004 Aug;71(2 Suppl):64-70.

  (4)   Murphy SC, Breman JG. Gaps in the childhood malaria burden in Africa: cerebral malaria, neurological sequelae, anemia,
        respiratory distress, hypoglycemia, and complications of pregnancy. Am J Trop Med Hyg 2001 Jan;64(1-2 Suppl):57-67.

  (5)   Reyburn H, Mbatia R, Drakeley C, Bruce J, Carneiro I, Olomi R, et al. Association of transmission intensity and age with
        clinical manifestations and case fatality of severe Plasmodium falciparum malaria. JAMA 2005 Mar 23;293(12):1461-70.

  (6)   Newton CR, Krishna S. Severe falciparum malaria in children: current understanding of pathophysiology and supportive
        treatment. Pharmacol Ther 1998 Jul;79(1):1-53.

  (7)   Boivin MJ, Bangirana P, Byarugaba J, Opoka RO, Idro R, Jurek AM, et al. Cognitive Impairment After Cerebral Malaria in
        Children: A Prospective Study. Pediatrics 2007 Jan 15.

  (8)   Bondi FS. The incidence and outcome of neurological abnormalities in childhood cerebral malaria: a long-term follow-up
        of 62 survivors. Trans R Soc Trop Med Hyg 1992 Jan;86(1):17-9.

  (9)   Carter JA, Neville BG, Newton CR. Neuro-cognitive impairment following acquired central nervous system infections in
        childhood: a systematic review. Brain Res Brain Res Rev 2003 Sep;43(1):57-69.

 (10)   Carter JA, Mung'Ala-Odera V, Neville BG, Murira G, Mturi N, Musumba C, et al. Persistent neurocognitive impairments
        associated with severe falciparum malaria in Kenyan children. J Neurol Neurosurg Psychiatry 2005 Apr;76(4):476-81.

 (11)   Idro R, Carter JA, Fegan G, Neville BG, Newton CR. Risk factors for persisting neurological and cognitive impairments
        following cerebral malaria. Arch Dis Child 2006 Feb;91(2):142-8.

 (12)   Ozen M, Gungor S, Atambay M, Daldal N. Cerebral malaria owing to Plasmodium vivax: case report. Ann Trop Paediatr
        2006 Jun;26(2):141-4.

 (13)   Thapa R, Patra V, Kundu R. Plasmodium vivax Cerebral Malaria. Indian Pediatr 2007 Jun;44(6):433-4.

 (14)   Tongren JE, Zavala F, Roos DS, Riley EM. Malaria vaccines: if at first you don't succeed.. Trends Parasitol 2004

 (15)   Hviid L. Naturally acquired immunity to Plasmodium falciparum malaria in Africa. Acta Trop 2005 Sep;95(3):270-5.

 (16)   Marsh K, Howard RJ. Antigens induced on erythrocytes by P. falciparum: expression of diverse and conserved
        determinants. Science 1986 Jan 10;231(4734):150-3.

 (17)   Marsh K, Snow RW. Host-parasite interaction and morbidity in malaria endemic areas. Philos Trans R Soc Lond B Biol
        Sci 1997 Sep 29;352(1359):1385-94.

 (18)   Kurtzhals JA, Goka BQ, Akanmori BD, Hviid L. The importance of strict patient definition in studies of malaria
        pathogenesis. Trends Parasitol 2001 Jul;17(7):313-4.

 (19)   Idro R, Jenkins NE, Newton CR. Pathogenesis, clinical features, and neurological outcome of cerebral malaria. Lancet
        Neurol 2005 Dec;4(12):827-40.

 (20)   World Health Organization. Management of severe malaria - A praktical handbook. 2 ed. 2006.

(21)   World Health Organization. Severe falciparum malaria. Transactions of the Royal Society of Tropical Medicine and
       Hygiene 2000 Apr;94:S1-S90.

(22)   Molyneux ME, Taylor TE, Wirima JJ, Borgstein A. Clinical features and prognostic indicators in paediatric cerebral
       malaria: a study of 131 comatose Malawian children. Q J Med 1989 May;71(265):441-59.

(23)   Taylor TE, Fu WJ, Carr RA, Whitten RO, Mueller JS, Fosiko NG, et al. Differentiating the pathologies of cerebral malaria
       by postmortem parasite counts. Nat Med 2004 Feb;10(2):143-5.

(24)   Marsh K, Forster D, Waruiru C, Mwangi I, Winstanley M, Marsh V, et al. Indicators of life-threatening malaria in African
       children. N Engl J Med 1995 May 25;332(21):1399-404.

(25)   Miller LH, Baruch DI, Marsh K, Doumbo OK. The pathogenic basis of malaria. Nature 2002 Feb 7;415(6872):673-9.

(26)   Hunt NH, Grau GE. Cytokines: accelerators and brakes in the pathogenesis of cerebral malaria. Trends Immunol 2003

(27)   de Souza JB, Riley EM. Cerebral malaria: the contribution of studies in animal models to our understanding of
       immunopathogenesis. Microbes and Infection 2002 Mar;4(3):291-300.

(28)   Engwerda C, Belnoue E, Gruner AC, Renia L. Experimental models of cerebral malaria. Curr Top Microbiol Immunol

(29)   Lou J, Lucas R, Grau GE. Pathogenesis of cerebral malaria: recent experimental data and possible applications for
       humans. Clin Microbiol Rev 2001 Oct;14(4):810-20, table.

(30)   Neill AL, Hunt NH. Pathology of Fatal and Resolving Plasmodium-Berghei Cerebral Malaria in Mice. Parasitology 1992

(31)   Curfs JH, Schetters TP, Hermsen CC, Jerusalem CR, van Zon AA, Eling WM. Immunological aspects of cerebral lesions
       in murine malaria. Clin Exp Immunol 1989 Jan;75(1):136-40.

(32)   Hearn J, Rayment N, Landon DN, Katz DR, de Souza JB. Immunopathology of cerebral malaria: Morphological evidence
       of parasite sequestration in murine brain microvasculature. Infect Immun 2000 Sep;68(9):5364-76.

(33)   Polder T, Jerusalem C, Eling W. Topographical distribution of the cerebral lesions in mice infected with Plasmodium
       berghei. Tropenmed Parasitol 1983 Dec;34(4):235-43.

(34)   Franke-Fayard B, Janse CJ, Cunha-Rodrigues M, Ramesar J, Buscher P, Que I, et al. Murine malaria parasite
       sequestration: CD36 is the major receptor, but cerebral pathology is unlinked to sequestration. Proc Natl Acad Sci U S A
       2005 Aug 9;102(32):11468-73.

(35)   Amani V, Boubou MI, Pied S, Marussig M, Walliker D, Mazier D, et al. Cloned lines of Plasmodium berghei ANKA differ in
       their abilities to induce experimental cerebral malaria. Infect Immun 1998 Sep;66(9):4093-9.

(36)   Roberts DJ, Craig AG, Berendt AR, Pinches R, Nash G, Marsh K, et al. Rapid switching to multiple antigenic and
       adhesive phenotypes in malaria. Nature 1992 Jun 25;357(6380):689-92.

(37)   Combes V, de Souza JB, Renia L, Hunt NH, Grau GE. Cerebral malaria: Which parasite? Which model? Drug Discovery
       Today: Disease Models 2005;2(2):141-7.

(38)   Brown H, Turner G, Rogerson S, Tembo M, Mwenechanya J, Molyneux M, et al. Cytokine expression in the brain in
       human cerebral malaria. Journal of Infectious Diseases 1999 Nov;180(5):1742-6.

(39)   Medana IM, Chaudhri G, Chan-Ling T, Hunt NH. Central nervous system in cerebral malaria: 'Innocent bystander' or
       active participant in the induction of immunopathology? Immunol Cell Biol 2001 Apr;79(2):101-20.

(40)   Berendt AR, Tumer GD, Newbold CI. Cerebral malaria: The sequestration hypothesis. Parasitol Today 1994;10(10):412-

(41)   Clark IA, Rockett KA. The cytokine theory of human cerebral malaria. Parasitol Today 1994;10(10):410-2.

(42)   Grau GE, de KS. Cerebral malaria: mediators, mechanical obstruction or more? Parasitol Today 1994 Oct;10(10):408-9.

(43)   Pino P, Vouldoukis I, Kolb JP, Mahmoudi N, sportes-Livage I, Bricaire F, et al. Plasmodium falciparum--infected
       erythrocyte adhesion induces caspase activation and apoptosis in human endothelial cells. J Infect Dis 2003 Apr

(44)   Hemmer CJ, Lehr HA, Westphal K, Unverricht M, Kratzius M, Reisinger EC. Plasmodium falciparum Malaria: reduction of
       endothelial cell apoptosis in vitro. Infect Immun 2005 Mar;73(3):1764-70.

(45)   Belnoue E, Kayibanda M, Vigario AM, Deschemin JC, Rooijen Nv, Viguier M, et al. On the Pathogenic Role of Brain-
       Sequestered {alpha}{beta} CD8+ T Cells in Experimental Cerebral Malaria. The Journal of Immunology 2002 Dec

(46)   Nitcheu J, Bonduelle O, Combadiere C, Tefit M, Seilhean D, Mazier D, et al. Perforin-dependent brain-infiltrating cytotoxic
       CD8+ T lymphocytes mediate experimental cerebral malaria pathogenesis. J Immunol 2003 Feb 15;170(4):2221-8.

(47)   Potter S, Chan-Ling T, Ball HJ, Mansour H, Mitchell A, Maluish L, et al. Perforin mediated apoptosis of cerebral
       microvascular endothelial cells during experimental cerebral malaria. Int J Parasitol 2006 Jan 19.

(48)   Baruch DI. Adhesive receptors on malaria-parasitized red cells. Baillieres Best Pract Res Clin Haematol 1999

(49)   Seydel KB, Milner DA, Jr., Kamiza SB, Molyneux ME, Taylor TE. The distribution and intensity of parasite sequestration
       in comatose malawian children. J Infect Dis 2006 Jul 15;194(2):208-5.

(50)   MacPherson GG, Warrell MJ, White NJ, Looareesuwan S, Warrell DA. Human cerebral malaria. A quantitative
       ultrastructural analysis of parasitized erythrocyte sequestration. Am J Pathol 1985 Jun;119(3):385-401.

(51)   Turner GD, Morrison H, Jones M, Davis TM, Looareesuwan S, Buley ID, et al. An immunohistochemical study of the
       pathology of fatal malaria. Evidence for widespread endothelial activation and a potential role for intercellular adhesion
       molecule-1 in cerebral sequestration. Am J Pathol 1994 Nov;145(5):1057-69.

(52)   Clark IA, Cowden WB. Why is the pathology of falciparum worse than that of vivax malaria? Parasitol Today 1999

(53)   Dondorp AM, Desakorn V, Pongtavornpinyo W, Sahassananda D, Silamut K, Chotivanich K, et al. Estimation of the total
       parasite biomass in acute falciparum malaria from plasma PfHRP2. PLoS Med 2005 Aug;2(8):e204.

(54)   Ho M, White NJ. Molecular mechanisms of cytoadherence in malaria. Am J Physiol 1999 Jun;276(6 Pt 1):C1231-C1242.

(55)   Tripathi AK, Sullivan DJ, Stins MF. Plasmodium falciparum-Infected Erythrocytes Decrease the Integrity of Human Blood-
       Brain Barrier Endothelial Cell Monolayers. J Infect Dis 2007 Apr 1;195(7):942-50.

(56)   Monso-Hinard C, Lou JN, Behr C, Juillard P, Grau GE. Expression of major histocompatibility complex antigens on
       mouse brain microvascular endothelial cells in relation to susceptibility to cerebral malaria. Immunology 1997

(57)   Turner GD, Ly VC, Nguyen TH, Tran TH, Nguyen HP, Bethell D, et al. Systemic endothelial activation occurs in both mild
       and severe malaria. Correlating dermal microvascular endothelial cell phenotype and soluble cell adhesion molecules
       with disease severity. Am J Pathol 1998 Jun;152(6):1477-87.

(58)   Perlmann P, Troye-Blomberg M. Malaria and the immune system in humans. Malaria Immunology, 2Nd Edition

(59)   Grau GE, Taylor TE, Molyneux ME, Wirima JJ, Vassalli P, Hommel M, et al. Tumor necrosis factor and disease severity
       in children with falciparum malaria. N Engl J Med 1989 Jun 15;320(24):1586-91.

(60)   Kwiatkowski D, Hill AV, Sambou I, Twumasi P, Castracane J, Manogue KR, et al. TNF concentration in fatal cerebral,
       non-fatal cerebral, and uncomplicated Plasmodium falciparum malaria. Lancet 1990 Nov 17;336(8725):1201-4.

(61)   Kurtzhals J. Low plasma concentrations of interleukin 10 in severe malarial anaemia compared with cerebral and
       uncomplicated malaria. Lancet 1999 Mar 6;353(9155):848.

(62)   Molyneux ME, Taylor TE, Wirima JJ, Grau GE. Tumour necrosis factor, interleukin-6, and malaria. Lancet 1991 May

(63)   Brown H, Turner G, Rogerson S, Tembo M, Mwenechanya J, Molyneux M, et al. Cytokine expression in the brain in
       human cerebral malaria. Journal of Infectious Diseases 1999 Nov;180(5):1742-6.

(64)   Brewster DR, Kwiatkowski D, White NJ. Neurological sequelae of cerebral malaria in children. Lancet 1990 Oct

(65)   McGuire W, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D. Variation in the TNF-alpha promoter region associated
       with susceptibility to cerebral malaria. Nature 1994 Oct 6;371(6497):508-10.

(66)   Kern P, Hemmer CJ, Gallati H, Neifer S, Kremsner P, Dietrich M, et al. Soluble tumor necrosis factor receptors correlate
       with parasitemia and disease severity in human malaria. J Infect Dis 1992 Oct;166(4):930-4.

(67)   Molyneux ME, Engelmann H, Taylor TE, Wirima JJ, Aderka D, Wallach D, et al. Circulating plasma receptors for tumour
       necrosis factor in Malawian children with severe falciparum malaria. Cytokine 1993 Nov;5(6):604-9.

(68)   Lucas R, Juillard P, Decoster E, Redard M, Burger D, Donati Y, et al. Crucial role of tumor necrosis factor (TNF) receptor
       2 and membrane-bound TNF in experimental cerebral malaria. Eur J Immunol 1997 Jul;27(7):1719-25.

(69)   Gimenez F, Barraud dL, Fernandez C, Pino P, Mazier D. Tumor necrosis factor alpha in the pathogenesis of cerebral
       malaria. Cell Mol Life Sci 2003 Aug;60(8):1623-35.

(70)   Stoelcker B, Hehlgans T, Weigl K, Bluethmann H, Grau GE, Mannel DN. Requirement for tumor necrosis factor receptor
       2 expression on vascular cells to induce experimental cerebral malaria. Infect Immun 2002 Oct;70(10):5857-9.

(71)   Artavanis-Tsakonas K, Tongren JE, Riley EM. The war between the malaria parasite and the immune system: immunity,
       immunoregulation and immunopathology. Clinical and Experimental Immunology 2003 Aug;133(2):145-52.

(72)   Engwerda CR, Mynott TL, Sawhney S, de Souza JB, Bickle QD, Kaye PM. Locally up-regulated lymphotoxin alpha, not
       systemic tumor necrosis factor alpha, is the principle mediator of murine cerebral malaria. Journal of Experimental
       Medicine 2002 May 20;195(10):1371-7.

(73)   Rudin W, Favre N, Bordmann G, Ryffel B. Interferon-gamma is essential for the development of cerebral malaria. Eur J
       Immunol 1997 Apr;27(4):810-5.

(74)   Yanez DM, Manning DD, Cooley AJ, Weidanz WP, van der Heyde HC. Participation of lymphocyte subpopulations in the
       pathogenesis of experimental murine cerebral malaria. J Immunol 1996 Aug 15;157(4):1620-4.

(75)   Amani V, Vigario AM, Belnoue E, Marussig M, Fonseca L, Mazier D, et al. Involvement of IFN-gamma receptor-
       medicated signaling in pathology and anti-malarial immunity induced by Plasmodium berghei infection. Eur J Immunol
       2000 Jun;30(6):1646-55.

(76)   Ho M, Sexton MM, Tongtawe P, Looareesuwan S, Suntharasamai P, Webster HK. Interleukin-10 inhibits tumor necrosis
       factor production but not antigen-specific lymphoproliferation in acute Plasmodium falciparum malaria. J Infect Dis 1995

(77)   Ringwald P, Peyron F, Vuillez JP, Touze JE, Le BJ, Deloron P. Levels of cytokines in plasma during Plasmodium
       falciparum malaria attacks. J Clin Microbiol 1991 Sep;29(9):2076-8.

(78)   Day NP, Hien TT, Schollaardt T, Loc PP, Chuong LV, Chau TT, et al. The prognostic and pathophysiologic role of pro-
       and antiinflammatory cytokines in severe malaria. J Infect Dis 1999 Oct;180(4):1288-97.

(79)   Weiser S, Miu J, Ball HJ, Hunt NH. Interferon-gamma synergises with tumour necrosis factor and lymphotoxin-alpha to
       enhance the mRNA and protein expression of adhesion molecules in mouse brain endothelial cells. Cytokine 2007

(80)   Kumaratilake LM, Ferrante A. T-cell cytokines in malaria: their role in the regulation of neutrophil- and macrophage-
       mediated killing of Plasmodium falciparum asexual blood forms. Res Immunol 1994 Jul;145(6):423-9.

(81)   Su Z, Stevenson MM. Central role of endogenous gamma interferon in protective immunity against blood-stage
       Plasmodium chabaudi AS infection. Infect Immun 2000 Aug;68(8):4399-406.

(82)   Gosselin D, Rivest S. Role of IL-1 and TNF in the brain: twenty years of progress on a Dr. Jekyll/Mr. Hyde duality of the
       innate immune system. Brain Behav Immun 2007 Mar;21(3):281-9.

(83)   Rockett KA, Awburn MM, Rockett EJ, Clark IA. Tumor necrosis factor and interleukin-1 synergy in the context of malaria
       pathology. Am J Trop Med Hyg 1994 Jun;50(6):735-42.

(84)   Dinarello CA. Biologic basis for interleukin-1 in disease. Blood 1996 Mar 15;87(6):2095-147.

(85)   Akanmori BD, Kurtzhals JA, Goka BQ, Adabayeri V, Ofori MF, Nkrumah FK, et al. Distinct patterns of cytokine regulation
       in discrete clinical forms of Plasmodium falciparum malaria. Eur Cytokine Netw 2000 Mar;11(1):113-8.

(86)   Moore KW, de Waal MR, Coffman RL, O'Garra A. Interleukin-10 and the interleukin-10 receptor. Annu Rev Immunol

(87)   Kossodo S, Monso C, Juillard P, Velu T, Goldman M, Grau GE. Interleukin-10 modulates susceptibility in experimental
       cerebral malaria. Immunology 1997 Aug;91(4):536-40.

 (88)   May J, Lell B, Luty AJ, Meyer CG, Kremsner PG. Plasma interleukin-10:Tumor necrosis factor (TNF)-alpha ratio is
        associated with TNF promoter variants and predicts malarial complications. J Infect Dis 2000 Nov;182(5):1570-3.

 (89)   Gazzinelli RT, Oswald IP, James SL, Sher A. IL-10 inhibits parasite killing and nitrogen oxide production by IFN-gamma-
        activated macrophages. J Immunol 1992 Mar 15;148(6):1792-6.

 (90)   Anstey NM, Weinberg JB, Hassanali M, Mwaikambo ED, Manyenga D, Misukonis MA, et al. Nitric oxide in Tanzanian
        children with malaria: Inverse relationship between malaria severity and nitric oxide production nitric oxide synthase type
        2 expression. Journal of Experimental Medicine 1996 Aug 1;184(2):557-67.

 (91)   Medana IM, Turner GD. Human cerebral malaria and the blood-brain barrier. Int J Parasitol 2006 Mar 10.

 (92)   Armah H, Wired EK, Dodoo AK, Adjei AA, Tettey Y, Gyasi R. Cytokines and adhesion molecules expression in the brain
        in human cerebral malaria. Int J Environ Res Public Health 2005 Apr;2(1):123-31.

 (93)   Wassmer SC, de Souza JB, Frere C, Candal FJ, Juhan-Vague I, Grau GE. TGF-{beta}1 Released from Activated
        Platelets Can Induce TNF-Stimulated Human Brain Endothelium Apoptosis: A New Mechanism for Microvascular Lesion
        during Cerebral Malaria. J Immunol 2006 Jan 15;176(2):1180-4.

 (94)   Badwey JA, Karnovsky ML. Active oxygen species and the functions of phagocytic leukocytes. Annu Rev Biochem

 (95)   Dawson VL, Dawson TM. Free radicals and neuronal cell death. Cell Death Differ 1996 Jan;3(1):71-8.

 (96)   Burlacu A, Jinga V, Gafencu AV, Simionescu M. Severity of oxidative stress generates different mechanisms of
        endothelial cell death. Cell Tissue Res 2001 Dec;306(3):409-16.

 (97)   Wei EP, Christman CW, Kontos HA, Povlishock JT. Effects of oxygen radicals on cerebral arterioles. Am J Physiol 1985
        Feb;248(2 Pt 2):H157-H162.

 (98)   Wozencraft AO, Croft SL, Sayers G. Oxygen radical release by adherent cell populations during the initial stages of a
        lethal rodent malarial infection. Immunology 1985 Nov;56(3):523-31.

 (99)   Thumwood CM, Hunt NH, Cowden WB, Clark IA. Antioxidants can prevent cerebral malaria in Plasmodium berghei-
        infected mice. Br J Exp Pathol 1989 Jun;70(3):293-303.

(100)   Huang PL, Huang Z, Mashimo H, Bloch KD, Moskowitz MA, Bevan JA, et al. Hypertension in mice lacking the gene for
        endothelial nitric oxide synthase. Nature 1995 Sep 21;377(6546):239-42.

(101)   Rockett KA, Awburn MM, Aggarwal BB, Cowden WB, Clark IA. In vivo induction of nitrite and nitrate by tumor necrosis
        factor, lymphotoxin, and interleukin-1: possible roles in malaria. Infect Immun 1992 Sep;60(9):3725-30.

(102)   Gyan B, Troye-Blomberg M, Perlmann P, Bjorkman A. Human monocytes cultured with and without interferon-gamma
        inhibit Plasmodium falciparum parasite growth in vitro via secretion of reactive nitrogen intermediates. Parasite Immunol
        1994 Jul;16(7):371-5.

(103)   Clark IA, Cowden WB. The pathophysiology of falciparum malaria. Pharmacol Ther 2003 Aug;99(2):221-60.

(104)   Samdani AF, Dawson TM, Dawson VL. Nitric oxide synthase in models of focal ischemia. Stroke 1997 Jun;28(6):1283-8.

(105)   Weiss G, Thuma PE, Biemba G, Mabeza G, Werner ER, Gordeuk VR. Cerebrospinal fluid levels of biopterin, nitric oxide
        metabolites, and immune activation markers and the clinical course of human cerebral malaria. J Infect Dis 1998

(106)   Kilbourn RG, Belloni P. Endothelial cell production of nitrogen oxides in response to interferon gamma in combination with
        tumor necrosis factor, interleukin-1, or endotoxin. J Natl Cancer Inst 1990 May 2;82(9):772-6.

(107)   Levesque MC, Hobbs MR, Anstey NM, Vaughn TN, Chancellor JA, Pole A, et al. Nitric oxide synthase type 2 promoter
        polymorphisms, nitric oxide production, and disease severity in Tanzanian children with malaria. J Infect Dis 1999

(108)   Gramaglia I, Sobolewski P, Meays D, Contreras R, Nolan JP, Frangos JA, et al. Low nitric oxide bioavailability contributes
        to the genesis of experimental cerebral malaria. Nat Med 2006 Nov 12.

(109)   Ziesche R, Petkov V, Williams J, Zakeri SM, Mosgoller W, Knofler M, et al. Lipopolysaccharide and interleukin 1 augment
        the effects of hypoxia and inflammation in human pulmonary arterial tissue. Proc Natl Acad Sci U S A 1996 Oct

(110)   Boubou MI, Collette A, Voegtle D, Mazier D, Cazenave PA, Pied S. T cell response in malaria pathogenesis: selective
        increase in T cells carrying the TCR V(beta)8 during experimental cerebral malaria. Int Immunol 1999 Sep;11(9):1553-62.

(111)   Hermsen C, van de WT, Mommers E, Sauerwein R, Eling W. Depletion of CD4+ or CD8+ T-cells prevents Plasmodium
        berghei induced cerebral malaria in end-stage disease. Parasitology 1997 Jan;114 ( Pt 1):7-12.

(112)   Dewalick S, Amante FH, McSweeney KA, Randall LM, Stanley AC, Haque A, et al. Cutting Edge: Conventional Dendritic
        Cells Are the Critical APC Required for the Induction of Experimental Cerebral Malaria. J Immunol 2007 May

(113)   Coltel N, Combes V, Hunt NH, Grau GE. Cerebral malaria -- a neurovascular pathology with many riddles still to be
        solved. Curr Neurovasc Res 2004 Apr;1(2):91-110.

(114)   Senaldi G, Vesin C, Chang R, Grau GE, Piguet PF. Role of polymorphonuclear neutrophil leukocytes and their integrin
        CD11a (LFA-1) in the pathogenesis of severe murine malaria. Infect Immun 1994 Apr;62(4):1144-9.

(115)   Chen L, Zhang Z, Sendo F. Neutrophils play a critical role in the pathogenesis of experimental cerebral malaria. Clin Exp
        Immunol 2000 Apr;120(1):125-33.

(116)   Chen L, Sendo F. Cytokine and chemokine mRNA expression in neutrophils from CBA/NSlc mice infected with
        Plasmodium berghei ANKA that induces experimental cerebral malaria. Parasitol Int 2001 Jul;50(2):139-43.

(117)   Stach JL, Dufrenoy E, Roffi J, Bach MA. T-cell subsets and natural killer activity in Plasmodium falciparum-infected
        children. Clin Immunol Immunopathol 1986 Jan;38(1):129-34.

(118)   Horstmann RD, Dietrich M, Bienzle U, Rasche H. Malaria-induced thrombocytopenia. Blut 1981 Mar;42(3):157-64.

(119)   Combes V, Coltel N, Faille D, Wassmer SC, Grau GE. Cerebral malaria: role of microparticles and platelets in alterations
        of the blood-brain barrier. Int J Parasitol 2006 May 1;36(5):541-6.

(120)   Wassmer SC, Combes V, Candal FJ, Juhan-Vague I, Grau GE. Platelets potentiate brain endothelial alterations induced
        by Plasmodium falciparum. Infect Immun 2006 Jan;74(1):645-53.

(121)   Schofield L. Intravascular infiltrates and organ-specific inflammation in malaria pathogenesis. Immunol Cell Biol 2007

(122)   Schofield L, Grau GE. Immunological processes in malaria pathogenesis. Nat Rev Immunol 2005 Sep;5(9):722-35.

(123)   Schofield L, Novakovic S, Gerold P, Schwarz RT, McConville MJ, Tachado SD. Glycosylphosphatidylinositol toxin of
        Plasmodium up-regulates intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and E-selectin expression
        in vascular endothelial cells and increases leukocyte and parasite cytoadherence via tyrosine kinase-dependent signal
        transduction. J Immunol 1996 Mar 1;156(5):1886-96.

(124)   Schofield L, Hewitt MC, Evans K, Siomos MA, Seeberger PH. Synthetic GPI as a candidate anti-toxic vaccine in a model
        of malaria. Nature 2002 Aug 15;418(6899):785-9.

(125)   Arese P, Schwarzer E. Malarial pigment (haemozoin): a very active 'inert' substance. Ann Trop Med Parasitol 1997

(126)   Parroche P, Lauw FN, Goutagny N, Latz E, Monks BG, Visintin A, et al. Malaria hemozoin is immunologically inert but
        radically enhances innate responses by presenting malaria DNA to Toll-like receptor 9. Proc Natl Acad Sci U S A 2007
        Feb 6;104(6):1919-24.

(127)   Coban C, Ishii KJ, Kawai T, Hemmi H, Sato S, Uematsu S, et al. Toll-like receptor 9 mediates innate immune activation
        by the malaria pigment hemozoin. J Exp Med 2005 Jan 3;201(1):19-25.

(128)   Guiyedi V, Chanseaud Y, Fesel C, Snounou G, Rousselle JC, Lim P, et al. Self-reactivities to the non-erythroid alpha
        spectrin correlate with cerebral malaria in gabonese children. PLoS ONE 2007;2:e389.

(129)   Dugue C, Perraut R, Youinou P, Renaudineau Y. Effects of anti-endothelial cell antibodies in leprosy and malaria. Infect
        Immun 2004 Jan;72(1):301-9.

(130)   Perlmann P, Perlmann H, ElGhazali G, Blomberg MT. IgE and tumor necrosis factor in malaria infection. Immunol Lett
        1999 Jan;65(1-2):29-33.

(131)   Fadeel B, Orrenius S, Zhivotovsky B. Apoptosis in human disease: a new skin for the old ceremony? Biochem Biophys
        Res Commun 1999 Dec 29;266(3):699-717.

(132)   Leist M, Jaattela M. Four deaths and a funeral: from caspases to alternative mechanisms. Nat Rev Mol Cell Biol 2001

(133)   Fadeel B, Orrenius S, Zhivotovsky B. Apoptosis in human disease: a new skin for the old ceremony? Biochem Biophys
        Res Commun 1999 Dec 29;266(3):699-717.

(134)   Sairanen T, Karjalainen-Lindsberg ML, Paetau A, Ijas P, Lindsberg PJ. Apoptosis dominant in the periinfarct area of
        human ischaemic stroke--a possible target of antiapoptotic treatments. Brain 2006 Jan;129(Pt 1):189-99.

(135)   Lackner P, Burger C, Pfaller K, Heussler V, Helbok R, Morandell M, et al. Apoptosis in experimental cerebral malaria:
        spatial profile of cleaved caspase-3 and ultrastructural alterations in different disease stages. Neuropathol Appl Neurobiol
        2007 Apr 18.

(136)   Kaiser K, Texier A, Ferrandiz J, Buguet A, Meiller A, Latour C, et al. Recombinant Human Erythropoietin Prevents the
        Death of Mice during Cerebral Malaria. J Infect Dis 2006 Apr 1;193(7):987-95.

(137)   Ballabh P, Braun A, Nedergaard M. The blood-brain barrier: an overview: structure, regulation, and clinical implications.
        Neurobiol Dis 2004 Jun;16(1):1-13.

(138)   Hawkins BT, Davis TP. The blood-brain barrier/neurovascular unit in health and disease. Pharmacol Rev 2005

(139)   Bechmann I, Galea I, Perry VH. What is the blood-brain barrier (not)? Trends Immunol 2007 Jan;28(1):5-11.

(140)   Brown H, Rogerson S, Taylor T, Tembo M, Mwenechanya J, Molyneux M, et al. Blood-brain barrier function in cerebral
        malaria in Malawian children. Am J Trop Med Hyg 2001 Mar;64(3-4):207-13.

(141)   Brown H, Hien TT, Day N, Mai NT, Chuong LV, Chau TT, et al. Evidence of blood-brain barrier dysfunction in human
        cerebral malaria. Neuropathol Appl Neurobiol 1999 Aug;25(4):331-40.

(142)   Brown HC, Chau TT, Mai NT, Day NP, Sinh DX, White NJ, et al. Blood-brain barrier function in cerebral malaria and CNS
        infections in Vietnam. Neurology 2000 Jul 12;55(1):104-11.

(143)   Warrell DA, Looareesuwan S, Phillips RE, White NJ, Warrell MJ, Chapel HM, et al. Function of the blood-cerebrospinal
        fluid barrier in human cerebral malaria: rejection of the permeability hypothesis. Am J Trop Med Hyg 1986 Sep;35(5):882-

(144)   Das BS, Mohanty S, Mishra SK, Patnaik JK, Satpathy SK, Mohanty D, et al. Increased cerebrospinal fluid protein and
        lipid peroxidation products in patients with cerebral malaria. Trans R Soc Trop Med Hyg 1991 Nov;85(6):733-4.

(145)   Newton CR, Crawley J, Sowumni A, Waruiru C, Mwangi I, English M, et al. Intracranial hypertension in Africans with
        cerebral malaria. Arch Dis Child 1997 Mar;76(3):219-26.

(146)   Waller D, Crawley J, Nosten F, Chapman D, Krishna S, Craddock C, et al. Intracranial pressure in childhood cerebral
        malaria. Trans R Soc Trop Med Hyg 1991 May;85(3):362-4.

(147)   Thumwood CM, Hunt NH, Clark IA, Cowden WB. Breakdown of the blood-brain barrier in murine cerebral malaria.
        Parasitology 1988 Jun;96 ( Pt 3):579-89.

(148)   Pino P, Taoufiq Z, Nitcheu J, Vouldoukis I, Mazier D. Blood-brain barrier breakdown during cerebral malaria: suicide or
        murder? Thromb Haemost 2005 Aug;94(2):336-40.

(149)   Hunt NH, Golenser J, Chan-Ling T, Parekh S, Rae C, Potter S, et al. Immunopathogenesis of cerebral malaria. Int J
        Parasitol 2006 May;36(5):569-82.

(150)   Penet MF, Viola A, Confort-Gouny S, Le Fur Y, Duhamel G, Kober F, et al. Imaging Experimental Cerebral Malaria In
        Vivo: Significant Role of Ischemic Brain Edema. J Neurosci 2005 Aug 10;25(32):7352-8.

(151)   Penet MF, Kober F, Confort-Gouny S, Le FY, Dalmasso C, Coltel N, et al. Magnetic Resonance Spectroscopy Reveals
        an Impaired Brain Metabolic Profile in Mice Resistant to Cerebral Malaria Infected with Plasmodium berghei ANKA. J Biol
        Chem 2007 May 11;282(19):14505-14.

(152)   Combes V, Taylor TE, Juhan-Vague I, Mege JL, Mwenechanya J, Tembo M, et al. Circulating Endothelial Microparticles
        in Malawian Children With Severe Falciparum Malaria Complicated With Coma. JAMA 2004 Jun 2;291(21):2542-a.

(153)   Renia L, Potter SM, Mauduit M, Rosa DS, Kayibanda M, Deschemin JC, et al. Pathogenic T cells in cerebral malaria. Int J
        Parasitol 2006 May 1;36(5):547-54.

(154)   Troye-Blomberg M, Romero P, Patarroyo ME, Bjorkman A, Perlmann P. Regulation of the immune response in
        Plasmodium falciparum malaria. III. Proliferative response to antigen in vitro and subset composition of T cells from
        patients with acute infection or from immune donors. Clin Exp Immunol 1984 Nov;58(2):380-7.

(155)   Pombo DJ, Lawrence G, Hirunpetcharat C, Rzepczyk C, Bryden M, Cloonan N, et al. Immunity to malaria after
        administration of ultra-low doses of red cells infected with Plasmodium falciparum. Lancet 2002 Aug 24;360(9333):610-7.

(156)   Medana IM, Day NP, Hien TT, Mai NT, Bethell D, Phu NH, et al. Axonal injury in cerebral malaria. Am J Pathol 2002

(157)   Kumar KA, Babu PP. Mitochondrial anomalies are associated with the induction of intrinsic cell death proteins-Bcl(2),
        Bax, cytochrome-c and p53 in mice brain during experimental fatal murine cerebral malaria. Neurosci Lett 2002 Sep

(158)   Schluesener HJ, Kremsner PG, Meyermann R. Widespread expression of MRP8 and MRP14 in human cerebral malaria
        by microglial cells. Acta Neuropathol (Berl) 1998 Dec;96(6):575-80.

(159)   Medana IM, Mai NT, Day NP, Hien TT, Bethell D, Phu NH, et al. Cellular stress and injury responses in the brains of adult
        Vietnamese patients with fatal Plasmodium falciparum malaria. Neuropathol Appl Neurobiol 2001 Dec;27(6):421-33.

(160)   Pamplona A, Ferreira A, Balla J, Jeney V, Balla G, Epiphanio S, et al. Heme oxygenase-1 and carbon monoxide
        suppress the pathogenesis of experimental cerebral malaria. Nat Med 2007 May 13.

(161)   Balla G, Jacob HS, Balla J, Rosenberg M, Nath K, Apple F, et al. Ferritin: a cytoprotective antioxidant strategem of
        endothelium. J Biol Chem 1992 Sep 5;267(25):18148-53.

(162)   Treeprasertsuk S, Krudsood S, Tosukhowong T, Maek AN, Vannaphan S, Saengnetswang T, et al. N-acetylcysteine in
        severe falciparum malaria in Thailand. Southeast Asian J Trop Med Public Health 2003 Mar;34(1):37-42.

(163)   Watt G, Jongsakul K, Ruangvirayuth R. A pilot study of N-acetylcysteine as adjunctive therapy for severe malaria. QJM
        2002 May;95(5):285-90.

(164)   Arreesrisom P, Dondorp AM, Looareesuwan S, Udomsangpetch R. Suppressive effects of the anti-oxidant N-
        acetylcysteine on the anti-malarial activity of artesunate. Parasitol Int 2007 May 10.

(165)   Bartt RE. Multiple sclerosis, natalizumab therapy, and progressive multifocal leukoencephalopathy. Curr Opin Neurol
        2006 Aug;19(4):341-9.

(166)   Ho M, Hoang HL, Lee KM, Liu N, MacRae T, Montes L, et al. Ectophosphorylation of CD36 regulates cytoadherence of
        Plasmodium falciparum to microvascular endothelium under flow conditions. Infect Immun 2005 Dec;73(12):8179-87.

(167)   Dondorp AM, Silamut K, Charunwatthana P, Chuasuwanchai S, Ruangveerayut R, Krintratun S, et al. Levamisole inhibits
        sequestration of infected red blood cells in patients with falciparum malaria. J Infect Dis 2007 Aug 1;196(3):460-6.

(168)   Jelkmann W. Effects of erythropoietin on brain function. Curr Pharm Biotechnol 2005 Feb;6(1):65-79.

(169)   Tan CC, Eckardt KU, Firth JD, Ratcliffe PJ. Feedback modulation of renal and hepatic erythropoietin mRNA in response
        to graded anemia and hypoxia. Am J Physiol 1992 Sep;263(3 Pt 2):F474-F481.

(170)   Koury MJ, Sawyer ST, Brandt SJ. New insights into erythropoiesis. Curr Opin Hematol 2002 Mar;9(2):93-100.

(171)   Buemi M, Cavallaro E, Floccari F, Sturiale A, Aloisi C, Trimarchi M, et al. The pleiotropic effects of erythropoietin in the
        central nervous system. J Neuropathol Exp Neurol 2003 Mar;62(3):228-36.

(172)   Hasselblatt M, Ehrenreich H, Siren AL. The brain erythropoietin system and its potential for therapeutic exploitation in
        brain disease. J Neurosurg Anesthesiol 2006 Apr;18(2):132-8.

(173)   Marti HH, Wenger RH, Rivas LA, Straumann U, Digicaylioglu M, Henn V, et al. Erythropoietin gene expression in human,
        monkey and murine brain. Eur J Neurosci 1996 Apr;8(4):666-76.

(174)   Pacary E, Petit E, Bernaudin M. Erythropoietin, a cytoprotective and regenerative cytokine, and the hypoxic brain.
        Neurodegener Dis 2006;3(1-2):87-93.

(175)   Masuda S, Okano M, Yamagishi K, Nagao M, Ueda M, Sasaki R. A novel site of erythropoietin production. Oxygen-
        dependent production in cultured rat astrocytes. J Biol Chem 1994 Jul 29;269(30):19488-93.

(176)   Juul SE, Yachnis AT, Rojiani AM, Christensen RD. Immunohistochemical localization of erythropoietin and its receptor in
        the developing human brain. Pediatr Dev Pathol 1999 Mar;2(2):148-58.

(177)   Siren AL, Knerlich F, Poser W, Gleiter CH, Bruck W, Ehrenreich H. Erythropoietin and erythropoietin receptor in human
        ischemic/hypoxic brain. Acta Neuropathol (Berl) 2001 Mar;101(3):271-6.

(178)   Siren AL, Ehrenreich H. Erythropoietin--a novel concept for neuroprotection. Eur Arch Psychiatry Clin Neurosci 2001

(179)   Wenger RH. Mammalian oxygen sensing, signalling and gene regulation. J Exp Biol 2000 Apr;203(Pt 8):1253-63.

(180)   Wenger RH. Cellular adaptation to hypoxia: O2-sensing protein hydroxylases, hypoxia-inducible transcription factors, and
        O2-regulated gene expression. FASEB J 2002 Aug;16(10):1151-62.

(181)   Brines ML, Ghezzi P, Keenan S, Agnello D, de Lanerolle NC, Cerami C, et al. Erythropoietin crosses the blood-brain
        barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 2000 Sep 12;97(19):10526-31.

(182)   Leist M, Ghezzi P, Grasso G, Bianchi R, Villa P, Fratelli M, et al. Derivatives of Erythropoietin That Are Tissue Protective
        But Not Erythropoietic. Science 2004 Jul 9;305(5681):239-42.

(183)   Chong ZZ, Kang JQ, Maiese K. Erythropoietin: cytoprotection in vascular and neuronal cells. Curr Drug Targets
        Cardiovasc Haematol Disord 2003 Jun;3(2):141-54.

(184)   Siren AL, Fratelli M, Brines M, Goemans C, Casagrande S, Lewczuk P, et al. Erythropoietin prevents neuronal apoptosis
        after cerebral ischemia and metabolic stress. Proc Natl Acad Sci U S A 2001 Mar 27;98(7):4044-9.

(185)   Masuda S, Nagao M, Takahata K, Konishi Y, Gallyas F, Jr., Tabira T, et al. Functional erythropoietin receptor of the cells
        with neural characteristics. Comparison with receptor properties of erythroid cells. J Biol Chem 1993 May

(186)   Campana WM, Misasi R, O'Brien JS. Identification of a neurotrophic sequence in erythropoietin. Int J Mol Med 1998

(187)   Brines M, Grasso G, Fiordaliso F, Sfacteria A, Ghezzi P, Fratelli M, et al. Erythropoietin mediates tissue protection
        through an erythropoietin and common beta-subunit heteroreceptor. Proc Natl Acad Sci U S A 2004 Oct

(188)   Lewczuk P, Hasselblatt M, Kamrowski-Kruck H, Heyer A, Unzicker C, Siren AL, et al. Survival of hippocampal neurons in
        culture upon hypoxia: effect of erythropoietin. Neuroreport 2000 Nov 9;11(16):3485-8.

(189)   Sakanaka M, Wen TC, Matsuda S, Masuda S, Morishita E, Nagao M, et al. In vivo evidence that erythropoietin protects
        neurons from ischemic damage. Proc Natl Acad Sci U S A 1998 Apr 14;95(8):4635-40.

(190)   Digicaylioglu M, Lipton SA. Erythropoietin-mediated neuroprotection involves cross-talk between Jak2 and NF-kappaB
        signalling cascades. Nature 2001 Aug 9;412(6847):641-7.

(191)   Wen TC, Sadamoto Y, Tanaka J, Zhu PX, Nakata K, Ma YJ, et al. Erythropoietin protects neurons against chemical
        hypoxia and cerebral ischemic injury by up-regulating Bcl-xL expression. J Neurosci Res 2002 Mar 15;67(6):795-803.

(192)   Chong ZZ, Kang JQ, Maiese K. Erythropoietin is a novel vascular protectant through activation of Akt1 and mitochondrial
        modulation of cysteine proteases. Circulation 2002 Dec 3;106(23):2973-9.

(193)   Diaz Z, Assaraf MI, Miller WH, Jr., Schipper HM. Astroglial cytoprotection by erythropoietin pre-conditioning: implications
        for ischemic and degenerative CNS disorders. J Neurochem 2005 Apr;93(2):392-402.

(194)   Villa P, Bigini P, Mennini T, Agnello D, Laragione T, Cagnotto A, et al. Erythropoietin Selectively Attenuates Cytokine
        Production and Inflammation in Cerebral Ischemia by Targeting Neuronal Apoptosis. The Journal of Experimental
        Medicine 2003 Sep 15;198(6):971-5.

(195)   Chattopadhyay A, Choudhury TD, Bandyopadhyay D, Datta AG. Protective effect of erythropoietin on the oxidative
        damage of erythrocyte membrane by hydroxyl radical. Biochem Pharmacol 2000 Feb 15;59(4):419-25.

(196)   Ribatti D, Presta M, Vacca A, Ria R, Giuliani R, Dell'Era P, et al. Human erythropoietin induces a pro-angiogenic
        phenotype in cultured endothelial cells and stimulates neovascularization in vivo. Blood 1999 Apr 15;93(8):2627-36.

(197)   Shingo T, Sorokan ST, Shimazaki T, Weiss S. Erythropoietin regulates the in vitro and in vivo production of neuronal
        progenitors by mammalian forebrain neural stem cells. J Neurosci 2001 Dec 15;21(24):9733-43.

(198)   Studer L, Csete M, Lee SH, Kabbani N, Walikonis J, Wold B, et al. Enhanced proliferation, survival, and dopaminergic
        differentiation of CNS precursors in lowered oxygen. J Neurosci 2000 Oct 1;20(19):7377-83.

(199)   Vairano M, Dello RC, Pozzoli G, Battaglia A, Scambia G, Tringali G, et al. Erythropoietin exerts anti-apoptotic effects on
        rat microglial cells in vitro. Eur J Neurosci 2002 Aug;16(4):584-92.

(200)   Risau W. Mechanisms of angiogenesis. Nature 1997 Apr 17;386(6626):671-4.

(201)   Marti HH. Erythropoietin and the hypoxic brain. J Exp Biol 2004 Aug;207(Pt 18):3233-42.

(202)   Li Y, Lu ZY, Ogle M, Wei L. Erythropoietin Prevents Blood Brain Barrier Damage Induced by Focal Cerebral Ischemia in
        Mice. Neurochem Res 2007 Jun 12.

(203)   Martinez-Estrada OM, Rodriguez-Millan E, Gonzalez-De VE, Reina M, Vilaro S, Fabre M. Erythropoietin protects the in
        vitro blood-brain barrier against VEGF-induced permeability. Eur J Neurosci 2003 Nov;18(9):2538-44.

(204)   Deininger MH, Winkler S, Kremsner PG, Meyermann R, Schluesener HJ. Angiogenic proteins in brains of patients who
        died with cerebral malaria. J Neuroimmunol 2003 Sep;142(1-2):101-11.

(205)   Bernaudin M, Marti HH, Roussel S, Divoux D, Nouvelot A, Mackenzie ET, et al. A potential role for erythropoietin in focal
        permanent cerebral ischemia in mice. J Cereb Blood Flow Metab 1999 Jun;19(6):643-51.

(206)   Sadamoto Y, Igase K, Sakanaka M, Sato K, Otsuka H, Sakaki S, et al. Erythropoietin prevents place navigation disability
        and cortical infarction in rats with permanent occlusion of the middle cerebral artery. Biochem Biophys Res Commun
        1998 Dec 9;253(1):26-32.

(207)   Grasso G, Buemi M, Alafaci C, Sfacteria A, Passalacqua M, Sturiale A, et al. Beneficial effects of systemic administration
        of recombinant human erythropoietin in rabbits subjected to subarachnoid hemorrhage. Proc Natl Acad Sci U S A 2002
        Apr 16;99(8):5627-31.

(208)   Matsushita H, Johnston MV, Lange MS, Wilson MA. Protective effect of erythropoietin in neonatal hypoxic ischemia in
        mice. Neuroreport 2003 Sep 15;14(13):1757-61.

(209)   Kaptanoglu E, Solaroglu I, Okutan O, Surucu HS, Akbiyik F, Beskonakli E. Erythropoietin exerts neuroprotection after
        acute spinal cord injury in rats: effect on lipid peroxidation and early ultrastructural findings. Neurosurg Rev 2004

(210)   Lipsic E, Schoemaker RG, van der MP, Voors AA, van Veldhuisen DJ, van Gilst WH. Protective effects of erythropoietin
        in cardiac ischemia: from bench to bedside. J Am Coll Cardiol 2006 Dec 5;48(11):2161-7.

(211)   Bullard AJ, Govewalla P, Yellon DM. Erythropoietin protects the myocardium against reperfusion injury in vitro and in
        vivo. Basic Res Cardiol 2005 Sep;100(5):397-403.

(212)   Ehrenreich H, Hasselblatt M, Dembowski C, Cepek L, Lewczuk P, Stiefel M, et al. Erythropoietin therapy for acute stroke
        is both safe and beneficial. Mol Med 2002 Aug;8(8):495-505.

(213)   Bath PM, Sprigg N. Colony stimulating factors (including erythropoietin, granulocyte colony stimulating factor and
        analogues) for stroke. Cochrane Database Syst Rev 2007;(2):CD005207.

(214)   Jelkmann W. Use of recombinant human erythropoietin as an antianemic and performance enhancing drug. Curr Pharm
        Biotechnol 2000 Jul;1(1):11-31.

(215)   Phrommintikul A, Haas SJ, Elsik M, Krum H. Mortality and target haemoglobin concentrations in anaemic patients with
        chronic kidney disease treated with erythropoietin: a meta-analysis. Lancet 2007 Feb 3;369(9559):381-8.

(216)   Casadevall N, Nataf J, Viron B, Kolta A, Kiladjian JJ, Martin-Dupont P, et al. Pure red-cell aplasia and antierythropoietin
        antibodies in patients treated with recombinant erythropoietin. N Engl J Med 2002 Feb 14;346(7):469-75.

(217)   Steinbrook R. Erythropoietin, the FDA, and oncology. N Engl J Med 2007 Jun 14;356(24):2448-51.

(218)   Haiden N, Cardona F, Schwindt J, Berger A, Kuhle S, Homoncik M, et al. Changes in thrombopoiesis and platelet
        reactivity in extremely low birth weight infants undergoing erythropoietin therapy for treatment of anaemia of prematurity.
        Thromb Haemost 2005 Jan;93(1):118-23.

(219)   Haiden N, Schwindt J, Cardona F, Berger A, Klebermass K, Wald M, et al. Effects of a combined therapy of
        erythropoietin, iron, folate, and vitamin B12 on the transfusion requirements of extremely low birth weight infants.
        Pediatrics 2006 Nov;118(5):2004-13.

(220)   Ledbetter DJ, Juul SE. Erythropoietin and the incidence of necrotizing enterocolitis in infants with very low birth weight. J
        Pediatr Surg 2000 Feb;35(2):178-81.

(221)   Romagnoli C, Zecca E, Gallini F, Girlando P, Zuppa AA. Do recombinant human erythropoietin and iron supplementation
        increase the risk of retinopathy of prematurity? Eur J Pediatr 2000 Aug;159(8):627-8.

(222)   Lipsic E, van der MP, Voors AA, Westenbrink BD, van den Heuvel AF, de Boer HC, et al. A single bolus of a long-acting
        erythropoietin analogue darbepoetin alfa in patients with acute myocardial infarction: a randomized feasibility and safety
        study. Cardiovasc Drugs Ther 2006 Apr;20(2):135-41.

(223)   Cheung WK, Goon BL, Guilfoyle MC, Wacholtz MC. Pharmacokinetics and pharmacodynamics of recombinant human
        erythropoietin after single and multiple subcutaneous doses to healthy subjects. Clin Pharmacol Ther 1998 Oct;64(4):412-

(224)   Maiese K, Li F, Chong ZZ. Erythropoietin in the brain: can the promise to protect be fulfilled? Trends Pharmacol Sci 2004

(225)   Hidalgo J, Aschner M, Zatta P, Vasak M. Roles of the metallothionein family of proteins in the central nervous system.
        Brain Res Bull 2001 May 15;55(2):133-45.

(226)   Hidalgo J, Penkowa M, Giralt M, Carrasco J, Molinero A. Metallothionein expression and oxidative stress in the brain.
        Methods Enzymol 2002;348:238-49.

(227)   Klaassen CD, Liu J, Choudhuri S. Metallothionein: an intracellular protein to protect against cadmium toxicity. Annu Rev
        Pharmacol Toxicol 1999;39:267-94.

(228)   Chung RS, West AK. A role for extracellular metallothioneins in CNS injury and repair. Neuroscience 2004;123(3):595-9.

(229)   Giralt M, Penkowa M, Lago N, Molinero A, Hidalgo J. Metallothionein-1+2 protect the CNS after a focal brain injury. Exp
        Neurol 2002 Jan;173(1):114-28.

(230)   Penkowa M, Carrasco J, Giralt M, Moos T, Hidalgo J. CNS wound healing is severely depressed in metallothionein I- and
        II-deficient mice. J Neurosci 1999 Apr 1;19(7):2535-45.

(231)   Penkowa M, Giralt M, Camats J, Hidalgo J. Metallothionein 1+2 protect the CNS during neuroglial degeneration induced
        by 6-aminonicotinamide. J Comp Neurol 2002 Mar 5;444(2):174-89.

(232)   van Lookeren CM, Thibodeaux H, van Bruggen N, Cairns B, Gerlai R, Palmer JT, et al. Evidence for a protective role of
        metallothionein-1 in focal cerebral ischemia. Proc Natl Acad Sci U S A 1999 Oct 26;96(22):12870-5.

(233)   Penkowa M, Hidalgo J. Metallothionein treatment reduces proinflammatory cytokines IL-6 and TNF-alpha and apoptotic
        cell death during experimental autoimmune encephalomyelitis (EAE). Exp Neurol 2001 Jul;170(1):1-14.

(234)   Snow RW, Guerra CA, Noor AM, Myint HY, Hay SI. The global distribution of clinical episodes of Plasmodium falciparum
        malaria. Nature 2005 Mar 10;434(7030):214-7.

(235)   Brown H, Turner G, Rogerson S, Tembo M, Mwenechanya J, Molyneux M, et al. Cytokine expression in the brain in
        human cerebral malaria. Journal of Infectious Diseases 1999 Nov;180(5):1742-6.

(236)   Barnham KJ, Masters CL, Bush AI. Neurodegenerative diseases and oxidative stress. Nat Rev Drug Discov 2004

(237)   Pino P, Vouldoukis I, Dugas N, Hassani-Loppion G, Dugas B, Mazier D. Redox-dependent apoptosis in human
        endothelial cells after adhesion of Plasmodium falciparum-infected erythrocytes. Ann N Y Acad Sci 2003 Dec;1010:582-6.

(238)   Hasnain SE, Begum R, Ramaiah KV, Sahdev S, Shajil EM, Taneja TK, et al. Host-pathogen interactions during
        apoptosis. J Biosci 2003 Apr;28(3):349-58.

(239)   Zhang F, Yin W, Chen J. Apoptosis in cerebral ischemia: executional and regulatory signaling mechanisms. Neurol Res
        2004 Dec;26(8):835-45.

(240)   Gitau EN, Newton CR. Review Article: blood-brain barrier in falciparum malaria. Trop Med Int Health 2005 Mar;10(3):285-

(241)   Curfs JH, Hermsen CC, Meuwissen JH, Eling WM. Immunization against cerebral pathology in Plasmodium berghei-
        infected mice. Parasitology 1992 Aug;105 ( Pt 1):7-14.

(242)   Curfs JH, Schetters TP, Hermsen CC, Jerusalem CR, van Zon AA, Eling WM. Immunological aspects of cerebral lesions
        in murine malaria. Clin Exp Immunol 1989 Jan;75(1):136-40.

(243)   Gasull T, Rebollo DV, Romero B, Hidalgo J. Development of a competitive double antibody radioimmunoassay for rat
        metallothionein. J Immunoassay 1993 Dec;14(4):209-25.

(244)   Penkowa M, Hidalgo J, Moos T. Increased astrocytic expression of metallothioneins I + II in brainstem of adult rats
        treated with 6-aminonicotinamide. Brain Res 1997 Nov 7;774(1-2):256-9.

(245)   Penkowa M, Quintana A, Carrasco J, Giralt M, Molinero A, Hidalgo J. Metallothionein prevents neurodegeneration and
        central nervous system cell death after treatment with gliotoxin 6-aminonicotinamide. J Neurosci Res 2004 Jul 1;77(1):35-

(246)   Zhang F, Yin W, Chen J. Apoptosis in cerebral ischemia: executional and regulatory signaling mechanisms. Neurol Res
        2004 Dec;26(8):835-45.

(247)   Medana IM, Day NP, Hien TT, Mai NT, Bethell D, Phu NH, et al. Axonal injury in cerebral malaria. Am J Pathol 2002

(248)   Turner G. Cerebral malaria. Brain Pathol 1997 Jan;7(1):569-82.

(249)   Penet MF, Viola A, Confort-Gouny S, Le Fur Y, Duhamel G, Kober F, et al. Imaging Experimental Cerebral Malaria In
        Vivo: Significant Role of Ischemic Brain Edema. J Neurosci 2005 Aug 10;25(32):7352-8.

(250)   Beckert S, Farrahi F, Perveen GQ, Aslam R, Scheuenstuhl H, Coerper S, et al. IGF-I-induced VEGF expression in
        HUVEC involves phosphorylation and inhibition of poly(ADP-ribose)polymerase. Biochem Biophys Res Commun 2006
        Jan 6.

(251)   Ferrara N. Vascular endothelial growth factor: basic science and clinical progress. Endocr Rev 2004 Aug;25(4):581-611.

(252)   Carter JA, Neville BG, White S, Ross AJ, Otieno G, Mturi N, et al. Increased prevalence of epilepsy associated with
        severe falciparum malaria in children. Epilepsia 2004 Aug;45(8):978-81.

(253)   Carter JA, Ross AJ, Neville BG, Obiero E, Katana K, Mung'Ala-Odera V, et al. Developmental impairments following
        severe falciparum malaria in children. Trop Med Int Health 2005 Jan;10(1):3-10.

(254)   Van Hensbroek MB, Palmer A, Jaffar S, Schneider G, Kwiatkowski D. Residual neurologic sequelae after childhood
        cerebral malaria. J Pediatr 1997 Jul;131(1 Pt 1):125-9.

(255)   Liou AK, Clark RS, Henshall DC, Yin XM, Chen J. To die or not to die for neurons in ischemia, traumatic brain injury and
        epilepsy: a review on the stress-activated signaling pathways and apoptotic pathways. Prog Neurobiol 2003

(256)   Kemp K, Akanmori BD, Kurtzhals JA, Adabayeri V, Goka BQ, Hviid L. Acute P. falciparum malaria induces a loss of
        CD28- T IFN-gamma producing cells. Parasite Immunol 2002 Nov;24(11-12):545-8.

(257)   Leiriao P, Mota MM, Rodriguez A. Apoptotic Plasmodium-infected hepatocytes provide antigens to liver dendritic cells. J
        Infect Dis 2005 May 15;191(10):1576-81.

(258)   Crocker IP, Tanner OM, Myers JE, Bulmer JN, Walraven G, Baker PN. Syncytiotrophoblast degradation and the
        pathophysiology of the malaria-infected placenta. Placenta 2004 Apr;25(4):273-82.

(259)   Carter JA, Neville BG, Newton CR. Neuro-cognitive impairment following acquired central nervous system infections in
        childhood: a systematic review. Brain Res Brain Res Rev 2003 Sep;43(1):57-69.

(260)   Medana IM, Day NP, Hien TT, Mai NT, Bethell D, Phu NH, et al. Axonal injury in cerebral malaria. Am J Pathol 2002

(261)   Thumwood CM, Hunt NH, Clark IA, Cowden WB. Breakdown of the blood-brain barrier in murine cerebral malaria.
        Parasitology 1988 Jun;96 ( Pt 3):579-89.

(262)   Penkowa M. Methallothionein expression and roles in the central nervous system. Biomedical Reviews 2002;13:1-15.

(263)   Carrasco J, Penkowa M, Hadberg H, Molinero A, Hidalgo J. Enhanced seizures and hippocampal neurodegeneration
        following kainic acid-induced seizures in metallothionein-I + II-deficient mice. Eur J Neurosci 2000 Jul;12(7):2311-22.

(264)   Penkowa M, Hidalgo J. Metallothionein I+II expression and their role in experimental autoimmune encephalomyelitis. Glia
        2000 Dec;32(3):247-63.

(265)   Carrasco J, Giralt M, Penkowa M, Stalder AK, Campbell IL, Hidalgo J. Metallothioneins are upregulated in symptomatic
        mice with astrocyte-targeted expression of tumor necrosis factor-alpha. Exp Neurol 2000 May;163(1):46-54.

(266)   Adlard PA, West AK, Vickers JC. Increased density of metallothionein I/II-immunopositive cortical glial cells in the early
        stages of Alzheimer's disease. Neurobiol Dis 1998 Nov;5(5):349-56.

(267)   Carrasco J, Giralt M, Molinero A, Penkowa M, Moos T, Hidalgo J. Metallothionein (MT)-III: generation of polyclonal
        antibodies, comparison with MT-I+II in the freeze lesioned rat brain and in a bioassay with astrocytes, and analysis of
        Alzheimer's disease brains. J Neurotrauma 1999 Nov;16(11):1115-29.

(268)   Sillevis Smitt PA, Mulder TP, Verspaget HW, Blaauwgeers HG, Troost D, de Jong JM. Metallothionein in amyotrophic
        lateral sclerosis. Biol Signals 1994 Jul;3(4):193-7.

(269)   Penkowa M, Espejo C, Ortega-Aznar A, Hidalgo J, Montalban X, Martinez Caceres EM. Metallothionein expression in the
        central nervous system of multiple sclerosis patients. Cell Mol Life Sci 2003 Jun;60(6):1258-66.

(270)   Hidalgo J, Carrasco J. Regulation of the synthesis of brain metallothioneins. Neurotoxicology 1998 Aug;19(4-5):661-6.

(271)   Espejo C, Penkowa M, Demestre M, Montalban X, Martinez-Caceres EM. Time-course expression of CNS inflammatory,
        neurodegenerative tissue repair markers and metallothioneins during experimental autoimmune encephalomyelitis.
        Neuroscience 2005;132(4):1135-49.

(272)   Brown H, Turner G, Rogerson S, Tembo M, Mwenechanya J, Molyneux M, et al. Cytokine expression in the brain in
        human cerebral malaria. Journal of Infectious Diseases 1999 Nov;180(5):1742-6.

(273)   Wiese L, Kurtzhals JA, Penkowa M. Neuronal apoptosis, metallothionein expression and proinflammatory responses
        during cerebral malaria in mice. Exp Neurol 2006 Jul;200(1):216-26.

(274)   Chang KH, Tam M, Stevenson MM. Modulation of the course and outcome of blood-stage malaria by erythropoietin-
        induced reticulocytosis. J Infect Dis 2004 Feb 15;189(4):735-43.

(275)   Rae C, McQuillan JA, Parekh SB, Bubb WA, Weiser S, Balcar VJ, et al. Brain gene expression, metabolism, and
        bioenergetics: interrelationships in murine models of cerebral and noncerebral malaria. FASEB J 2004 Mar;18(3):499-

(276)   Wiese L, Hempel C, Kirkby N, Penkowa M, Kurtzhals JAL. Erythropoietin treatment increases survival and reduces
        neuronal apoptosis during murine cerebral malaria. (submitted) 2007 Mar.

(277)   Ehrenreich H, Aust C, Krampe H, Jahn H, Jacob S, Herrmann M, et al. Erythropoietin: novel approaches to
        neuroprotection in human brain disease. Metab Brain Dis 2004 Dec;19(3-4):195-206.

(278)   Villa P, van BJ, Larsen AK, Gerwien J, Christensen S, Cerami A, et al. Reduced functional deficits, neuroinflammation,
        and secondary tissue damage after treatment of stroke by nonerythropoietic erythropoietin derivatives. J Cereb Blood
        Flow Metab 2006 Jul 12.

(279)   Medana IM, Esiri MM. Axonal damage: a key predictor of outcome in human CNS diseases. Brain 2003 Mar;126(Pt


To top