Persistent cognitive impairment after cerebral malaria: models, mechanisms and adjunctive therapies

Keywords::

• antioxidants
• cerebral malaria
• cognitive dysfunction
• experimental malaria
• mouse models
• oxidant stress

The global burden of malaria is enormous: more than 40% of the world’s population is at risk for infections caused by Plasmodium parasites, including Plasmodium falciparum[1]. A recent report estimated that 451 million cases of clinical P. falciparum infection occurred in 2007 [2]. P. falciparum is the principal cause of syndromes of severe malaria, including cerebral malaria [3,4]. Coma is the defining feature of cerebral malaria, with variable incidence of other neurologic and systemic manifestations [4]. While only approximately 1% of P. falciparum infections progress to cerebral malaria, the syndrome is fatal in 10–20% of affected patients, causing 300,000–500,000 deaths each year [4]. Prevention of plasmodial infection, development of new and more effective antimalarial drugs, prevention of pathogen resistance to currently used antimalarials and, ultimately, eradication of the malarial parasites are compelling public health and scientific objectives [1]. Identification of adjunctive therapies that can reduce organ damage and clinical complications in severe malarial syndromes is also a priority [4,5].

African children are disproportionately at risk for developing cerebral malaria [4]. There is now evidence that many patients who are effectively treated and survive cerebral malaria have long-term physical or neurologic sequelae, and thus present a second clinical challenge [4,6–9]. In recent studies neurocognitive dysfunction was detected in as many as 26% of African children when follow-up testing was performed 6 months to 9 years after treatment of acute cerebral malaria [6–9]. Malaria with complicated seizures, a separate disorder, also causes late neurocognitive dysfunction [6,7]. This clinical outcome has not been studied in endemic regions of Asia or South America [2]. Nevertheless, 500,000–800,000 children develop cerebral malaria each year in Sub-Saharan Africa alone, suggesting that persistent cognitive dysfunction in survivors is a significant complication regardless of its incidence in other parts of the world. The potential impact on long-term learning and behavioral function, and the societal consequences that may result from deficiencies in these cerebral activities, are staggering, and will be amplified if persistent cognitive impairment is prevalent in other clinical presentations such as malaria with complicated seizures [6,7]. The added impact of post-malarial neurologic sequelae, such as deafness, blindness and epilepsy, on access to education and consequent cognitive abilities remains to be charted [10,11]. It is also possible that cognitive dysfunction occurs after successful therapy of cerebral malaria in adults, although the syndrome is much less common in older age groups [4]. Thus, while much remains to be learned about cognitive impairment after severe malaria, current evidence indicates that it is a substantial clinical problem. Mechanisms that underlie cognitive dysfunction after cerebral malaria, and measures that can be used to prevent and treat it are, however, largely unexplored.

Precise identification of molecular mechanisms that underlie sustained cognitive dysfunction after cerebral malaria in humans is likely to be daunting, making development of specific therapeutic measures equally challenging. Early symptoms of cerebral malaria are difficult to differentiate from other conditions such as encephalitis or meningitis [4]. In addition, there are many other confounding clinical variables. Histopathologic studies, while essential, are difficult to achieve in many populations at risk for cerebral malaria [4], and also reveal variability in brain pathology [12]. It is not yet clear what cellular pathologic changes are critical determinants of outcome in cerebral malaria [4,13], including, by extension, cognitive dysfunction after its treatment. Because multiple areas of the brain are involved in cerebral malaria, cellular and molecular changes that lead to death in the acute syndrome and to cognitive or neurologic dysfunction in survivors may be different or overlapping. Use of noninvasive neuroimaging procedures to study these issues is likely to be limited in many areas on the global malaria map in the immediate future because of cost and ethical constraints [4]. Clinical trials involving candidate adjunctive agents have the potential to yield both mechanistic and therapeutic information. They are costly and time consuming, however, and are frequently powered so that only dramatic benefits can be detected; they also usually focus on reversal of coma, short-term mortality, and other features of acute severe malarial syndromes as end points [14,15].

In complex and challenging conditions such as cerebral malaria, experimental models involving isolated cells and/or laboratory animals must often be utilized in parallel with observational studies of the ‘human model’, that is, affected patients [16]. Together with collaborators and colleagues, we recently reported studies employing mouse models of experimental cerebral malaria that were aimed at pathogenesis and mechanisms of persistent cognitive dysfunction after rescue therapy. Mice of genetic backgrounds that are susceptible or resistant to experimental cerebral malaria induced by rodent plasmodial parasites were examined using a complex multisystem screening assessment to identify early neurologic alterations, thus allowing rescue antimalarial treatment, and a follow-up battery of behavioral tests to investigate late cognitive function [17]. Cognitive impairment in contextual and aversive memory occurred in mice of susceptible genetic backgrounds infected with specific parasites (C57/black 6 infected with Plasmodium berghei ANKA; swiss webster infected with Plasmodium yoelli XL) and was sustained for a 30-day study period. Desruisseaux et al. previously reported cognitive dysfunction in the acute phase of P. berghei ANKA (PbA) infection [18], but it was unknown if cognitive defects persist after rescue therapy. Another central observation in our study is that persistent cognitive dysfunction can occur when effective rescue therapy is given early in experimental cerebral malaria, when neurologic manifestations are relatively subtle and before the onset of coma, seizures and other manifestations [17]. This has important implications if it also occurs in children or adults who are treated in early phases of severe P. falciparum infection and do not develop coma and frank cerebral malaria defined by consensus criteria [4].

We also found evidence for oxidant stress as a mechanism contributing to persistent cognitive dysfunction [17]. Previous reports indicate that reactive oxygen species are generated in malarial syndromes and that ‘oxidant stress’ is a potential target for intervention [14,19]. Oxidative damage to the membranes of parasitized and uninfected red blood cells, causing rigidity and impaired deformation, may contribute to sequestration of red blood cells in microvascular beds [19]. Sequestration in capillaries and post-capillary venules of the brain, leading to obstruction and parenchymal hypoxia, is thought by many investigators to be a sine qua non of clinical cerebral malaria [15], although contributions of red blood cell sequestration to neurobehavioral responses and cognitive impairment are unexplored. Using the murine models outlined above, we found that oxidant stress is induced in the brains of mice with early cerebral malaria [17]. Furthermore, administration of two antioxidants in combination, N-acetylcysteine and desferroxamine, as adjunctive treatment together with rescue antimalarial therapy prevented late cognitive impairment, and reduced microvascular sequestration in several areas of the brain [17]. Administration of adjunctive agents together with primary antiplasmodial drugs, and not simply alone, is critical in evaluation of candidate adjunctive therapies in preclinical studies [15,20,21]. In our investigation, this approach also provided evidence that oxidant attack and/or signaling is a critical mechanistic feature of sustained cognitive dysfunction in the experimental murine model. Interestingly, oxidant stress is reported to be a central mechanism in organ dysfunction syndromes in another systemic inflammatory syndrome, sepsis, and administration of desferroxamine and N-acetylcysteine in combination improved outcomes, including late memory loss, in experimental rodent models of sepsis [22,23]. Severe malaria and bacterial sepsis have similarities, although microvascular pathology and red blood cell rheology appear to be different [15]. Thus, beneficial effects of antioxidant agents on neurocognitive outcomes in experimental models of malaria [17] and sepsis [22,23] suggests that oxidant damage or injurious signaling may occur in key extravascular neuronal cells and pathways in the two conditions.

Our observations using the murine models led us to suggest that the specific combination of N-acetylcysteine and desferroxamine might be examined in clinical trials as an adjunctive therapy together with antimalarial drugs to improve or prevent cognitive dysfunction after cerebral malaria in patients [17]. Both antioxidants have previously been studied in humans with severe malaria [14,15,24–26], although late cognitive end points have not been examined. But are our findings in mice at all relevant to the human condition, and could they be informative in consideration of such a trial? Currently there is heated controversy over the relevance of experimental cerebral malaria, particularly the PbA-infected mouse, to clinical cerebral malaria in patients. One view is that murine models, and specifically, the PbA model, is at best of questionable value as an approach to identifying pathologic mechanisms, evaluating therapeutic interventions, and as basis for clinical trials of adjunctive interventions [15]. This opinion is contrary to that of others in the field [4], and generated a flurry of retorts defending murine cerebral malaria as a relevant experimental system with many similarities to the human clinical syndrome [13,20,21,27,28]. We share the latter opinion. Although few, if any, animal models perfectly replicate the human conditions they are used to mimic, when carefully chosen, interpreted and correlated, murine models of cerebral malaria can be used to explore possible mechanisms and examine features of the natural history that would be difficult or impossible to delineate in humans [29]. Experimental models can also be employed to critically vet candidate primary or adjunctive therapies [4,5,30,31] in a proof-of-principle fashion that provides key information as priorities for clinical trials are considered [15]. Mouse models of cerebral malaria are likely to substantively contribute to rigorous translational research in the field [20], narrowing the apparent gap between experimental and clinical investigation [15]. This may then have a positive impact on patients with this lethal and, in survivors, potentially debilitating neurologic syndrome.

Acknowledgements

The authors are grateful to the students and postdoctoral fellows who carried out studies mentioned in this editorial, and to their collaborators at the Universidade Federal do Rio de Janeiro and Universidade do Extremo sul Catarinese. Thoughtful and critical comments from Don Granger regarding this manuscript are appreciated.

Financial & competing interests disclosure

Work mentioned was partially supported by the National Institutes of Health (R03 NS04512 and R37 [R01] HL44525-22 to Guy A Zimmerman). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.