Induction of the vascular endothelial growth factor pathway in the brain of adults with fatal falciparum malaria is a non-specific response to severe disease

Abstract

Aims: Pathological or neuroprotective mechanisms in the brain in severe malaria may arise from microvascular obstruction with malaria-parasitized erythrocytes. This study aimed to investigate the role of hypoxia and induction of the vascular endothelial growth factor (VEGF) pathway in the neuropathophysiology of severe malaria.

Methods and results: Immunohistochemistry was performed on post mortem brain tissue sections from 20 cases of severe malaria and examined for the expression of transcriptional regulators of VEGF [hypoxia-inducible factor-1 alpha (HIF-1alpha), HIF-2alpha], DEC-1, VEGF, VEGF receptors 1 and 2, and the activated, phosphorylated VEGF receptor 2 (pKDR). HIFs showed limited protein expression and/or translocation to cell nuclei in severe malaria, but DEC-1, which is more stable and regulated by HIF-1alpha, was observed. There was heterogeneous expression of VEGF and its receptors in severe malaria and non-malarial disease controls. pKDR expression on vessels was greater in malaria cases than in controls but did not correlate with parasite sequestration. VEGF uptake by malaria parasites was observed.

Conclusions: VEGF and its receptor expression levels in severe malaria reflect a non-specific response to severe systemic disease. Potential manipulation of events at the vasculature by the parasite requires further investigation.

Figure 1

Hypoxia-inducible factor (HIF)-1α, HIF-2α and DEC-1 immunoreactivity of post mortem brain tissue of positive controls (peroxidase, brown) and severe malaria cases (alkaline phosphatase, red). A, Nuclear reactivity for HIF-1α in a renal clear cell carcinoma. B, Nuclear and cytoplasmic reactivity for HIF-2α in the same renal tumour. C, HIF-2α reactivity in a case of generalized hypoxic-ischaemic injury. In the cerebellum there is strong reactivity of HIF-2α on macrophages and glial cells around the depths of the sulci. D, HIF-2α reactivity surrounding a metastatic deposit of adenocarcinoma in the brain. Peritumoral macrophages show strong immunopositivity. E, HIF-1α is not observed on any of the sections from severe malaria cases. F, Nuclear and cytoplasmic HIF-2α is observed at low levels in at least one brain region of all severe malaria cases. G, DEC-1 reactivity of endothelial cells in a positive control case with infarction. H, Nuclear DEC-1 reactivity in neurons and glial cells in a severe malaria case.

Figure 2

Frequency of immunoreactivity for markers in the brain of fatal cases of severe malaria. The degree of reactivity per tissue section was semiquantified using the following scale: 0, no reactivity; 1, <1% of cells/vessels reactive; 2, 1–10% of cells/vessels reactive; 3, >10% cells/vessels reactive. Each score represents the average of results from three brain regions including the temporal cortex, diencephalon (thalamus, projections of internal capsule) and medulla of the brainstem for an individual patient. *Statistically higher median levels in the severe malaria group compared with non-neurological controls from the UK with multi-organ dysfunction.

Figure 4

A, A vessel containing sequestered malaria parasites that are strongly positive for vascular endothelial growth factor (VEGF). Neurons are also immunolabelled for VEGF in this section (*). B, A vessel containing parasites that are VEGF+ (filled arrow). An adjacent vessel with parasites that are not positive for VEGF (empty arrow). This vessel shows light labelling for VEGF on the endothelial cells. Neurons (*) and glia (arrowheads) are also labelled for VEGF in this section. C, Graph showing percent sequestration (% vessels with parasite sequestration) in the cortex, diencephalon and brainstem with and without VEGF-labelled parasites. In all but one case, VEGF+ parasites are observed in sections with high sequestration rates. D, A large vessel containing numerous intravascular leucocytes. VEGF-labelled parasites appear cytoadhered to the endothelium. Interaction between the endothelium, parasites and monocytes can be observed (see insert, 317% enlargement). Light labelling of the endothelium can be observed in several segments of the vessel (arrows). E, A large vessel without intravascular parasites shows strong reactivity for VEGF on endothelial cells and pericytes. The serum is also immunolabelled for VEGF. A smaller vessel in the bottom left corner of the image shows VEGF labelling on perivascular glia (arrowheads). VEGF+ neurons can also be observed (*).

Figure 3

Vascular endothelial growth factor (VEGF), VEGF receptor 1 and activated VEGF receptor 2 (KDR) immunoreactivity in post mortem brain tissue of severe malaria cases and controls. A, A vessel with endothelial cells expressing VEGF (arrows). B, Endothelial cells, perivascular cells and neurons labelled for VEGF receptor 1. C, Neuronal and glial labelling for activated KDR. D–F, Different paravascular patterns of immunoreactivity for activated KDR: diffuse (D); following white matter tracts (E); glial-associated (arrows, F). G–I, Activated KDR reactivity associated with axonal fibre tracts: longitudinal section showing peri-axonal glial reactivity (arrows) in a severe malaria case (G); longitudinal section showing axonal reactivity (arrows) in a case with cerebral infarction (H); transverse section showing axonal reactivity (white arrows) in close proximity to a vessel also showing KDR reactivity of endothelial cells (black arrows) and pericyte/smooth muscle cells in a severe malaria case.