Host matrix metalloproteinases in cerebral malaria: new kids on the block against blood-brain barrier integrity?

Abstract

Cerebral malaria (CM) is a life-threatening complication of falciparum malaria, associated with high mortality rates, as well as neurological impairment in surviving patients. Despite disease severity, the etiology of CM remains elusive. Interestingly, although the Plasmodium parasite is sequestered in cerebral microvessels, it does not enter the brain parenchyma: so how does Plasmodium induce neuronal dysfunction? Several independent research groups have suggested a mechanism in which increased blood-brain barrier (BBB) permeability might allow toxic molecules from the parasite or the host to enter the brain. However, the reported severity of BBB damage in CM is variable depending on the model system, ranging from mild impairment to full BBB breakdown. Moreover, the factors responsible for increased BBB permeability are still unknown. Here we review the prevailing theories on CM pathophysiology and discuss new evidence from animal and human CM models implicating BBB damage. Finally, we will review the newly-described role of matrix metalloproteinases (MMPs) and BBB integrity. MMPs comprise a family of proteolytic enzymes involved in modulating inflammatory response, disrupting tight junctions, and degrading sub-endothelial basal lamina. As such, MMPs represent potential innovative drug targets for CM.

Figure 1

Most commonly accepted hypotheses for pathophysiological mechanisms underlying clinical progress towards cerebral malaria (CM). The diagram summarizes the three distinct hypotheses on CM etiology and their typical features: i) the mechanical hypothesis is associated with iRBC cytoadherence and their reduced deformability, causing following anemia, rosette formation and microvascular obstruction; ii) the permeability hypothesis is based on BBB impairment and subsequent increase in vascular permeability, allowing toxic compounds to reach the brain parenchyma and causing neurological dysfunction; iii) the humoral hypothesis focuses on the enhanced production by the host of pro-inflammatory molecules, including cytokines and chemokines, and other soluble factors such as ROS, which are putatively responsible for inflammation, fever and coma during CM.

Figure 2

Blood–brain barrier structure: cerebral microvascular inter-endothelial junctions (adherens and tight junctions) and cell-matrix adhesion molecules. Diagram showing the structures of CNS inter-endothelial junctions, including adherens junctions and tight junctions, and of cell-matrix adhesion complexes, including talin, filamin, tensin or α-actinin filaments associated with integrins in the extracellular matrix. The core of adherens junctions results after the interactions among transmembrane glycoproteins, such as VE-cadherin, whose cytoplasmic face is linked to the catenin family members, including p120-catenin, β-catenin, and α-catenin. Tight junctions are composed of a branching network of sealing strands, each of which is formed by extracellular domains of transmembrane proteins, claudins and occludin, joining one another directly. These transmembrane proteins associate with different peripheral membrane proteins such as ZO-1 located on the intracellular side of plasma membrane, anchoring the strands to the actin component of the cytoskeleton.

Figure 3

Nomenclature and structure of mammalian matrix metalloproteinases(MMPs). The mammalian MMP family encompasses 25 members, categorized by different numbers (standard MMP nomenclature) or named depending on their matrix substrates (enzyme nomenclature). Each MMP displays some conserved structural domains, including: i) an N-terminal signal peptide required for secretion; ii) a cleavable pro-domain maintaining enzymatic latency; iii) a catalytic and Zn-binding domain; and iv) a C-terminal hemopexin domain. Optional MMP motifs include a fibronectin-type domain, a vitronectin motif, a furine cleavage site, three head-to-tail cysteine-rich repeats, and (for MT-MMPs) a C-terminal transmembrane domain or GPI anchor occasionally associated with a cytoplasmic domain.

Figure 4

Principal mechanisms underlying Hz-dependent dysregulation of matrix metalloproteinase-9 (MMP-9) and related molecules in human monocytes. The most relevant Hz-dependent mechanisms underlying aberrant MMP-9 function in malaria are based on current evidence from in vitro models of cultured human monocytes. After Hz phagocytosis, some lipoperoxidation products generated by Hz autocatalysis such as 15-HETE can promote early and late activation of PKC and p38 MAPK. These kinases have been associated with cytosolic I-kBα phosphorylation and degradation, resulting in subsequent nuclear translocation of NF-kB p50 and p65 subunits. Consequently, the transcription and protein expression of several pro-inflammatory molecules including TNF-α, IL-1β, and MIP-1α, and of some proteolytic enzymes or inhibitors such as lysozyme, pro-MMP-9, and TIMP-1 is enhanced. p38 MAPK can also promote monocyte degranulation, releasing pro-MMP-9, TIMP-1 and lysozyme into the extracellular environment. After secretion of these molecules, several proteolytic events can occur. Active MMP-9, generated from MMP-3 processing of the pro-enzyme, can further modulate TNF-α shedding from cell membrane in a similar manner as TACE, whereas ICE activates IL-1β after cleaving its pro-peptide. Soluble TNF-α, IL-1β, and MIP-1α have been shown to play a key role in mediating Hz effects on MMP-9, lysozyme, and TIMP-1 production, possibly generating some auto-enhancing loops. Hz-enhanced MMP-9 could favour CM development through complementary proteolytic activities (see Figure 5). On the other hand, TIMP-1 is primarily referred to as a MMP-9 inhibitor, thus TIMP-1 Hz-enhanced levels could supposedly be protective. However, several MMP-independent functions such as inhibition of cell apoptosis and growth have been recently described for TIMP-1. Thus, Hz-enhanced TIMP-1 protein may play a role in prolonged survival of impaired Hz-fed monocytes, in their altered maturation to dendritic cells and in their reduced ability to coordinate erythropoiesis. Finally, enhanced plasma levels of human lysozyme have been depicted as a risk factor for severe malaria.

Figure 5

Multiple putative roles of matrix metalloproteinases (MMPs) in cerebral malaria (CM) according to their biochemical functions. MMPs could play an active role during CM development through several complementary mechanisms: i) by disrupting endothelial tight junctions after protein degradation of ZO-1, claudins and occludin, thus causing an increased BBB permeability; ii) by promoting TNF-α shedding, IL-1β activation and CXCL8/IL-8 potentiation after proteolytic cleavage of their pro-domains, therefore inducing exacerbated pro-inflammatory response; iii) by processing some CM-associated hemostatic factors such as tpA, upA, and PAI-1, thus increasing the risk for thrombotic events.