The authors have declared that no competing interests exist.
Conceived and designed the experiments: FE GERG VC JW NHH. Performed the experiments: FE JW. Analyzed the data: FE GERG VC. Contributed reagents/materials/analysis tools: NHH. Wrote the paper: FE.
In patients with cerebral malaria (CM), higher levels of cell-specific microparticles (MP) correlate with the presence of neurological symptoms. MP are submicron plasma membrane-derived vesicles that express antigens of their cell of origin and phosphatidylserine (PS) on their surface, facilitating their role in coagulation, inflammation and cell adhesion. In this study, the
Cerebral malaria (CM) is a potentially fatal neurological syndrome characterised by unrousable coma. Since the detection of high levels of plasma microparticles (MP) in patients with CM, it has been demonstrated that inhibition of MP production confers protection from murine CM. However, the precise mechanisms of action of these MP during CM have not been completely deciphered. In this study, we used experimental models of CM to measure the production and origins of MP over the course of infection. We found low baseline circulating MP in healthy mice and these were subsequently raised at the time of the neurological syndrome. Phenotypic analyses showed that circulating MP were predominantly from activated host cells that have previously been established to participate in CM pathogenesis. We show for the first time transferred MP impairing endothelial integrity and inducing CM-like pathology in the brain and lung of healthy animals. Our study dissects what tissues these MP localise to exert their effects, as little is known about their fate following the initial release. These data suggest a causal relationship between MP and the development of CM and also warrant further investigation into the representation of MP as a marker of CM risk.
Cell activation by various agonists and apoptosis trigger the vesiculation of microparticles (MP) from all cell types
Once described as inert biological bystanders MP have now emerged as novel therapeutic targets in the treatment of diseases
CM is a multisystem multi-organ dysfunction that develops as a syndrome following
Little is understood about the role of MP in CM pathogenesis, although markedly high plasma levels of circulating platelet, erythrocytic, leucocytic and in particular endothelial cell-derived MP (PMP, EryMP, LMP, EMP respectively) have been detected in patients with CM
Using murine experimental models of CM
In our study, PbK infection at the non-encephalitogenic dose (PbK) was used to model NCM, PbK infection at the encephalitogenic dose (PbK1/2) and PbA infection (PbA) to model CM (
PbK-infection at the non-encephalitogenic dose was used to model NCM (closed triangle), PbK-infection at the encephalitogenic dose (PbK1/2, open triangle) and PbA-infection to model CM (closed square) (n = 10). (
We collected PFP following
(
Plasma MP numbers were compared at the time of CM onset (day 6–8) between the groups of infected mice, i.e. PbK-, PbK1/2- and PbA (
To evaluate whether
PbK (closed triangle), PbK1/2 (open triangle) or PbA (closed square) (
We studied the levels of cell-specific MP in the plasma at the time of CM onset (
(
Fluorescently labeled MP, purified from the plasma of healthy and PbA infected donor mice, were adoptively transferred into healthy or PbA infected recipient mice. The presence of these MP was assessed in the blood of recipient mice via flow cytometry over 60 minutes post injection. MP from healthy donors were detectable in significantly lower numbers than those from PbA infected donors. Injecting MP from PbA infected donors induced an acute rise of MP in recipient mice (mean ± SEM MP/µL, 1018.5±733.49 healthy recipient, 808.23±329.75 PbA recipient) that cleared within 2 minutes following injection (
(
Brain smears were prepared from all recipient mice. Fluorescence microscopic analysis showed that, after transfer, only MP from PbA infected mice showing signs of CM (CM+) were found within the vessels of the brain of CM+ recipient mice, but this was not evident in NCM mice. MP were found to be lodged along the endothelium within the lumen and at the bifurcation of some but not all microvessels. When the recipient mice were healthy, no MP could be visualised within the cerebral microvessels (
Donor | Recipient | MP present | |
healthy | healthy | + | |
healthy | PbA | + | |
PbA | healthy | ++ | |
PbA | PbA | +++ | |
healthy | healthy | − | |
healthy | PbA | − | |
PbA | healthy | − | |
PbA | PbA | + | |
healthy | healthy | − | |
healthy | PbA | − | |
PbA | healthy | − | |
PbA | PbA | + | |
healthy | healthy | − | |
healthy | PbA | − | |
PbA | healthy | − | |
PbA | PbA | +/− | |
healthy | healthy | − | |
healthy | PbA | − | |
PbA | healthy | − | |
PbA | PbA | +/− | |
healthy | healthy | − | |
healthy | PbA | − | |
PbA | healthy | − | |
PbA | PbA | − |
To determine the major route of clearance for the adoptively transferred MP we imaged the spleen, kidney, liver, lung and heart of recipient mice. Observations made on all combinations MP-donor/MP-recipient and the presence of MP within these organs are represented in
MP from both healthy and PbA infected donors were present in the red pulp of the spleen of all recipient mice (
EMP purified from resting or TNF-stimulated MVECs
(
To identify possible markers that could be facilitating the interaction of EMP at the site of the lesion, the surface phenotype of EMP was determined. We found that CD54 and CD106 (
This study investigates the levels and cellular origin of MP produced during murine malaria and for the first time describes the clearance and fate of CM+ MP
Ablating the increase of MP numbers either genetically
Elevated levels of circulating Annexin V+ MP were detected in the plasma of CM+ mice (i.e. infected with PbA or PbK1/2) at the time of neurological development. This finding confirms what has already been observed upon CM onset in experimental and hCM
At the time of CM onset, cell-specific MP numbers were higher in PbA infected mice than in PbK- and PbK1/2 infected animals or healthy controls. The sum of positive MP for single staining of cell markers gave a closer approximation of total circulating MP, as Annexin V staining indicates only PS-positive MP. MP can be PS negative or have low undetectable PS on the surface and remain unbound to Annexin V
Our data show that during the acute stage of PbA-infection, a peak of PMP can be detected, and this is absent in the PbK- and PbK1/2-infections. This finding is consistent with the substantial loss of platelets as MP in the acute phase of the PbA-infection
EryMP were elevated in CM+ mice, both in PbA- and PbK1/2 infected mice, at CM onset. This finding is supported by human studies, whereby EryMP numbers are increased in patients with
Adoptively transferred CM+ MP are detectable in recipient mice but quickly subsided, indicating clearance from circulation. Some of these MP were found to be arrested in cerebral microvessels of CM+ recipients and also in the spleen, kidney and, to a lesser extent, the lung and liver. Recent studies in humans have shown that parasites induce the loss of endothelial protein C receptors in the cerebral microvessels, leaving them vulnerable to enhanced local thrombin generation and coagulopathy
Little is known on the mechanisms underlying the clearance of plasma MP from circulation
All cells are able to produce and release MP, although the emerging progeny of MP are heterogeneous and do not share the same properties. EMP represent the most abundant MP detected in pathologies that arise due to vascular injury or endothelial dysfunction although not all their roles are noxious
Exactly how EMP alter endothelial integrity in CM is unclear. Flow cytometry revealed that our EMP express endoglin, CD54 and CD106. Soluble endoglin (sCD105) overexpression is linked to typical systemic and vascular inflammation states such as pre-eclampsia and HELLP syndrome
Transferred EMP induced atelectasis in lungs from healthy recipient mice and increased alveolar cellularity, resembling the pathology seen in CM+ lung. EMP were shown to induce endothelial dysfunction, promote vasodilation, pulmonary oedema and acute lung injury in pathophysiological concentrations
Taken together, our findings offer new evidence for a causal relationship between MP and the pathogenesis of CM. To our knowledge, this is the first time that MP have been localised at the neurovascular lesion
During CM, activation of cells by parasite moieties, toxins, cytokines and/or cell death, delivers MP into the circulation. We propose that the interactions between MP, endothelium, circulating host vascular cells and their released circulating soluble factors influence the course of infection leading to the development of CM. Since the first human studies on MP in CM
In addition, molecules such as Pantethine that inhibit MP release have conferred protection
Infections were performed as previously described
Seven to 8 weeks old female CBA mice (Animal Resource Centre, Perth) were housed under pathogen-free conditions. These mice are susceptible to
Parasitaemia was determined from thin tail blood smears on day 4 p.i and every second day until end point, using light microscopy and Diff-Quick staining. CM was diagnosed if an infected mouse presented with ruffled fur, severe motor impairment (ataxia, hemiplegia or paraplegia) or convulsions and was allocated a score of 3 or 4 as described previously
Mouse venous blood was collected by retro-orbital venepuncture under anaesthesia in 0.129 mol/L sodium citrate (ratio of blood to anticoagulant 4∶1). Samples were centrifuged at 1 500 g for 15 min at room temperature. Harvested supernatant was further centrifuged at 18 000 g for 4 min, twice, to achieve platelet-free plasma (PFP).
Total MP numbers were quantified by detection of PS using FITC-Annexin V (Beckman Coulter) labelling, which is Ca2+ dependent. The cellular origin of these MP was determined using cell-specific monoclonal antibodies as detailed in
Cell-type | Marker | Alternate name | Clone | Supplier | Concentration used |
Endothelial cell | CD105 | Endoglin | MJ7/18 | eBioscience | 5 µg/mL |
Erythrocyte | TER-119/CD235a | Erythroid cell marker | TER-119 | eBioscience | 2 µg/mL |
Leucocyte | CD45 | Leucocyte Common Antigen | 30-F11 | Becton Dickinson Pharmingen | 2 µg/mL |
Monocyte | CD11b | Integrin αM | M1/70 | eBioscience | 2 µg/mL |
Platelet | CD41 | Integrin α IIb chain | MWReg30 | Becton Dickinson Pharmingen | 5 µg/mL |
Intercellular Adhesion Molecule-1 | CD54 | ICAM-1 | YN1/1.7.4 | eBioscience | 10 µg/mL |
Vascular Cell Adhesion Molecule-1 | CD106 | VCAM-1 | 429 | eBioscience | 10 µg/mL |
Briefly, 20 µl of PFP were incubated with Annexin V-FITC diluted 1 : 2 in 10× binding buffer (BB) or antibodies for 30 min. Following incubation, 20 µL of Flow-count™ Fluorospheres (1000/µL) (Beckman Coulter) were added to each sample to act as a calibrated internal standard of known size and concentration. Samples were resuspended in 200 µL of 1× BB and analysed on a Beckman-Coulter FC500-MPL flow cytometer. Data were acquired for 60 s and analysed using CXP analysis software (Beckman Coulter). MP were first discriminated based on their size (<1 µm) on a log-forward light scatter and log-side light scatter (FSC-SSC) dot plot and then for their positivity for binding of either Annexin V or specific antibodies (
Blood was collected by venepuncture of the retro-orbital plexus in sterile citrated tubes from healthy and PbA infected mice displaying full blown syndrome. In order to maximise the number of MP purified without changing the phenotype of MP produced by the blood cells and to minimise the number of donor mice, whole blood was activated using calcium (Ca2+)-ionophore (Calcimycin A23187 2 mmol/L, SIGMA) vortexed and incubated at 37°C for 40 min. Calcium ionophore activation of whole blood increases the number of MP by two – three fold. Blood was processed for PFP as mentioned above. To obtain a purified population of MP devoid of blood proteins and calcium ionophore, PFP was further centrifuged at 18 000 g for 1 h at 15°C. Supernatant was used as a MP-free control after checking by flow cytometry. The MP pellet was gently resuspended in Diluent C (Sigma), prior to labelling using PKH67 Green Fluorescent Cell Linker Kit for General Cell Membrane Labelling (SIGMA). Briefly, under dark sterile conditions 1 µL of PKH67 was added to 250 µL of Diluent C and then added to 250 µL of MP suspension. Following continuous pipetting for 1 min, the suspension was incubated in the dark for 4 min. Labelling was stopped by adding 2 mL of foetal bovine serum (FBS) and 10 mL of RMPI-1640 containing 10% FBS (Gibco). The suspension was centrifuged at 18 000 g for 1 h at 15°C to pellet MP. PKH67+MP were resuspended in PBS and numbers were calculated after flow cytometry analysis.
PKH67+MP suspensions and MP-free supernatant were injected intravenously into healthy and PbA infected recipient mice 6 days p.i. Briefly, mice were anesthetised and received 400×103 MP in 200 µL of PBS. Mice were allowed to recover. Blood was collected via tail vein at selected time points (1, 2, 3, 4, 5, 10, 20, 30, 40, 50 and 60 min) and analysed by flow cytometry prior to euthanasia and subsequent collection of organs. Briefly, 5 µL of mixed blood and citrate and 5 µL of Flow-count fluorospheres (Beckman Coulter) were resuspended in 200 µL of PBS. MP present in the samples were counted based on their size and PKH67 labelling.
Following euthanasia, brains were collected and cut along the sagittal plane. By placing a small (1 mm×1 mm) section of fresh brain between two glass microscope slides and pressing these together, brain smears were created. Smears were allowed to air dry completely prior to fixation and subsequent labelling. The rest of the brain, together with lung, spleen, liver, kidney and heart tissue, were placed in cryoprotective embedding medium (OCT) and snap frozen in hexane cooled with liquid nitrogen. Tissue was then cut into 5 µm sections. Sections and brain smears were fixed in precooled (30 min at −20°C) absolute ethanol and acetone (3 : 1) for 10 min. Following blocking with filtered 5% (w/v) bovine serum albumin in PBS for 20 min, slides were incubated with Texas Red labeled Lycopersicon Esculentum (Tomato) Lectin (LEL, TL, Vector Laboratories) for 45 min at room temperature. Slides were washed and stained with 4′, 6-diamidino-2-phenylindole (DAPI) fluorescent stain (Invitrogen) and mounted in Fluoromount-G (Southern Biotech). Images were obtained using an Olympus IX71 inverted microscope and also the confocal microscope Olympus FV1000, as noted.
EMP were generated from mouse brain microvascular endothelial cells (B3 cell line)
Recipient mice were separated into two groups, healthy and PbA infected. PbA infected mice were infected 5 days prior to the EMP purifications to allow MP and mice to be ready at the same time. These groups were further divided to account for the experimental conditions (
Formalin fixed, paraffin embedded brain and lung were Haematoxylin-Eosin stained. Slides were imaged at 100× magnification using an Olympus IX71 inverted microscope. Qualitative assessment of the tissue was performed by two independent researchers using parameters stipulated in the histopathological scale (
Score | ||||||
Tissue | Parameter | 0 | 1 | 2 | 3 | 4 |
Sequestration | None present | * | ** | *** | **** | |
PVS | No changes | * | ** | *** | **** | |
Haemorrhage | None present | * | ** | *** | **** | |
Cellularity | None present | * | ** | *** | **** | |
Atelectasis | None present | * | ** | *** | **** | |
Plasma | None present | * | ** | *** | **** | |
Haemorrhage | None present | * | ** | *** | **** |
Data were analysed using GraphPad Prism version 5.00 for Windows, GraphPad Software, San Diego California USA. Survival curves were analysed using the Log-rank (Mantel-Cox) Test and the Gehan-Breslow-Wilcoxon Test. To compare several groups, we used non-parametric analysis of variance (ANOVA, Kruskall-Wallis) with a Dunn's post-test. To compare mean total and cell-specific MP levels between two groups the Wilcoxon test was used; *p<0.05, **p<0.001, ***p<0.0001.
(PDF)