Elimination of intravascular thrombi prevents early mortality and reduces gliosis in hyper-inflammatory experimental cerebral malaria

Elimination of intravascular thrombi prevents early mortality and reduces gliosis in... Background: Cerebral malaria (CM) is the most lethal outcome of Plasmodium infection. There are clear correlations between expression of inflammatory cytokines, severe coagulopathies, and mortality in human CM. However, the mechanisms intertwining the coagulation and inflammation pathways, and their roles in CM, are only beginning to be understood. In mice with T cells deficient in the regulatory cytokine IL-10 (IL-10 KO), infection with Plasmodium chabaudi leads to a hyper-inflammatory response and lethal outcome that can be prevented by anti-TNF treatment. However, inflammatory T cells are adherent within the vasculature and not present in the brain parenchyma, suggesting a novel form of cerebral inflammation. We have previously documented behavioral dysfunction and microglial activation in infected IL-10 KO animals suggestive of neurological involvement driven by inflammation. In order to understand the relationship of intravascular inflammation to parenchymal dysfunction, we studied the congestion of vessels with leukocytes and fibrin(ogen) and the relationship of glial cell activation to congested vessels in the brains of P. chabaudi-infected IL-10 KO mice. Methods: Using immunofluorescence microscopy, we describe severe thrombotic congestion in these animals. We stained for immune cell surface markers (CD45, CD11b, CD4), fibrin(ogen), microglia (Iba-1), and astrocytes (GFAP) in the brain at the peak of behavioral symptoms. Finally, we investigated the roles of inflammatory cytokine tumor necrosis factor (TNF) and coagulation on the pathology observed using neutralizing antibodies and low-molecular weight heparin to inhibit both inflammation and coagulation, respectively. Results: Many blood vessels in the brain were congested with thrombi containing adherent leukocytes, including CD4 T cells and monocytes. Despite containment of the pathogen and leukocytes within the vasculature, activated microglia and astrocytes were prevalent in the parenchyma, particularly clustered near vessels with thrombi. Neutralization of TNF, or the coagulation cascade, significantly reduced both thrombus formation and gliosis in P. chabaudi-infected IL-10 KO mice. (Continued on next page) * Correspondence: rostephe@utmb.edu Department of Microbiology and Immunology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555, USA Department of Internal Medicine, Division of Infectious Diseases, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0435, USA Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 2 of 17 (Continued from previous page) Conclusions: These findings support the contribution of cytokines, coagulation, and leukocytes within the brain vasculature to neuropathology in malaria infection. Strikingly, localization of inflammatory leukocytes within intravascular clots suggests a mechanism for interaction between the two cascades by which cytokines drive local inflammation without considerable cellular infiltration into the brain parenchyma. Keywords: Malaria, Brain, Inflammation, Neuropathology, Vascular congestion, Monocyte, Astrocyte, Microglia, Immunofluorescence, Confocal microscopy, Background significant enrichment [18]. The P. chabaudi life cycle is With 212 million new cases and 429,000 estimated synchronous. Mature schizonts disappear from the cir- deaths in 2015, malaria remains one of the most eco- culation almost completely and are found sequestered nomically impactful infectious diseases worldwide [1]. A primarily in the liver and lungs of mice in a partially small percentage of Plasmodium falciparum infections ICAM1-dependent manner [19]. Interestingly, patho- results in severe malarial disease. However, a significant logical damage within each organ in P. chabaudi does proportion of severe malaria infections includes cerebral not correspond to the degree of organ-specific seques- malaria (CM), which is a leading cause of death in tration of the parasite [18]. Sequestration is a hallmark sub-Saharan African children and represents a major of autopsy in fatal P. falciparum-induced CM cases [20, burden worldwide [2]. CM accounts for an estimated 21], and specific parasite variants are associated with se- 500,000 cases per year and correlates with high parasite- vere malaria [22–25]; however, it is challenging to defini- mic burden, severe inflammation, and cerebral edema tively prove that parasite sequestration in the brain is [2]. Furthermore, about 20% of patients with CM die causal to CM. despite timely treatment [3], and neurological sequelae Activated immune cells and pro-inflammatory cyto- in survivors is common [4]. Several host genetic factors kines are also strongly implicated in the mortality in hu- have been implicated in pathology. For example, muta- man disease [26, 27]. A low ratio of IL-10 to TNF in tions in the promoters of the inflammatory cytokine patients predicts more severe malaria, as do mutations tumor necrosis factor (TNF), which drives the in the IL-10 and TNF genes [28, 29]. Mouse models anti-malaria response of phagocytes, and the regulatory show that this is because IL-10 is required to protect an- cytokine IL-10, which protects the host from excessive imals from lethal pathology, as it regulates the immunopathology, have been correlated with severe dis- pro-inflammatory cytokines IL-12 and TNF [30], which ease in both mice and humans [5–10]. However, inflam- drive as yet poorly defined neuroimmunopathology. matory cytokines also allow parasite sequestration and IL-10 KO mice lacking IFN-γ receptor signaling are also leukocyte adhesion by upregulating adhesion molecules rescued from mortality, even though they exhibit higher on the vascular endothelium [11–13]. levels of parasitemia [31]. IL-10 is primarily made by + + The role of inflammatory cytokines increased by the CD4 IFN-γ effector T cells (Teff) in P. chabaudi infec- absence of IL-10 has been studied extensively in the tion, not Tregs, and is downstream of IL-27 [32, 33], Plasmodium chabaudi mouse model of severe malaria and we have shown that CD4 Teff are found solely [14]. P. chabaudi is a rodent parasite that leads to mild within the cerebral vasculature, not in the brain paren- malaria in C57BL/6 (WT) mice. However, in chyma [16]. IL-10-deficient (IL-10 KO) mice, P. chabaudi infection While there are studies of host genetics and those cor- leads to hyper-inflammation and death. The syndrome relating systemic inflammatory cytokines with poor out- includes increased levels of the pro-inflammatory cyto- comes in severe malaria [26, 27], no significant kines TNF and IFN-γ [14] and lethal disease character- inflammatory infiltrate within the brain parenchyma has ized by cerebral pathology including cerebral edema and been documented in human or mouse studies of the dis- hemorrhage [15]. In addition, we have recently demon- ease [20, 21, 34–40]. As a result, the contribution of ac- strated pathological behavioral phenotypes indicative of tivated peripheral leukocytes to brain pathology has neurological and cognitive dysfunction in this model been poorly appreciated. Interestingly, despite the lack [16]. Strikingly, there is no significant parasite sequestra- of infiltrating inflammatory cells in the brain paren- tion in the brains of these mice. While a few parasites chyma, we have documented increased microglial activa- have been detected in the brain vasculature via electron tion in this model [16]. This is intriguing because glia microscopy [17], a more recent examination of the brain are found behind the multi-layered blood-brain barrier using highly sensitive luminescence technology to detect (BBB), while activated peripheral immune cells are luciferin-expressing P. chabaudi parasites did not show within the vasculature [16]. This prompted the question Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 3 of 17 of how the inflammatory cells within the vasculature coagulation in vascular congestion in CM and also impli- could amplify cytokine production in the absence of a cate a novel mechanism of inflammation-induced neuro- lymphoid structure, such as that developing in neuroim- pathology possibly initiated by leukocytes contained munopathologies with parenchymal infiltrates. within the vasculature. These findings may be relevant be- Congestion of the brain and retinal vasculature has been cause the inflammation-driven neuropathology in this documented in human cerebral malaria and is associated model shares many features with human cerebral malaria, with poor prognoses in human cases of CM [41, 42]. Sev- including intravascular leukocytes and thrombi, systemic eral factors are likely to contribute to congestion in human hyper-inflammation, edema, and death. patients: parasite sequestration, leukocyte adhesion, and co- agulation defects. Parasite-infected erythrocytes can both Methods bind to the vascular endothelium, leading to activation and Mice tm1Cgn vascular dysfunction, and activate the coagulation cascade C57BL/6J (WT) and B6.129P2-Il10 /J (IL-10 KO) [43, 44]. Coagulation defects are also seen in both murine mice (Jackson Laboratory, Bar Harbor, ME) were bred in experimental cerebral malaria and in human cerebral mal- The University of Texas Medical Branch Animal Re- aria [45–47] and can be promoted by the parasite itself source Center. Experimental mice were female and be- [45]. Vascular thrombi were observed in CM2 patients in tween 6 and 12 weeks of age at the time of infection. All Malawi, who are documented to have both sequestration animals were kept in a specific pathogen-free housing and cerebral hemorrhages [20]. This supports the finding with ad libitum access to food and water. Animals were that disseminated intravascular coagulation (DIC) was ob- cared for according to the Guide for the Care and Use of served in 19% of CM patients and correlated with poor Laboratory Animals under Institutional Animal Care outcomes [48]. However, the role of coagulation in neuro- and Use Committee-approved protocols. UTMB Animal pathology is obscured by contradictory outcomes in stud- Resource Center facilities operate in compliance with ies of the effect of the anticoagulant, heparin [49, 50]. In the USDA Animal Welfare Act, the Guide for the Care clinical trials, heparin significantly reduced death in a clin- and Use of Laboratory Animals, under OLAW accredit- ical trial in children with CM in Indonesia (from 13/17 to ation, and IACUC-approved protocols. 2/16, [50]) and reduced patient’s coma and hospitalization time [49]. However, it is not currently recommended for Parasite and infection treatment due to the potential of systemic hemorrhagic Frozen stocks of Plasmodium chabaudi chabaudi side effects of this older drug, suggested by work in (AS)-infected RBCs (iRBCs) (Jean Langhorne, Francis non-human primates [51] and case studies of malarious Crick Institute, London, UK) kept at − 80 °C were soldiers in Asia with pulmonary involvement [52], though thawed and injected intraperitoneally (i.p.) into WT not seen in clinical trials. The presence of monocytes and mice. Parasitized blood from these animals was diluted T cells in the brain vasculature [20], but not in the brain in Krebs-Ringer bicarbonate buffer (Sigma-Aldrich, St. parenchyma [34], is also documented. This has often been Louis, MO) and normal saline to deliver 10 iRBCs i.p. interpreted as a “lack of inflammation,” despite strong evi- in 200 μl into experimental WT or IL-10 KO mice. Thin dence, both genetic and serological, that cytokines play a blood smears were collected at regular intervals to critical role in killing parasite and inducing pathology [53]. monitor for peripheral parasitemia by staining with In an attempt to understand the role of adherent intra- Diff-Quik (Siemens Healthcare Diagnostics, Newark, vascular leukocytes and coagulation in promoting neur- DE) or Giemsa stain (Ricca Chemical Company, Arling- onal malfunction, we investigated the contents of ton, TX) and counted on a light microscope. congested vessels and their effects on the brain paren- chyma, as measured by gliosis. Furthermore, we tested Animal body temperature and weight the role of coagulation in pathology by studying the ef- Internal body temperatures were assessed daily during in- fect of anticoagulants on mortality and histological fea- fection using rounded stainless steel rectal probes and a tures of inflammation-driven neuropathology in P. BIO-TK8851 digital rodent model thermometer (Bioseb, chabaudi infection of IL-10 KO mice. We found that Pinellas Park, FL). Probes were sanitized with CaviCide thrombi were prevalent throughout the brain and coin- (Metrex Research Corp., Romulus, MI) between each use. cide with localization of adherent leukocytes. In Animal weights were measured using an OHAUS Scout addition, areas of coagulation and leukocytes Pro SP601 portable balance (OHAUS, Parsippany, NJ). co-localized with parenchymal gliosis. We also found a striking reduction of mortality and a significantly recov- Animal behavior evaluation ered parenchymal histology on elimination of coagula- Beginning on day 5 post-infection, daily assessments tion suggesting a pathological role for thrombi in this were performed on all animals using an abbreviated ver- model. These observations suggest an important role of sion of the modified SmithKline Beecham, Harwell, Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 4 of 17 Imperial College, Royal London Hospital Phenotype As- MA), or rabbit (catalog no. Z0334, Agilent Technologies, sessment (SHIRPA) protocol [54]. This brief behavioral Carpinteria, CA) anti-GFAP, mouse anti-CD11b biotin assessment was developed based on the full assessment (clone M1/70, catalog no. 13-0112-85, eBioscience, San in a previous study [16]. Higher scores were awarded for Diego, CA), and rat anti-CD45 biotin (clone 104, catalog measures showing higher functional ability. The proce- no. 13-0454-85, eBioscience, Sand Diego, CA). Secondary dures were carried out in an open testing environment antibodies used were goat anti-rat AlexaFluor-488 (catalog away from the home cage and took approximately 5 min no. A11006, Thermo Fisher Scientific, Waltham, MA) and per animal. goat anti-rabbit AlexaFluor-568 (catalog no. A11011, The abbreviated SHIRPA used involves a selection of Thermo Fisher Scientific, Waltham, MA). nine semi-quantitative tests for general health and sen- Streptavidin-FITC (catalog no. 11-4317-87, eBioscience, sory function, baseline behaviors, and neurological re- San Diego, CA) was used as a tertiary step for biotinylated flexes. We observed undisturbed behavior with the antibodies. CellTrace Violet (catalog no. C34557, Thermo mouse placed in an inverted beaker on top of a metal Fisher Scientific, Waltham, MA)-labeled CD4 T cells were grid suspended above the home cage for 3 min, during adoptively transferred into IL-10 KO mice for later which body position and spontaneous activity were co-localization with brain vasculature after i.v. perfusion assessed. Body position scores ranged from 0 (com- with DyLight488-labeled tomato lectin (catalog no. pletely flat) to 5 (repeated vertical leaping). Spontaneous DL-1174, Vector Laboratories, Burlingame, CA). Images of activity scores ranged from 0 (none) to 4 (rapid/dart immunohistochemistry (IHC) sections were taken with an movement). At the end of the observation period, palpe- Olympus IX 71 inverted brightfield microscope using a × bral closure, which was scored from 0 (eyes closed) to 2 20 air objective, while the immunofluorescence images (eyes wide open), and qualitative grip strength, scored were taken with a confocal microscope (Olympus FV 1000) from 0 (none) to 4 (unusually strong), are tested by ap- with the DAPI channel for nuclei, Alexa 488 channel for plying a gentle horizontal force on the animal’s tail as it Iba1 tagged with Alexa 488, and Alexa 647 channel for CD grips the metal grid. The animal is then placed in an 31 tagged with Alexa 647. IHC images of Iba1-stained sec- open arena in which several behaviors are measured. tions were contrast-enhanced and segmented by threshold Gait is observed as the animal traverses the arena and is for microglia using ImageJ (NIH, Version 1.48u). These scored from 0 (incapacity) to 3 (normal). During move- were used to create binary images. Individual microglia ment, tail elevation is scored, ranging from 0 (dragging) were identified using a semi-automatic algorithm employ- to 2 (elevated). Touch escape measures the reaction to a ing the particle analysis function on image and average area finger stroke and is scored from 0 (no response) to 3 (es- per microglia; the microglia density and total immunoreac- cape response to approach). Palpation of the animal’s tive area were calculated from the binary images. Area frac- sternum determines heart rate: 0 (slow) to 2 (fast), and tion of small processes is a ratio of immunoreactive area finally, righting reflex is scored by releasing the animal without microglia to total immunoreactive area which from an upside-down position near the surface and ob- indicates the degree of ramification. Transformation index, serving the responding effort to upright itself, scored and indicator of activation, was calculated as T-Index = from 0 (fails to right) to 3 (lands on feet). The expected (Perimeter )/(4π × Area) per microglia. To quantitatively score of a healthy, uninfected IL-10 KO or WT mouse is describe the degree of ramification, we calculated the area 22. A score of 15 was identified as the humane endpoint fraction of small thin processes to total immunoreactive based on the finding that any female animal that drops area. Ramification could be seen in IHC images as glia below that score by day 9 will succumb to infection (see with long and thin processes that appeared segmented due Additional file 1: Figure S1). to branching in and out of the tissue section plane. The astrocyte-thrombus association index was defined in which Histochemistry the ratio of X (the number of astrocytes contacting a Immunofluorescence of cryosections was examined after thrombus divided by the total number of thrombi) was cal- 48 h of post-fixation of mouse brains in 4% PFA and 72 h culated, and values were normalized based on the following of cryoprotection in 30% sucrose. Fixed frozen sagittal sec- equation, (X − X )/(X − X ), where X =1.3 (lower i min max min min tions (30 μm) were made using Tissue Plus® Optimal Cut- limit of astrocyte-thrombi interaction seen in uninfected ting Temperature Compound (Fisher Healthcare, IL-10 KO brains) and X = 3.25 (~ 75% astrocyte/thrombi max Houston, TX) and mounted on glass slides with Fluoro- association) approximated the lower and upper limit of as- mount mounting medium (Novus Biologicals, Littleton, trocytes interacting with thrombi based on our data. CO). Sections were incubated overnight at 4 °C with pri- mary antibodies rabbit anti-fibrinogen (catalog no. A0080, Cell and in vivo labeling Agilent Technologies, Carpinteria, CA), rat (clone 2.2B10, Some infected IL-10 KO and WT animals were injected 6 + catalog no 13-0300, Thermo Fisher Scientific, Waltham, with 2 × 10 CTV CD4 T cells 3.5 h before sacrifice Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 5 of 17 (i.p.) and 40 μg of DyLight488 labeled Lycopersicon escu- between two #1.5 cover glass. To visualize large regions of lentum (tomato) Lectin (catalog no. DL-1174, Vector La- optically cleared brain tissue using two-photon microscopy, boratories, Burlingame, CA) 20 min before sacrifice image stack mosaic and stitching were applied. Image stack (i.v.). CellTrace Violet (catalog no. C34557, Thermo stitching was done with a 10% overlap on a field of view of Fisher Scientific, Waltham, MA) labeling was performed 2327.3 × 237.3 μm providing 232.73 μm of co-registration as previously described [55]. in X and Y coordinates. Images were analyzed using ImageJ (FIJI), Olympus Fluoview FV1000-ASW 2.0 Viewer (con- Anti-TNF antibody treatment focal), Imaris Image Analysis Software (confocal and Mice receiving anti-TNF antibody (clone XT3.11, Bio X two-photon microscopy; Bitplane USA, Concord, MA), and Cell, West Lebanon, NH) were treated with 0.2 μg/day NIS Elements (confocal; Nikon Instruments, Melville, NY). for 5 days starting on day 5 post-infection (days 5–9). Positive fibrinogen and elevated GFAP staining in each field Untreated mice received isotype rat IgG1 as a control. was quantified by applying a signal intensity threshold and the percent area covered was calculated via the outlined CLARITY and optical clearing areas of positive staining that met the signal intensity Fixed brain sections (IL-10 KO and WT) were subjected threshold per field of view. The percentage of total area in- to the passive CLARITY optical clearing method [56] for cluded was calculated using ImageJ software (FIJI, NIH). large-scale labeling and imaging. In brief, mice were anesthetized and perfused transcardially with a mixture Ammonia assay of 4% (wt/vol) PFA, 4% (wt/vol) acrylamide, 0.05% (wt/ Tissue and serum ammonia was quantified using a com- vol) bis-acrylamide, and 0.25% (wt/vol) VA044 (hydrogel mercial colorimetric ammonia assay kit (ab83360, solution) in PBS. Brains were extracted and incubated in Abcam, Cambridge, MA). Briefly, brain and liver sam- hydrogel solution at 4 °C for 3 days. Solution ples were collected from infected IL-10 KO and WT temperature was then increased for 3 h to 37 °C to initi- mice at the peak of behavioral symptoms, washed in ate polymerization. Hydrogel-embedded brains were sec- cold PBS, resuspended in 100 μl assay buffer, and ho- tioned into 2-mm-thick sagittal sections and placed in mogenized using a Dounce homogenizer to produce clearing solution (sodium borate buffer, 200 mM, pH single-cell suspensions. After 2–5 min of centrifugation 8.5) containing 4% (wt/vol) SDS) for 3 weeks at 40 °C at 4 °C, cells were counted via hemocytometer and under gentle agitation. Samples were immunostained for seeded into a 96-well plate to provide 1–5×10 cells/ GFAP to assess astrogliosis. After immunostaining, sam- well. Serum samples were counted and seeded directly ples were optically cleared using increasing serial con- into plates without processing (5–10 μl/well). The col- centrations (10–100%) of 2,2′-thiodiethanol (TDE) in orimetric assay was conducted using OxiRed probe. Milli-Q water (EMD Millipore, Darmstadt, Germany) to Color change was recorded at OD 570 nm using a spec- achieve optimal refractive index matching with tissue. trophotometer microplate reader and compared to an ammonium chloride standard curve (detects 0–10 nmol/ Microscopy well) after 60 min of incubation at 37 °C. Fixed cryosections (30 μm thickness, fluorescent or con- focal microscopy) were imaged with a Nikon Eclipse 80i Statistics epifluorescence microscope and a Fluoview 1000MPE sys- Where indicated, groups were compared by t test (2 tem configured with an upright BX61 microscope (Olym- groups) or one-way ANOVA (3 or more groups), pus, Center Valley, PA). Fixed, CLARITY-processed followed by post hoc Bonferroni method or Tukey’s test sections (2 mm thickness, two-photon confocal micros- to identify significance between individual groups. Each copy) were imaged using a Prairie Ultima IV (Prairie point represents the average value per animal after ana- Technologies/Bruker, Middleton, WI) upright multipho- lysis of 10 fields, unless otherwise specified. Statistical ton microscope. For two-photon fluorescence microscopy, analysis was performed in Prism (GraphPad, La Jolla, a × 10 0.3 N.A. objective (UPLFL10X, Olympus) and a × CA), *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001. Error bars 25 1.05 N.A. super-objective (XLSLPLN25XGMP, Olym- represent ± SEM. pus) were used for image collection. Illumination for exci- tation of fluorescence was provided by a femtosecond Results laser (Mai Tai, SpectraPhysics, Santa Clara, CA) tuned to Congestion of brain blood vessels with thrombi + + + 800 nm. Fluorescence was collected using a two-photon containing CD45 , CD11b , and CD4 leukocytes in P. standard M filter set including filters with bandwidth 604 chabaudi-infected IL-10 KO mice ± 45 nm, a filter with bandwidth 525 ± 70 nm, and di- To investigate vascular abnormalities in P. chabaudi-in- chroic mirror cutoff at 575 nm. Samples were mounted fected IL-10 KO mice, we examined sagittal sections of on a 30-mm cage plate (CP06, ThorLabs, Newton, NJ) perfused and fixed brain tissue for evidence of vascular Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 6 of 17 leakage as indicated by extravascular fibrinogen at the systemic production of fibrinogen is a risk factor for co- peak of infection (day 8 post-infection). Brains from agulation, it does not lead to clotting by itself [58]. How- infection-matched, disease-resistant WT mice were used ever, an increase in liver fibrinogen production is not as controls (Fig. 1a). In addition to the expected sites of sufficient for accumulation of fibrin, which is triggered by perivascular fibrinogen (evidence of fibrinogen leakage), the coagulation cascade [57, 58]. we also found foci of fibrin(ogen) staining within the Studies of both human CM and murine experimental vascular lumen of brain blood vessels in IL-10 KO mice. cerebral malaria (ECM) have documented congestion of As we had performed transcardial perfusion prior to sac- the brain and retinal vasculature, but the role of thrombi in rifice, this data is suggestive of intravascular thrombi. reduced blood flow is not clear. By imaging through Quantification of fibrin(ogen) staining in the IL-10 KO 200 μm of tissue, we found that both large and small vessels mice showed an increase in the area of the brain with retain intravascular fibrin(ogen) (Fig. 2a), often to the point bright fibrinogen immunoreactivity (percent area of Alexa of complete occlusion of the vascular lumen (Fig. 2b), rem- Fluor 568 pixels, 10 fields/mouse) compared to infected iniscent of thrombosis. The coagulation cascade leads to WT, or uninfected, which were indistinguishable from cleavage of fibrinogen into fibrin during the formation of a each other (Fig. 1b). There was also a large increase in clot [59]. The polyclonal antiserum used to detect fibrino- staining of fibrinogen in the livers of infected IL-10 KO gen here also detects fibrin and other degradation products compared to WT, which had some lighter staining as well of fibrinogen [60, 61]. Therefore, we interpret this staining that was not quantifiable over background levels in unin- pattern to represent fibrin clots. The appearance of sherical fected mice (Fig. 1c). This could potentially be due to an gaps in fibrin staining led us to hypothesize that in addition increase in fibrinogen production by the IL-10 KO mouse to red blood cells and platelets, immune cells could be downstream of inflammation, as fibrinogen is an acute retained within the thrombi of congested vessels. In order phase response protein [57]. However, while increased to identify them, we stained IL-10 KO brains for the Fig. 1 IL-10 KO mice have residual fibrin deposition in and around brain vasculature and increased liver fibrinogen. a Confocal images (× 20) showing immunofluorescent staining of fixed, frozen brain sections (30 μm) from P. chabaudi-infected IL-10 KO and WT mice (day 8 p.i., n =4 mice/group). Fibrin (red) and tomato lectin (green, vascular endothelium). b Fibrin (red) was quantified by surveying 10 fields per brain section (× 10). Graph showing average percent area of fibrin-positive staining above threshold in each field. c Immunofluorescent staining (× 10) and quantitation of fibrinogen (red) in liver from infected IL-10 KO, WT, and uninfected controls (n = 4 mice/group). One-way ANOVA, followed by post hoc Bonferroni method, was used to determine statistical significance. *p < 0.05, **p < 0.01. Scale bar represents 100 μm Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 7 of 17 Fig. 2 Vascular congestion in IL-10 KO mice with malaria includes thrombi-containing monocytes and T cells. Immunofluorescent staining of fixed, frozen brain sections (30 μm) from P. chabaudi-infected IL-10 KO mice (day 8 p.i., n =4 mice). a Confocal images (× 40) of IL-10 KO brain stained for fibrin (red). b Successive single-plane confocal images (× 40) of a 30-μm z-stack showing complete occlusion of a large vessel with residual fibrinogen (red). c Immunofluorescence staining of IL-10 KO brains showing fibrin staining of blood vessels (red) and leukocytes expressing CD45 (green, × 60) and d CD11b (green, × 40). e CTV CD4 T cells (blue) from infected IL-10 KO mice were adoptively transferred into infection-matched IL-10 KO (day 7 p.i.) recipients 3.5 h before sacrifice. Frozen brain sections (day 7 p.i.) were stained for fibrin (red). Max intensity projection of a 30-μm z-stack (× 240) displayed from brain tissue of IL-10 KO mice co-stained with WT control samples (n =3–4 mice per group). Scale bars represent 30 μm(a), 50 μm(b–d), or 10 μm(d) pan-leukocyte marker, CD45 (Fig. 2c), and the monocyte experimental cerebral malaria [62], we analyzed the ex- marker, CD11b (Fig. 2d). Staining showed that many, but tent of astrocyte activation in IL-10 KO mice infected + + not all, CD45 and CD11b leukocytes are contained within with P. chabaudi. In order to visualize extensive activa- areas of residual fibrinogen staining. We previously quanti- tion of astrocytes, we utilized CLARITY followed by op- fied CD11b cells within the brains of P. chabaudi-infected tical clearing, a tissue processing technique which IL-10 KO mice using flow cytometry. In that analysis, we removes relatively opaque lipids, transforming thick sa- + + showed that the CD11b cells were also Ly6C , indicating gittal brain sections (2 mm) to make them optically that they are inflammatory monocytes [16]. There was a transparent. This process diminishes excess light scatter- hi large and significant increase in cerebral Ly6C inflamma- ing during image acquisition by confocal or two-photon tory monocytes in IL-10 KO compared to that in infected microscopy, allowing for increased imaging depth be- int WT mice, while a Ly6C population of resident macro- yond that possible in unprocessed tissue. The ability to phages was not increased. obtain image stacks over the full 2 mm thickness com- We were also interested to see if CD4 T cells, the pri- bined with image stitching allowed for image acquisition mary producers of IL-10 in this infection, were also found of the entire thick sagittal section. Whole brain sections localized with fibrin(ogen) in the vessels. Therefore, CD4 stained for glial fibrillary acidic protein (GFAP), which is T cells (CellTrace Violet ) from IL-10 KO mice 7 days upregulated on activated astrocytes, were imaged to de- post-infection (p.i.) were adoptively transferred into termine the extent of astrocyte activation in susceptible infection-matched IL-10 KO recipients, which underwent IL-10 KO mice (Fig. 3a, c, e) and resistant WT animals transcardial perfusion and brain tissue collection 3.5 h (Fig. 3b, d, f). A higher GFAP signal was observed in later. Transferred CD4 T cells were indeed identified in multiple areas of the IL-10 KO brain compared to WT, the brain, and often within a fibrin(ogen) clot (Fig. 2e). including the hippocampus, thalamus, and caudate puta- While the number of leukocytes is not large, activated leu- men, suggesting astrocyte activation via increased pro- kocytes have the potential to promote activation of the duction of inflammatory cytokines (Fig. 3a, b). While neuroglial cells surrounding the vasculature, namely, as- GFAP is expressed on most astrocytes, even in unin- trocytes. Therefore, we next tested brain sections from in- fected animals, the level of expression is significantly fected IL-10 KO animals for astrogliosis. lower than on activated astrocytes [63]. Interestingly, there was little GFAP signal in the cortex, a result that is Inflammatory cytokine TNF induces astrocyte activation in consistent with findings in human CM autopsy [20]. For clusters near thrombotic cerebral vasculature in IL-10 KO quantitation of astrogliosis, we focused our analysis on mice with malaria the hippocampal formation (Fig. 3c, d), as a representa- As astrocytes play an important role in maintaining the tive region in which astrogliosis was evident. This region integrity of the BBB, including in the context of is amenable to being isolated from other regions by Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 8 of 17 Fig. 3 Increased astrocyte activation in IL-10 KO mice with malaria. Mice were infected with P. chabaudi and sacrificed 8 days post-infection. Thick sagittal brain sections (2 mm) were CLARITY-processed, optically cleared, stained with GFAP (red), and imaged by two-photon confocal microscopy. a, c, e IL-10 KO and b, d, f WT brains from the peak of P. chabaudi infection (day 8 p.i., n = 5 mice/group). a, b Single fields of the entire tissue section (× 10) stitched together. c, d Hippocampus of the thick brain section is masked for increased resolution and quantitation in c IL-10 KO and d WT animals (n = 3 mice/group). e, f Representative high-resolution image (× 25) of astrocytes from the hippocampus showing the e IL-10 KO and f WT control brains. g Quantification of the percent area of astrocyte staining above threshold in the hippocampal formation of the P. chabaudi-infected IL-10 KO and WT brains. Number of fields for IL-10 KO (n = 15) and WT (n = 9). Scale bars represent 1 mm (a, b), 200 μm (c, d), and 50 μm(e, f). Student’s t test was used to determine statistical significance. **p < 0.01 image processing due to its well-defined margin and thus congestion characterized by fibrin staining and astrocyte bright allowed for comparison of GFAP cells in the full vol- activation in this hyper-inflammatory response, we next ume of the hippocampal region in each section. As shown sought to determine the role that inflammatory cyto- in high-resolution 3D micrographs (Fig. 3e, f), in addition kines play in this process. to upregulation of GFAP, astrocytes in IL-10 KO mice Immunopathology in IL-10 KO mice infected with P. showed distinct morphological changes, appearing hyper- chabaudi is generated by the hyper-inflammatory cyto- trophied and with more processes compared to infected kine response generated in the absence of this regulatory bright WT. The GFAP fraction of the hippocampal forma- cytokine primarily made by T cells [32]. Neutralizing tion in infected IL-10 KO mice was significantly increased TNF is known to improve survival and also improve all compared to WT mice (Fig. 3g). While elevated serum measures of symptomatic pathology, while Ifngr1 defi- ammonia from potential liver damage can activate astro- ciency in IL-10 KO mice improves survival [14, 31]. Im- cytes [64], there was no significant difference in ammonia portantly, neutralizing the other major regulatory production between WT and IL-10 KO mice (Add- cytokine, transforming growth factor-β, increases mor- itional file 2: Figure S2). As inflammation, or vascular tality of the IL-10 KO to 100%, suggesting that the bal- damage, can also lead to astrocyte activation, we next in- ance of inflammatory and regulatory cytokines in the vestigated whether vascular congestion and astrocyte acti- immune response to malaria infection determines the le- vation occurred in close proximity. thality of P. chabaudi in IL-10 KO mice [14]. However, To investigate the potential connection between vas- the role of TNF in brain pathology, including its behav- cular congestion and astrocyte activation, we performed ioral results, has not yet been investigated in this model. immunofluorescent staining of peak-infected (day 7 p.i.) As an indication of brain pathology, we used a and uninfected IL-10 KO brains for fibrin(ogen) and semi-quantitative P. chabaudi-specific SHIRPA health astrocyte activation. In the hippocampal formation, we assessment abbreviated from one we previously de- observed an increase in residual fibrin(ogen) staining in scribed [16]. We have now identified a smaller set of be- the infected IL-10 KO brains compared to WT (Fig. 4). havioral symptoms, described in the “Methods” section, Interestingly, the astrocytes showed an increase in GFAP that specifically change at the time that IL-10 KO mice staining and polarity and were more frequently found in begin to succumb to infection. The SHIRPA screen was contact with fibrin-containing vessels in infected IL-10 highly predictive of outcome, as the SHIRPA scores of KO brains compared to infected WT and uninfected mice that died during infection were significantly lower IL-10 KO controls (Additional file 3: Figure S3). How- than those of mice that survived (Additional file 1: Fig- ever, it was noted that not all areas with residual fibrin ure S1). In addition, we were able to use the abbreviated staining were located near highly activated astrocytes. SHIRPA to identify animals predicted to succumb to Uninfected mice showed neither residual fibrinogen de- hyper-inflammatory experimental cerebral malarial dis- position, nor an increase in GFAP immunoreactivity. ease. Any P. chabaudi-infected IL-10 KO mouse that Having established a link between microvascular scored below 17, out of a maximum of 22, on the Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 9 of 17 Fig. 4 Activated astrocytes cluster along thrombus-containing brain vasculature. IL-10 KO mice were either infected with P. chabaudi and sacrificed 8 days post-infection or used as uninfected controls. Representative epifluorescence images (× 20) of the hippocampal formation in cryosections (30 μm) from infected (day 8 p.i.) IL-10 KO brains (left, middle) and uninfected IL-10 KO brains (right) immunostained for GFAP (green), fibrinogen (red), and DAPI (blue). IL-10 KO mice were co-stained with WT control samples (n =5–6 mice per group). Scale bars represent 50 μm abbreviated SHIRPA screen before day 9 post-infection had with P. chabaudi, we treated infected IL-10 KO mice a statistically significant chance of succumbing to infection, with the anticoagulant drug, enoxaparin sodium, a with an odds ratio of 23.7 (95% CI 4.0–126.0, χ test), low-molecular weight heparin (LMWH), starting on day meaning they had almost 24 times more probability to suc- 4 post-infection through to the end of peak illness at day cumb to disease. However, two out of 49 mice (4.1%) that 12 post-infection, when all control animals had died. were predicted to die actually survived. In addition, due to Mice were treated twice per day and monitored using the speed of progression from undetectable morbidity to the abbreviated SHIRPA screen. Blood smears were also mortality, some animals (11/28, 39%) will die naturally collected on day 9 post-infection to monitor parasite without ever exhibiting a low SHIRPA score. burden. Strikingly, LMWH treatment of IL-10 KO mice To test the role of TNF in neuroimmunopathology and rescued them from fatal neurologic disease before day 9 astrocyte activation in this infection, we treated IL-10 KO post-infection (Fig. 6a). However, LMWH-treated IL-10 mice with neutralizing anti-TNF antibody or isotype control KO mice were still susceptible to delayed mortality, as antibody for 5 days (days 5–9p.i.) [14]. To monitor for fi- two out of four ENO-treated mice (50%) died after day 9 brinogen accumulation and astrocyte activation, mice were post-infection. This may represent death from severe sacrificed at day 8 p.i., at the onset of severe disease, and anemia that typically presents after the peak of P. cha- brain tissue was stained for confocal microscopy. We ob- baudi infection [65]. The differential mortality between served an increase in astrocyte activation and increased re- treatment groups was not due to differences in parasit- sidual fibrinogen in isotype-treated IL-10 KO animals emia at the peak of infection on day 9 p.i., while behav- (Fig. 5a), but neither of these changes were observed in the ioral scores were significantly improved with LMWH IL-10 KO group treated with neutralizing anti-TNF anti- treatment (Fig. 6b). As a control to assure the quality of bodies (Fig. 5b), similar to isotype-treated WT mice the treatment, we quantitated fibrinogen deposition in (Fig. 5c). These changes were significant, with a complete the brains of treated animals and confirmed that LMWH reductioninfibrinogen accumulation (Fig. 5d) and astro- eliminated thrombi completely (Fig. 6c). Strikingly, we cyte activation (Fig. 5e). Furthermore, animals were pro- found that astrogliosis was significantly reduced by anti- tected from behavioral symptoms during anti-TNF coagulant treatment, though not to the levels seen in un- treatment (Fig. 5f). Behavioral symptoms declined after infected animals (Fig. 6d). In conclusion, LMWH treatment stopped, but we did not observe any late mortal- treatment decreased astrocyte activation and intravascu- ity. As expected, excess production of fibrinogen in the liver lar fibrin clotting, suggesting that thrombi in cerebral was also reduced by anti-TNF treatment (Fig. 5g). As vasculature play a critical role in astrogliosis and lethal anti-TNF blocks many components of the acute phase re- pathology from malaria without affecting parasitemia. action besides coagulation, we proceeded with more spe- Microglia are important sentinels and potent ampli- cific tests for the importance of coagulation to fiers of inflammation within the CNS. In response to en- hyper-inflammatory experimental cerebral malaria. vironmental cues and inflammatory stimuli, microglia become activated and undergo characteristic morpho- Anticoagulant treatment eliminates early mortality and logical changes. Therefore, we quantified both upregula- reduces glial cell activation in IL-10 KO mice with malaria tion of Iba1, a marker of activation, and morphological To test the hypothesis that thrombi contribute to the changes characteristic of microglial activation in brain fatal neurological phenotype of IL-10 KO mice infected sections from either uninfected or P. chabaudi-infected Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 10 of 17 Fig. 5 Anti-TNF antibody treatment prevents astrocyte activation and mortality in IL-10 KO mice with malaria. Mice were infected with P. chabaudi and followed throughout the acute phase of infection (day 12 p.i.) or sacrificed 8 days post-infection for immunofluorescent staining. One group of IL-10 KO mice received anti-TNF IgG treatment (n = 5), while another group of IL-10 KO mice (n = 5) and a group of WT mice received isotype IgG as control (n =5). a Representative confocal images (× 20) of cryosections stained for astrocytes (GFAP; green) and fibrinogen (red) with DAPI (blue) in sagittal brain sections in anti-TNF antibody-treated IL-10 KO mice, b isotype IgG-treated IL-10 KO mice, c and isotype IgG-treated WT mice. d Brain fibrinogen and e GFAP staining for reactive astrocytes in the hippocampus were quantified by calculating the percent area per field of immunostaining above signal threshold. Ten fields per animal were assessed, with the graph showing the mean value per animal. f General behavior as measured by the abbreviated SHIRPA screen of anti-TNF antibody-treated (IL-10 KO, n = 5) and isotype IgG-treated (IL-10 KO, n =5; WT, n = 5) mice infected with P. chabaudi. Green arrows represent the dosing schedule of either anti-TNF IgG or isotype control IgG. g Liver fibrinogen quantitation. Data shown is representative of two independent experiments (n = 9 total mice/group). One-way ANOVA, followed by post hoc Bonferroni method, was used to determine statistical significance. *p <0.05, **p < 0.01, ***p < 0.001. Scale bars represent 50 μm mice on day 8 p.i. (Fig. 7a). We observed dramatic four quantitative assessments: (1) total immunoreactive changes in the microglia in the IL-10 KO compared to area (% of total Iba1-positive pixels in a field); (2) aver- WT, and we observed further changes in the age immunoreactive area per microglia; (3) transform- anticoagulant-treated animals. To interpret these ation index, a measure of microglial ramification; and changes, we quantitated the extent of microglial activa- (4) area fraction of small processes, which is normalized tion in these images based on morphology. We used to total immunoreactive area. The latter was done to Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 11 of 17 Fig. 6 IL-10 KO mice are rescued from fatal neurologic disease with LMWH treatment. a Two groups of IL-10 KO mice (n = 4) were either treated with 1000 IU/kg (20 IU/dose) enoxaparin Na (ENO) i.p. twice a day (12 h apart) or given saline starting at day 4 post-infection until the middle of the anemic period of disease (day 12 post-infection). b Survival was monitored daily, and blood smears were collected on day 9 post-infection. Behavior was monitored daily using the abbreviated SHIRPA screen (n = 4 mice/group). c Fibrinogen quantification in the brains of untreated and LMWH-treated mice at the peak of infection (day 9 post-infection, n = 4 mice/group). d 30-μm brain hippocampus cryosections stained for astrocytes (GFAP, green). GFAP staining quantified by calculating the percent area per field of immunostaining above signal threshold. One-way ANOVA, followed by post hoc Bonferroni method was used to determine statistical significance. *p < 0.05, **p < 0.01. Scale bars represent 50 μm capture differences in small/fragmented processes, as The morphological changes in infected IL-10 KO mice small processes were not observed in the IL-10 KO group, show significant changes in microglial activation state, while they were present in the LMWH group, although suggestive of increased intracranial inflammation. Interest- not as numerous as the WT group (Fig. 7b). The last ingly, all features of activation show significant improve- graph, therefore, shows how much Iba1-reactive area each ment towards homeostasis after clearance of thrombi group has with respect to the area occupied by microglia following LMWH treatment. Therefore, these findings soma, which was significantly lower in the untreated IL-10 demonstrate a critical role of inflammation-driven coagu- KO group. We interpret this to mean that activated lation in experimental cerebral malaria pathology. microglia retract their dendrites, which then appear thicker, as opposed to the thinner processes that cover Discussion more three-dimensional area in homeostasis. All of these The presence of peripheral immune cells adherent measures suggest that microglial activation is reduced, but within the vasculature in mouse models of CM and in not back to homeostatic levels, by LMWH treatment, brain vessels on autopsy of cerebral malaria patients [66] similar to what we found for astrogliosis above. suggest that such cells play an important role in mediat- In order to determine the relative localization of acti- ing neuropathology [67]. Current paradigms to explain vated microglia and cerebral vasculature, immunofluor- CM pathogenesis support an important role for inflam- escent staining was performed on microglia (Iba1) and mation in the generation and amplification of neuro- CD31 blood vessels (Fig. 7c). We observed increasing pathology but do not explain the derivation of these microglial polarity and thickening of dendrites in IL-10 cytokines in the brain. The derivation and contribution KO animals, with decreased numbers of small processes of cerebral thrombi to CM pathology is also poorly in the microglia of untreated IL-10 KO mice. The understood. The vascular findings in this study suggest- localization of microglia near vessels in infected animals ive of pervasive (Fig. 1) and complete (Fig. 2) blockade is clearly seen when viewed as a 3D stack. Enumeration of the vasculature by inflammation-induced thrombi are of the number of microglia that interacted with a blood striking. These abnormalities have not been described in vessel, defined as either body or process on the blood P. chabaudi infection before. Coagulation is clearly of vessel, indicated 79% of glia interacted with a vessel in major relevance for our understanding of pathological the KO group vs. 54% in the WT (p < 0.05), and while mechanisms in cerebral malaria [21, 58, 68]. Potentially mean value for LMWH-treated IL-0 KO mice was 69%, pathogenic serum levels of both pro- and anticoagula- it was not statistically significant from either KO or WT. tion proteins have been documented in human CM [69, Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 12 of 17 Fig. 7 Microglia changes in IL-10 KO mice infected with P. chabaudi. a Representative images of day 8 p.i. WT, IL-10 KO, and LMWH-treated IL-10 KO mice (n = 4 mice/group) 30-μm brain cryosections stained with anti-Iba-1 antibodies and visualized using DAB. b Quantitative analysis of microglia morphology in WT, IL-10 KO, and LMWH-treated IL-10 KO mice using ImageJ software. c Immunofluoresence imaging of microglia (Iba-1-Alexa 488, green), endothelial cells (CD31-Alexa 567, red), and nuclei (DAPI, blue) in 30-μm brain cryosections from WT, IL-10 KO, and LMWH-treated IL-10 KO mice during the peak of infection. Right, 3D reconstruction showing the spatial orientation of microglia cells in relationship to microvasculature in a P. chabaudi-infected IL-10 KO mouse. One-way ANOVA, followed by post hoc Tukey’s test, was used to determine statistical significance. **p < 0.01. Scale bars represent 20 and 50 μm 70]. Systemic inflammation has also recently been shown formation and activation of cells in the brain paren- to contribute to intravascular clotting via mechanisms chyma in the absence of local parasite adhesion. Studies involving neutrophils and monocyte interaction with of Plasmodium berghei (ANKA) (PbA) infection have platelets in CM [71, 72], linking inflammation and clot- established the importance of the inflammatory response ting, which in turn promote sequestration. Recent stud- in the development of neurocognitive dysfunction [74– ies also show that the anticoagulation endothelial 76]. PbA infection shows pathogenic immune cell accu- protein C receptor (EPCR) may bind the parasite and be mulation in cerebral blood vessels as a result of inflam- downregulated, thus promoting clotting and suggesting matory TNF and IP-10 secretion [77, 78] and a mechanism for the induction of coagulation by P. fal- intercellular adhesion molecule-1 (ICAM-1) on the vas- ciparum sequestration [45, 73]. Interestingly, studies cular endothelium [79]. PbA infection has also been point to the bi-directional amplification of the clotting shown to induce astrocyte activation and degeneration cascade and inflammation suggesting an important inter- near sites of monocyte vascular adhesion [62, 80]. How- section that is likely to be crucial to pathology in CM ever, the signals leading to the breakdown of local astro- [58]. cyte barrier function in malaria have not yet been The data presented here confirm that inflammatory defined. The activation of astrocytes is a feature of many cells within the vasculature can drive both clot neurological diseases, including cerebral malaria [81, 82]. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 13 of 17 Our results demonstrate a causal link between 91]. Selection of treatments with relatively moderate hyper-inflammation, hyper-coagulation, glial cell activa- anticoagulation activity is likely essential to achieving tion, and mortality (Figs. 3 and 4). Gliosis across mul- therapeutic goals while avoiding hemorrhagic complica- tiple areas of the brain was observed in infected IL-10 tions. LMWH, as the name implies, involves only the ac- KO mice, with astrocytes and microglia associating tivity of the smaller heparin proteins, which act with highly with the vasculature compared to the WT higher specificity on factor Xa, exhibit less thrombin in- group—yet both microglial and astrocyte gliosis were hibition, and produce a more reliable therapeutic profile. significantly reduced upon LMWH treatment, indicating Our studies show that LMWH treatment is protective this direct link. within the context of hyper-inflammatory cerebral mal- This is important because resolution of CM in African aria and prevents intravascular thrombi formation in the children and Asian adults can be resistant to brains of mice exhibiting behavioral dysfunction (Fig. 6). anti-malarial drug treatment, suggesting that parasite This is particularly important in that both astrocyte and alone does not cause the full cerebral malaria syndrome. microglial activation were dependent on this coagulation Furthermore, it is not yet clear how parasite adhesion event to some degree (Figs. 6 and 7). Activation of alone drives the neuropathology evident from patient microglia has been shown to be an important compo- symptoms [83]. However, because of the overlap of in- nent of neuroinflammation and behavioral dysfunction flammation with parasite-dependent factors, determining associated with PbA infection [92–94]. Widespread the independent contributions of each presents an on- microglial activation, not always restricted to areas of going challenge to investigators. The impact of parasite parasite sequestration, has also been identified in cases adhesion to the vascular endothelium on coagulation, of human CM [95, 96]. However, these findings are vascular integrity, and congestion has been shown in in novel in the context of P. chabaudi infection. Further- vitro endothelial cultures and animal models of cerebral more, the spatial relationship of intravascular coagula- malaria [19, 43, 67, 84, 85]. Sequestration is seen in most tion with glial cell activation is also previously unknown fatal pediatric and adult CM cases [20, 21] and is used in any malaria infection and should be examined in hu- as a critical hallmark of disease. We chose to study the man CM autopsy samples. role of inflammatory cytokines in isolation from the po- Efforts to manipulate the inflammatory response and tential contribution of sequestration using an clotting cascade have provided mixed results in clinical inflammation-induced cerebral malaria model. The re- trials to date [97–99], highlighting the importance of un- sults confirm that inflammation can cause many of the derstanding the interactions between various arms of the pathological changes seen in CM, though not all. host response within the pathogenesis of cerebral mal- In this study, we show that both the congestion aria. In summary, our experiments support the import- phenotype associated with intravascular clotting and ance of intravascular coagulation and leukocytes astrocyte activation can be reversed via neutralization of producing inflammatory cytokines in malaria-induced TNF (Fig. 5), or anticoagulant therapy (Fig. 6). Serum cerebral pathology. The activation of surveilling microglia TNF concentration correlates with severity of human and vascular/neuronal-supportive astrocytes downstream malaria [86]. However, TNF blockade has thus far of systemic inflammation could promote the generation of proven ineffective in preventing death in childhood cere- neuropathology secondary to malaria infection. Identifica- bral malaria [87, 88]. As different reagents displayed dif- tion of both T cells and monocytes within fibrin clots sug- ferential effects, the timing, dose, or precise antigenic gests a new working model where inflammatory cells specificity of treatments may yet be improved for adju- promote cerebral damage even from their localization vant therapy. Strikingly, these data also show that fatal within the cerebral vasculature. It is possible that leuko- neurological disease in IL-10 KO mice is dependent on cytes within the structure of intravascular thrombi serve intravascular coagulation, as it can be prevented by to amplify pathological inflammatory cytokines leading to LMWH treatment (Fig. 6). This demonstrates a central immunopathology in the brain. These data demonstrate role for thrombi in driving the disease mortality and the interaction of the anti-parasitic and hemostatic ele- promoting neuropathology in P. chabaudi infection of ments of host defense, promoting a new appreciation of IL-10 KO mice. As anti-TNF and anticoagulants have the interplay between mechanisms important for develop- similar effects in this model, it is likely that cytokines ment of fatal cerebral malaria. and the coagulation cascade promote each other, as in other systems. Despite the WHO recommendation Conclusions against the use of heparin since 1984, citing excessive Our study has identified intravascular thrombi within bleeding [89], there are several clinical trials showing the cerebral vasculature during severe P. chabaudi infec- significant beneficial effects of anticoagulant usage on tion and showed that they contribute to lethal immuno- mortality and length of coma in human CM [49, 50, 90, pathology. Furthermore, vascular congestion with an Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 14 of 17 accumulation of leukocytes is spatially associated with Acknowledgements We would like to acknowledge Andrew S. Mendiola for suggesting the astrocyte and microglial activation in this model, with fibrinogen stain and for the immunofluorescence protocols from the the former being driven by TNF. The most striking find- Cardona laboratory. We appreciate Ping Wu, Tiffany Dunn, Yongquan Jiang, ing is that dissipation of these inflammatory foci within Paula Villarreal, and the Biomedical Imaging Network at UTMB for the technical assistance in the sample prep and confocal microscopy. We fibrin-rich thrombi by LMWH treatment leads to a sig- appreciate the feedback from the UTMB Joint Immunology Lab Meeting and nificant decrease in early lethal pathology. These find- the Neuro-Infectious Diseases working group organized by the UTMB ings begin to define the parameters of inflammation in Institute for Human Infection and Immunity, particularly Kathryn Cunningham and Kelly Dineley. We are also indebted to Linsey Yeager for the brain during cerebral malaria, and the downstream the critical review of the manuscript. We are grateful for the excellent animal pathology linked to hyper-inflammation. Previously, care by the UTMB Animal Resource Center and the training from the Rodent findings of cytokine gene linkage to CM were under- In Vivo Assessment Core. stood in terms of increasing parasite binding within the Funding capillary bed. Our findings demonstrate that inflamma- Funding for these studies was from the UTMB Institute for Human Infections tory cytokines contribute both pathogenic coagulation and Immunity (IHII) and the John S. Dunn Foundation (RS, KDW), McLaughlin and activation of sentinel glia in the brain parenchyma, Endowment Predoctoral Grant and Jeane B. Kempner Award (KDW), and NIH (R01AI363327 (RS), R01NS078501 (AEC), and T32AI363327 (KDW)). This study which are capable of causing neurological sequelae, even was also supported by a grant from the University of Texas System in the absence of localized sequestration, although to a Neuroscience and Neurotechnology Research Institute (RS, GV, #363327). The lesser degree than more virulent parasites. These find- funders had no role in the study design, data collection and interpretation, or the decision to submit the work for publication. ings, therefore, contribute to the current understanding of the etiologies of cerebral pathology and neurovascular Availability of data and materials abnormalities in malaria infection. While the effective- All data generated or analyzed during this study are included in this ness and safety of this approach must be validated, the published article and its supplementary information files. positive effect of anticoagulants could inform develop- ment of future adjunctive therapy for CM patients. Authors’ contributions Epifluorescent and confocal imaging studies were designed by KDW, GV, RS, conducted by KDW, with assistance from LFO and ODS, and analyzed by KDW, GV, and RS. Two-photon imaging studies were designed by GV and RS, Additional files conducted by LFO and ODS, and analyzed by LFO, ODS, and GV. Fibrinogen and microglia imaging studies were designed by KDW, RS, and AEC, con- ducted by KDW and SMC, and analyzed by KDW, RS, GV, LFO, and RP. Adop- Additional file 1: Figure S1. IL-10 KO mouse behavioral scores are tive transfer, antibody treatment, and animal behavior studies were designed predictive of outcome during P. chabaudi Infection. Left: representative by KDW and RS, conducted by KDW, and analyzed by KDW and RS. Tissue experiment showing SHIRPA scores of infected male IL-10 KO mice ammonia quantitation were performed and analyzed by VHC and ODS. grouped by eventual outcome (survived = blue, n =6; died = black, n =5). Statistical analyses and figures were generated by KDW, LFO, ODS, VHC, and Right: graph of the lowest abbreviated SHIRPA score in individual mice before PHK. KDW and RS drafted the manuscript, with essential input from GV, LFO, day 9 post-infection with infected IL-10 KO mice grouped according to ODS, and AEC. All authors read and approved the final manuscript. outcome. Showing concatenated data from multiple experiments (n =48). Error bar represents SEM, ***p < 0.001, Wilcoxon signed rank test. (TIF 326 kb) Additional file 2: Figure S2. IL-10 KO mice ammonia levels are not Ethics approval elevated above WT during P. chabaudi infection. WT and IL-10 KO mice Animals were cared for according to the Guide for the Care and Use of were infected with P. chabaudi and monitored during the peak of Laboratory Animals under the Institutional Animal Care and Use Committee- infection. WT mice were sacrificed at the peak of infection (day 10 p.i.) approved protocol #1006031. UTMB Animal Resource Center facilities operate in and IL-10 KO mice upon severe morbidity as determined via SHIRPA compliance with the USDA Animal Welfare Act, the Guide for the Care and Use of score. Organ and plasma ammonia levels were measured using a Laboratory Animals, under OLAW accreditation, and IACUC-approved protocols. colorimetric ammonia assay. (TIF 218 kb) Additional file 3: Figure S3. P. chabaudi-infected IL-10 KO mice show Competing interests astrocyte association with thrombi. Left, normalized astrocyte-thrombus The authors declare that they have no competing interests. association ratio. Right, representative confocal images of experimental groups stained for astrocytes (green) and fibrinogen (red). N =3–5 mice/ group. Error bar represents 30 μm. (TIF 12464 kb) Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details Abbreviations Department of Microbiology and Immunology, University of Texas Medical 3D: Three-dimensional; BBB: Blood-brain-barrier; CM: Cerebral malaria; Branch, 301 University Boulevard, Galveston, TX 77555, USA. Center for CNS: Central nervous system; CXCR3: C-X chemokine receptor 3; Biomedical Engineering, University of Texas Medical Branch, 301 University DIC: Disseminated intravascular coagulation; ECM: Experimental cerebral Boulevard, Galveston, TX 77555, USA. Department of Biology, One UTSA malaria; GFAP: Glial fibrillary acidic protein; i.p.: Intraperitoneal; ICAM- Circle, University of Texas at San Antonio, San Antonio, TX 78249, USA. 1: Intracellular adhesion molecule-1; IFN-γ: Interferon gamma; Department of Neuroscience and Cell Biology, University of Texas Medical IHC: Immunohistochemistry; IL-10 KO: IL-10-deficient; iRBCs: Infected red Branch, 301 University Boulevard, Galveston, TX 77555, USA. Department of blood cells; MHC-II: Major histocompatibility complex class II; Internal Medicine, Division of Infectious Diseases, University of Texas Medical PbA: Plasmodium berghei (ANKA); SHIRPA: SmithKline Beecham, Harwell, Branch, 301 University Boulevard, Galveston, TX 77555-0435, USA. Institute Imperial College, Royal London Hospital Phenotype Assessment; TNF: Tumor for Human Infections and Immunity, University of Texas Medical Branch, 301 necrosis factor; WT: Wild-type, C57Bl/6J University Boulevard, Galveston, TX 77555, USA. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 15 of 17 Received: 14 February 2018 Accepted: 17 May 2018 congestion are associated with coma in human cerebral malaria. J Infect Dis. 2012;205:663–71. 22. Jensen AT, Magistrado P, Sharp S, Joergensen L, Lavstsen T, Chiucchiuini A, Salanti A, Vestergaard LS, Lusingu JP, Hermsen R, et al. Plasmodium References falciparum associated with severe childhood malaria preferentially expresses 1. World Malaria Report 2016. Geneva: World Health Organization; 2016. http:// PfEMP1 encoded by group A var genes. J Exp Med. 2004;199:1179–90. www.who.int/malaria/publications/world-malaria-report-2016/report/en/. 23. Bernabeu M, Danziger SA, Avril M, Vaz M, Babar PH, Brazier AJ, Herricks T, Accessed 13 March 2018. Maki JN, Pereira L, Mascarenhas A, et al. Severe adult malaria is associated 2. Seydel KB, Kampondeni SD, Valim C, Potchen MJ, Milner DA, Muwalo FW, with specific PfEMP1 adhesion types and high parasite biomass. Proc Natl Birbeck GL, Bradley WG, Fox LL, Glover SJ, et al. Brain swelling and death in Acad Sci U S A. 2016;113:E3270–9. children with cerebral malaria. N Engl J Med. 2015;372:1126–37. 24. Lavstsen T, Turner L, Saguti F, Magistrado P, Rask TS, Jespersen JS, Wang CW, 3. Idro R, Jenkins NE, Newton CR. Pathogenesis, clinical features, and Berger SS, Baraka V, Marquard AM, et al. Plasmodium falciparum erythrocyte neurological outcome of cerebral malaria. Lancet Neurol. 2005;4:827–40. membrane protein 1 domain cassettes 8 and 13 are associated with severe 4. Idro R, Kakooza-Mwesige A, Balyejjussa S, Mirembe G, Mugasha C, Tugumisirize J, malaria in children. Proc Natl Acad Sci U S A. 2012;109:E1791–800. Byarugaba J. Severe neurological sequelae and behaviour problems after cerebral 25. Avril M, Tripathi AK, Brazier AJ, Andisi C, Janes JH, Soma VL, Sullivan DJ Jr, malaria in Ugandan children. BMC Res Notes. 2010;3:104. Bull PC, Stins MF, Smith JD. A restricted subset of var genes mediates 5. Armah HB, Wilson NO, Sarfo BY, Powell MD, Bond VC, Anderson W, Adjei AA, adherence of Plasmodium falciparum-infected erythrocytes to brain Gyasi RK, Tettey Y, Wiredu EK, et al. Cerebrospinal fluid and serum biomarkers endothelial cells. Proc Natl Acad Sci U S A. 2012;109:E1782–90. of cerebral malaria mortality in Ghanaian children. Malar J. 2007;6:147. 26. Grau GE, Piguet PF, Vassalli P, Lambert PH. Tumor-necrosis factor and other 6. McGuire W, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D. Variation in cytokines in cerebral malaria: experimental and clinical data. Immunol Rev. the TNF-alpha promoter region associated with susceptibility to cerebral 1989;112:49–70. malaria. Nature. 1994;371:508–10. 27. Kwiatkowski DP, Luoni G. Host genetic factors in resistance and 7. Ouma C, Davenport GC, Were T, Otieno MF, Hittner JB, Vulule JM, Martinson susceptibility to malaria. Parassitologia. 2006;48:450–67. J, Ong’echa JM, Ferrell RE, Perkins DJ. Haplotypes of IL-10 promoter variants 28. May J, Lell B, Luty AJ, Meyer CG, Kremsner PG. Plasma interleukin-10: tumor are associated with susceptibility to severe malarial anemia and functional necrosis factor (TNF)-alpha ratio is associated with TNF promoter variants changes in IL-10 production. Hum Genet. 2008;124:515–24. and predicts malarial complications. J Infect Dis. 2000;182:1570–3. 8. Burt RA, Baldwin TM, Marshall VM, Foote SJ. Temporal expression of an H2-linked 29. Ho M, Schollaardt T, Snape S, Looareesuwan S, Suntharasamai P, White NJ. locus inhost responsetomouse malaria. Immunogenetics. 1999;50:278–85. Endogenous interleukin-10 modulates proinflammatory response in 9. Knight JC, Udalova I, Hill AVS, Greenwood BM, Peshu N, Marsh K, Plasmodium falciparum malaria. J Infect Dis. 1998;178:520–5. Kwiatkowski D. A polymorphism that affects OCT-1 binding to the TNF 30. Freitas do Rosario AP, Langhorne J. T cell-derived IL-10 and its impact on the promoter region is associated with severe malaria. Nat Genet. 1999;22:145. regulation of host responses during malaria. Int J Parasitol. 2012;42:549–55. 10. Fortin A, Stevenson MM, Gros P. Complex genetic control of susceptibility 31. Li C, Corraliza I, Langhorne J. A defect in interleukin-10 leads to enhanced to malaria in mice. Genes Immun. 2002;3:177–86. malarial disease in Plasmodium chabaudi chabaudi infection in mice. Infect 11. Amante FH, Haque A, Stanley AC, Rivera Fde L, Randall LM, Wilson YA, Yeo Immun. 1999;67:4435–42. G, Pieper C, Crabb BS, de Koning-Ward TF, et al. Immune-mediated 32. Freitas do Rosario AP, Lamb T, Spence P, Stephens R, Lang A, Roers A, mechanisms of parasite tissue sequestration during experimental cerebral Muller W, O'Garra A, Langhorne J. IL-27 promotes IL-10 production by malaria. J Immunol. 2010;185:3632–42. effector Th1 CD4+ T cells: a critical mechanism for protection from severe 12. Chakravorty SJ, Carret C, Nash GB, Ivens A, Szestak T, Craig AG. Altered immunopathology during malaria infection. J Immunol. 2012;188:1178–90. phenotype and gene transcription in endothelial cells, induced by 33. Couper KN, Blount DG, Wilson MS, Hafalla JC, Belkaid Y, Kamanaka M, Flavell Plasmodium falciparum-infected red blood cells: pathogenic or protective? RA, de Souza JB, Riley EM. IL-10 from CD4CD25Foxp3CD127 adaptive Int J Parasitol. 2007;37:975–87. regulatory T cells modulates parasite clearance and pathology during 13. Palomo J, Reverchon F, Piotet J, Besnard AG, Couturier-Maillard A, Maillet I, malaria infection. PLoS Pathog. 2008;4:e1000004. Tefit M, Erard F, Mazier D, Ryffel B, Quesniaux VF. Critical role of IL-33 34. Turner GD, Morrison H, Jones M, Davis TM, Looareesuwan S, Buley ID, Gatter receptor ST2 in experimental cerebral malaria development. Eur J Immunol. KC, Newbold CI, Pukritayakamee S, Nagachinta B, et al. An 2015;45:1354–65. immunohistochemical study of the pathology of fatal malaria. Evidence for 14. Li C, Sanni LA, Omer F, Riley E, Langhorne J. Pathology of Plasmodium widespread endothelial activation and a potential role for intercellular chabaudi chabaudi infection and mortality in interleukin-10-deficient mice are adhesion molecule-1 in cerebral sequestration. Am J Pathol. 1994;145:1057–69. ameliorated by anti-tumor necrosis factor alpha and exacerbated by anti- 35. MacPherson GG, Warrell MJ, White NJ, Looareesuwan S, Warrell DA. Human transforming growth factor beta antibodies. Infect Immun. 2003;71:4850–6. cerebral malaria. A quantitative ultrastructural analysis of parasitized 15. Sanni LA, Jarra W, Li C, Langhorne J. Cerebral edema and cerebral erythrocyte sequestration. Am J Pathol. 1985;119:385–401. hemorrhages in interleukin-10-deficient mice infected with Plasmodium 36. Turner G. Cerebral malaria. Brain Pathol. 1997;7:569–82. chabaudi. Infect Immun. 2004;72:3054–8. 37. Spitz S. The pathology of acute falciparum malaria. Mil Surg. 1946;99:555–72. 16. Wilson KD, Stutz SJ, Ochoa LF, Valbuena GA, Cravens PD, Dineley KT, Vargas 38. Berendt AR, Tumer GDH, Newbold CI. Cerebral malaria: the sequestration G, Stephens R. Behavioural and neurological symptoms accompanied by hypothesis. Parasitol Today. 1994;10:412–4. cellular neuroinflammation in IL-10-deficient mice infected with 39. Pongponratn E, Turner GD, Day NP, Phu NH, Simpson JA, Stepniewska K, Plasmodium chabaudi. Malar J. 2016;15:428. Mai NT, Viriyavejakul P, Looareesuwan S, Hien TT, et al. An ultrastructural 17. Mota MM, Jarra W, Hirst E, Patnaik PK, Holder AA. Plasmodium chabaudi- study of the brain in fatal Plasmodium falciparum malaria. Am J Trop Med infected erythrocytes adhere to CD36 and bind to microvascular endothelial Hyg. 2003;69:345–59. cells in an organ-specific way. Infect Immun. 2000;68:4135–44. 40. White NJ, Turner GD, Medana IM, Dondorp AM, Day NP. The murine 18. Brugat T, Cunningham D, Sodenkamp J, Coomes S, Wilson M, Spence PJ, Jarra cerebral malaria phenomenon. Trends Parasitol. 2010;26:11–5. W, Thompson J, Scudamore C, Langhorne J. Sequestration and histopathology 41. Lewallen S, Bakker H, Taylor TE, Wills BA, Courtright P, Molyneux ME. Retinal in Plasmodium chabaudi malaria are influenced by the immune response in findings predictive of outcome in cerebral malaria. Trans R Soc Trop Med an organ-specific manner. Cell Microbiol. 2014;16:687–700. Hyg. 1996;90:144–6. 19. Cunningham DA, Lin JW, Brugat T, Jarra W, Tumwine I, Kushinga G, Ramesar J, Franke-Fayard B, Langhorne J. ICAM-1 is a key receptor mediating 42. Patnaik JK, Das BS, Mishra SK, Mohanty S, Satpathy SK, Mohanty D. Vascular cytoadherence and pathology in the Plasmodium chabaudi malaria model. clogging, mononuclear cell margination, and enhanced vascular Malar J. 2017;16:185. permeability in the pathogenesis of human cerebral malaria. Am J Trop 20. Dorovini-Zis K, Schmidt K, Huynh H, Fu W, Whitten RO, Milner D, Kamiza S, Med Hyg. 1994;51:642–7. Molyneux M, Taylor TE. The neuropathology of fatal cerebral malaria in 43. Franke-Fayard B, Janse CJ, Cunha-Rodrigues M, Ramesar J, Buscher P, Que I, Malawian children. Am J Pathol. 2011;178:2146–58. Lowik C, Voshol PJ, den Boer MA, van Duinen SG, et al. Murine malaria 21. Ponsford MJ, Medana IM, Prapansilp P, Hien TT, Lee SJ, Dondorp AM, Esiri parasite sequestration: CD36 is the major receptor, but cerebral pathology is MM, Day NP, White NJ, Turner GD. Sequestration and microvascular unlinked to sequestration. Proc Natl Acad Sci U S A. 2005;102:11468–73. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 16 of 17 44. Bridges DJ, Bunn J, van Mourik JA, Grau G, Preston RJ, Molyneux M, Combes 70. Reid HA, Nkrumah FK. Fibrin-degradation products in cerebral malaria. V, O'Donnell JS, de Laat B, Craig A. Rapid activation of endothelial cells Lancet. 1972;1:218–21. enables Plasmodium falciparum adhesion to platelet-decorated von 71. Feintuch CM, Saidi A, Seydel K, Chen G, Goldman-Yassen A, Mita-Mendoza Willebrand factor strings. Blood. 2010;115:1472–4. NK, Kim RS, Frenette PS, Taylor T, Daily JP. Activated neutrophils are 45. Moxon CA, Wassmer SC, Milner DA, Jr., Chisala NV, Taylor TE, Seydel KB, associated with pediatric cerebral malaria vasculopathy in Malawian Molyneux ME, Faragher B, Esmon CT, Downey C, et al: Loss of endothelial children. MBio. 2016;7:e01300–15. protein C receptors links coagulation and inflammation to parasite 72. Hochman SE, Madaline TF, Wassmer SC, Mbale E, Choi N, Seydel KB, Whitten sequestration in cerebral malaria in African children. Blood 2013, 122:842–851. RO, Varughese J, Grau GE, Kamiza S, et al. Fatal pediatric cerebral malaria is 46. O'Regan N, Gegenbauer K, O'Sullivan JM, Maleki S, Brophy TM, Dalton N, Chion associated with intravascular monocytes and platelets that are increased A, Fallon PG, Grau GE, Budde U, et al. A novel role for von Willebrand factor in with HIV coinfection. MBio. 2015;6:e01390–15. the pathogenesis of experimental cerebral malaria. Blood. 2016;127:1192–201. 73. Sampath S, Brazier AJ, Avril M, Bernabeu M, Vigdorovich V, Mascarenhas A, 47. Sun G, Chang WL, Li J, Berney SM, Kimpel D, van der Heyde HC. Inhibition Gomes E, Sather DN, Esmon CT, Smith JD. Plasmodium falciparum adhesion of platelet adherence to brain microvasculature protects against severe domains linked to severe malaria differ in blockade of endothelial protein C Plasmodium berghei malaria. Infect Immun. 2003;71:6553–61. receptor. Cell Microbiol. 2015;17:1868–82. 48. Moxon CA, Chisala NV, Mzikamanda R, MacCormick I, Harding S, Downey C, 74. Desruisseaux MS, Gulinello M, Smith DN, Lee SC, Tsuji M, Weiss LM, Spray Molyneux M, Seydel KB, Taylor TE, Heyderman RS, Toh CH. Laboratory DC, Tanowitz HB. Cognitive dysfunction in mice infected with Plasmodium evidence of disseminated intravascular coagulation is associated with a fatal berghei strain ANKA. J Infect Dis. 2008;197:1621–7. outcome in children with cerebral malaria despite an absence of clinically 75. Lackner P, Beer R, Heussler V, Goebel G, Rudzki D, Helbok R, Tannich E, evident thrombosis or bleeding. J Thromb Haemost. 2015;13:1653–64. Schmutzhard E. Behavioural and histopathological alterations in mice with 49. Rampengan TH. Cerebral malaria in children. Comparative study between cerebral malaria. Neuropathol Appl Neurobiol. 2006;32:177–88. heparin, dexamethasone and placebo. Paediatrica Indonesiana. 1991;31: 76. Lacerda-Queiroz N, Rodrigues DH, Vilela MC, Miranda AS, Amaral DC, 59–66. Camargos ER, Carvalho LJ, Howe CL, Teixeira MM, Teixeira AL. Inflammatory 50. Munir M, Tjandra H, Rampengan TH, Mustadjab I, Wulur FH. Heparin in the changes in the central nervous system are associated with behavioral treatment of cerebral malaria. Paediatr Indones. 1980;20:47–50. impairment in Plasmodium berghei (strain ANKA)-infected mice. Exp 51. Reid HA, Sucharit P. Ancrod, heparin, and -aminocaproic acid in simian Parasitol. 2010;125:271–8. Knowlesi malaria. Lancet. 1972;2:1110–2. 77. Nie CQ, Bernard NJ, Norman MU, Amante FH, Lundie RJ, Crabb BS, Heath 52. Stone WJ, Hanchett JE, Knepshield JH. Acute renal insufficiency due to WR, Engwerda CR, Hickey MJ, Schofield L, Hansen DS. IP-10-mediated T cell falciparum malaria. Review of 42 cases. Arch Intern Med. 1972;129:620–8. homing promotes cerebral inflammation over splenic immunity to malaria 53. Grau GE, Craig AG. Cerebral malaria pathogenesis: revisiting parasite and infection. PLoS Pathog. 2009;5:e1000369. host contributions. Future Microbiol. 2012;7:291–302. 78. Campanella GS, Tager AM, El Khoury JK, Thomas SY, Abrazinski TA, Manice 54. Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE. Behavioral LA, Colvin RA, Luster AD. Chemokine receptor CXCR3 and its ligands CXCL9 and functional analysis of mouse phenotype: SHIRPA, a proposed protocol and CXCL10 are required for the development of murine cerebral malaria. for comprehensive phenotype assessment. Mamm Genome. 1997;8:711–3. Proc Natl Acad Sci U S A. 2008;105:4814–9. 55. Stephens R, Langhorne J. Effector memory Th1 CD4 T cells are maintained 79. Rudin W, Eugster HP, Bordmann G, Bonato J, Müller M, Yamage M, Ryffel B. in a mouse model of chronic malaria. PLoS Pathog. 2010;6:e1001208. Resistance to cerebral malaria in tumor necrosis factor-alpha/beta-deficient 56. Chung K, Wallace J, Kim S-Y, Kalyanasundaram S, Andalman AS, Davidson TJ, mice is associated with a reduction of intercellular adhesion molecule-1 up- Mirzabekov JJ, Zalocusky KA, Mattis J, Denisin AK, et al. Structural and regulation and T helper type 1 response. Am J Pathol. 1997;150:257–66. molecular interrogation of intact biological systems. Nature. 2013;497:332–7. 80. Ampawong S, Chaisri U, Viriyavejakul P, Nontprasert A, Grau GE, 57. Bertini R, Bianchi M, Villa P, Ghezzi P. Depression of liver drug metabolism Pongponratn E. Electron microscopic features of brain edema in rodent and increase in plasma fibrinogen by interleukin 1 and tumor necrosis cerebral malaria in relation to glial fibrillary acidic protein expression. Int J factor: a comparison with lymphotoxin and interferon. Int J Clin Exp Pathol. 2014;7:2056–67. Immunopharmacol. 1988;10:525–30. 81. Ma N, Madigan MC, Chan-Ling T, Hunt NH. Compromised blood-nerve 58. Francischetti IM, Seydel KB, Monteiro RQ. Blood coagulation, inflammation, barrier, astrogliosis, and myelin disruption in optic nerves during fatal and malaria. Microcirculation. 2008;15:81–107. murine cerebral malaria. Glia. 1997;19:135–51. 59. Blombäck B, Hessel B, Hogg D, Therkildsen L. A two-step fibrinogen—fibrin 82. Medana IM, Turner GDH. Human cerebral malaria and the blood–brain transition in blood coagulation. Nature. 1978;275:501–5. barrier. Int J Parasitol. 2006;36:555–68. 60. Raut S, McEvoy F, Gaffney PJ. Development of an ELISA for the 83. O'Sullivan JM, Preston RJ, O'Regan N, O'Donnell JS. Emerging roles for quantification of fibrin in canine tumours. Thromb Res. 1999;96:11–7. hemostatic dysfunction in malaria pathogenesis. Blood. 2016;127:2281–8. 61. Clemmensen I. Three new E-antigenic fibrinogen fractions found in a 84. Hearn J, Rayment N, Landon DN, Katz DR, de Souza JB. Immunopathology commercial plasmin preparation. Science Tools LKB Instr J. 1973;20:7–8. of cerebral malaria: morphological evidence of parasite sequestration in 62. Medana IM, Chaudhri G, Chan-Ling T, Hunt NH. Central nervous system in murine brain microvasculature. Infect Immun. 2000;68:5364–76. cerebral malaria: ‘innocent bystander’ or active participant in the induction 85. Franke-Fayard B, Fonager J, Braks A, Khan SM, Janse CJ. Sequestration and of immunopathology? Immunol Cell Biol. 2001;79:101–20. tissue accumulation of human malaria parasites: can we learn anything from rodent models of malaria? PLoS Pathog. 2010;6:e1001032. 63. Ridet JL, Privat A, Malhotra SK, Gage FH. Reactive astrocytes: cellular and molecular cues to biological function. Trends Neurosci. 1997;20:570–7. 86. Kwiatkowski D, Sambou I, Twumasi P, Greenwood BM, Hill AVS, Manogue 64. Haussinger D, Schliess F. Pathogenetic mechanisms of hepatic KR, Cerami A, Castracane J, Brewster DR. TNF concentration in fatal cerebral, encephalopathy. Gut. 2008;57:1156–65. non-fatal cerebral, and uncomplicated Plasmodium falciparum malaria. 65. Yap GS, Stevenson MM. Blood transfusion alters the course and outcome of Lancet. 1990;336:1201–4. Plasmodium chabaudi AS infection in mice. Infect Immun. 1994;62:3761–5. 87. van Hensbroek MB, Palmer A, Onyiorah E, Schneider G, Jaffar S, Dolan G, 66. Brown H,Hien TT, DayN,Mai NT,ChuongLV, Chau TT, LocPP, Phu Memming H, Frenkel J, Enwere G, Bennett S, et al. The effect of a NH, Bethell D, Farrar J, et al. Evidence of blood-brain barrier monoclonal antibody to tumor necrosis factor on survival from childhood dysfunction in human cerebral malaria. Neuropathol Appl Neurobiol. cerebral malaria. J Infect Dis. 1996;174:1091–7. 1999;25:331–40. 88. Kwiatkowski D, Molyneux ME, Stephens S, Curtis N, Klein N, Pointaire P, Smit 67. Nacer A, Movila A, Baer K, Mikolajczak SA, Kappe SH, Frevert U. M, Allan R, Brewster DR, Grau GE, et al. Anti-TNF therapy inhibits fever in Neuroimmunological blood brain barrier opening in experimental cerebral cerebral malaria. Q J Med. 1993;86:91–8. malaria. PLoS Pathog. 2012;8:e1002982. 89. Severe and complicated malaria. World Health Organization Malaria Action 68. Milner DA, Whitten RO, Kamiza S, Carr R, Liomba G, Dzamalala C, Seydel KB, Programme. Trans R Soc Trop Med Hyg. 1986;80 Suppl:3–50. Molyneux ME, Taylor TE. The systemic pathology of cerebral malaria in 90. Havlik I, Looareesuwan S, Vannaphan S, Wilairatana P, Krudsood S, Thuma African children. Front Cell Infect Microbiol. 2014;4:104. PE, Kozbor D, Watanabe N, Kaneko Y. Curdlan sulphate in human severe/ 69. Borochovitz D, Crosley AL, Metz J. Disseminated intravascular coagulation cerebral Plasmodium falciparum malaria. Trans R Soc Trop Med Hyg. 2005; with fatal haemorrhage in cerebral malaria. Br Med J. 1970;2:710. 99:333–40. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 17 of 17 91. Jaroonvesama N. Intravascular coagulation in falciparum malaria. Lancet. 1972;1:221–3. 92. Sarkar S, Keswani T, Sengupta A, Mitra S, Bhattacharyya A. Differential modulation of glial cell mediated neuroinflammation in Plasmodium berghei ANKA infection by TGF beta and IL 6. Cytokine. 2017;99:249–59. 93. Reverchon F, Mortaud S, Sivoyon M, Maillet I, Laugeray A, Palomo J, Montecot C, Herzine A, Meme S, Meme W, et al. IL-33 receptor ST2 regulates the cognitive impairments associated with experimental cerebral malaria. PLoS Pathog. 2017;13:e1006322. 94. Capuccini B, Lin J, Talavera-Lopez C, Khan SM, Sodenkamp J, Spaccapelo R, Langhorne J. Transcriptomic profiling of microglia reveals signatures of cell activation and immune response, during experimental cerebral malaria. Sci Rep. 2016;6:39258. 95. Schluesener HJ, Kremsner PG, Meyermann R. Widespread expression of MRP8 and MRP14 in human cerebral malaria by microglial cells. Acta Neuropathol. 1998;96:575–80. 96. Deininger MH, Kremsner PG, Meyermann R, Schluesener H. Macrophages/ microglial cells in patients with cerebral malaria. Eur Cytokine Netw. 2002; 13:173–85. 97. Warrell DA, Looareesuwan S, Warrell MJ, Kasemsarn P, Intaraprasert R, Bunnag D, Harinasuta T. Dexamethasone proves deleterious in cerebral malaria. N Engl J Med. 1982;306:313–9. 98. Di Perri G, Di Perri HG, Monteiro GB, Bonora S, Hennig C, Cassatella M, Micciolo R, Vento S, Dusi S, Bassetti D, Concia E. Pentoxifylline as a supportive agent in the treatment of cerebral malaria in children. J Infect Dis. 1995;171:1317–22. 99. Hemmer CJ, Kern P, Holst FGE, Nawroth PP, Dietrich M. Neither heparin nor acetylsalicylic acid influence the clinical course in human Plasmodium falciparum malaria: a prospective randomized study. Am J Trop Med Hyg. 1991;45:608–12. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Neuroinflammation Springer Journals

Elimination of intravascular thrombi prevents early mortality and reduces gliosis in hyper-inflammatory experimental cerebral malaria

Free
17 pages

Loading next page...
1
 
/lp/springer_journal/elimination-of-intravascular-thrombi-prevents-early-mortality-and-bOOt4KqbSs
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Neurology; Neurobiology; Immunology
eISSN
1742-2094
D.O.I.
10.1186/s12974-018-1207-4
Publisher site
See Article on Publisher Site

Abstract

Background: Cerebral malaria (CM) is the most lethal outcome of Plasmodium infection. There are clear correlations between expression of inflammatory cytokines, severe coagulopathies, and mortality in human CM. However, the mechanisms intertwining the coagulation and inflammation pathways, and their roles in CM, are only beginning to be understood. In mice with T cells deficient in the regulatory cytokine IL-10 (IL-10 KO), infection with Plasmodium chabaudi leads to a hyper-inflammatory response and lethal outcome that can be prevented by anti-TNF treatment. However, inflammatory T cells are adherent within the vasculature and not present in the brain parenchyma, suggesting a novel form of cerebral inflammation. We have previously documented behavioral dysfunction and microglial activation in infected IL-10 KO animals suggestive of neurological involvement driven by inflammation. In order to understand the relationship of intravascular inflammation to parenchymal dysfunction, we studied the congestion of vessels with leukocytes and fibrin(ogen) and the relationship of glial cell activation to congested vessels in the brains of P. chabaudi-infected IL-10 KO mice. Methods: Using immunofluorescence microscopy, we describe severe thrombotic congestion in these animals. We stained for immune cell surface markers (CD45, CD11b, CD4), fibrin(ogen), microglia (Iba-1), and astrocytes (GFAP) in the brain at the peak of behavioral symptoms. Finally, we investigated the roles of inflammatory cytokine tumor necrosis factor (TNF) and coagulation on the pathology observed using neutralizing antibodies and low-molecular weight heparin to inhibit both inflammation and coagulation, respectively. Results: Many blood vessels in the brain were congested with thrombi containing adherent leukocytes, including CD4 T cells and monocytes. Despite containment of the pathogen and leukocytes within the vasculature, activated microglia and astrocytes were prevalent in the parenchyma, particularly clustered near vessels with thrombi. Neutralization of TNF, or the coagulation cascade, significantly reduced both thrombus formation and gliosis in P. chabaudi-infected IL-10 KO mice. (Continued on next page) * Correspondence: rostephe@utmb.edu Department of Microbiology and Immunology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555, USA Department of Internal Medicine, Division of Infectious Diseases, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0435, USA Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 2 of 17 (Continued from previous page) Conclusions: These findings support the contribution of cytokines, coagulation, and leukocytes within the brain vasculature to neuropathology in malaria infection. Strikingly, localization of inflammatory leukocytes within intravascular clots suggests a mechanism for interaction between the two cascades by which cytokines drive local inflammation without considerable cellular infiltration into the brain parenchyma. Keywords: Malaria, Brain, Inflammation, Neuropathology, Vascular congestion, Monocyte, Astrocyte, Microglia, Immunofluorescence, Confocal microscopy, Background significant enrichment [18]. The P. chabaudi life cycle is With 212 million new cases and 429,000 estimated synchronous. Mature schizonts disappear from the cir- deaths in 2015, malaria remains one of the most eco- culation almost completely and are found sequestered nomically impactful infectious diseases worldwide [1]. A primarily in the liver and lungs of mice in a partially small percentage of Plasmodium falciparum infections ICAM1-dependent manner [19]. Interestingly, patho- results in severe malarial disease. However, a significant logical damage within each organ in P. chabaudi does proportion of severe malaria infections includes cerebral not correspond to the degree of organ-specific seques- malaria (CM), which is a leading cause of death in tration of the parasite [18]. Sequestration is a hallmark sub-Saharan African children and represents a major of autopsy in fatal P. falciparum-induced CM cases [20, burden worldwide [2]. CM accounts for an estimated 21], and specific parasite variants are associated with se- 500,000 cases per year and correlates with high parasite- vere malaria [22–25]; however, it is challenging to defini- mic burden, severe inflammation, and cerebral edema tively prove that parasite sequestration in the brain is [2]. Furthermore, about 20% of patients with CM die causal to CM. despite timely treatment [3], and neurological sequelae Activated immune cells and pro-inflammatory cyto- in survivors is common [4]. Several host genetic factors kines are also strongly implicated in the mortality in hu- have been implicated in pathology. For example, muta- man disease [26, 27]. A low ratio of IL-10 to TNF in tions in the promoters of the inflammatory cytokine patients predicts more severe malaria, as do mutations tumor necrosis factor (TNF), which drives the in the IL-10 and TNF genes [28, 29]. Mouse models anti-malaria response of phagocytes, and the regulatory show that this is because IL-10 is required to protect an- cytokine IL-10, which protects the host from excessive imals from lethal pathology, as it regulates the immunopathology, have been correlated with severe dis- pro-inflammatory cytokines IL-12 and TNF [30], which ease in both mice and humans [5–10]. However, inflam- drive as yet poorly defined neuroimmunopathology. matory cytokines also allow parasite sequestration and IL-10 KO mice lacking IFN-γ receptor signaling are also leukocyte adhesion by upregulating adhesion molecules rescued from mortality, even though they exhibit higher on the vascular endothelium [11–13]. levels of parasitemia [31]. IL-10 is primarily made by + + The role of inflammatory cytokines increased by the CD4 IFN-γ effector T cells (Teff) in P. chabaudi infec- absence of IL-10 has been studied extensively in the tion, not Tregs, and is downstream of IL-27 [32, 33], Plasmodium chabaudi mouse model of severe malaria and we have shown that CD4 Teff are found solely [14]. P. chabaudi is a rodent parasite that leads to mild within the cerebral vasculature, not in the brain paren- malaria in C57BL/6 (WT) mice. However, in chyma [16]. IL-10-deficient (IL-10 KO) mice, P. chabaudi infection While there are studies of host genetics and those cor- leads to hyper-inflammation and death. The syndrome relating systemic inflammatory cytokines with poor out- includes increased levels of the pro-inflammatory cyto- comes in severe malaria [26, 27], no significant kines TNF and IFN-γ [14] and lethal disease character- inflammatory infiltrate within the brain parenchyma has ized by cerebral pathology including cerebral edema and been documented in human or mouse studies of the dis- hemorrhage [15]. In addition, we have recently demon- ease [20, 21, 34–40]. As a result, the contribution of ac- strated pathological behavioral phenotypes indicative of tivated peripheral leukocytes to brain pathology has neurological and cognitive dysfunction in this model been poorly appreciated. Interestingly, despite the lack [16]. Strikingly, there is no significant parasite sequestra- of infiltrating inflammatory cells in the brain paren- tion in the brains of these mice. While a few parasites chyma, we have documented increased microglial activa- have been detected in the brain vasculature via electron tion in this model [16]. This is intriguing because glia microscopy [17], a more recent examination of the brain are found behind the multi-layered blood-brain barrier using highly sensitive luminescence technology to detect (BBB), while activated peripheral immune cells are luciferin-expressing P. chabaudi parasites did not show within the vasculature [16]. This prompted the question Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 3 of 17 of how the inflammatory cells within the vasculature coagulation in vascular congestion in CM and also impli- could amplify cytokine production in the absence of a cate a novel mechanism of inflammation-induced neuro- lymphoid structure, such as that developing in neuroim- pathology possibly initiated by leukocytes contained munopathologies with parenchymal infiltrates. within the vasculature. These findings may be relevant be- Congestion of the brain and retinal vasculature has been cause the inflammation-driven neuropathology in this documented in human cerebral malaria and is associated model shares many features with human cerebral malaria, with poor prognoses in human cases of CM [41, 42]. Sev- including intravascular leukocytes and thrombi, systemic eral factors are likely to contribute to congestion in human hyper-inflammation, edema, and death. patients: parasite sequestration, leukocyte adhesion, and co- agulation defects. Parasite-infected erythrocytes can both Methods bind to the vascular endothelium, leading to activation and Mice tm1Cgn vascular dysfunction, and activate the coagulation cascade C57BL/6J (WT) and B6.129P2-Il10 /J (IL-10 KO) [43, 44]. Coagulation defects are also seen in both murine mice (Jackson Laboratory, Bar Harbor, ME) were bred in experimental cerebral malaria and in human cerebral mal- The University of Texas Medical Branch Animal Re- aria [45–47] and can be promoted by the parasite itself source Center. Experimental mice were female and be- [45]. Vascular thrombi were observed in CM2 patients in tween 6 and 12 weeks of age at the time of infection. All Malawi, who are documented to have both sequestration animals were kept in a specific pathogen-free housing and cerebral hemorrhages [20]. This supports the finding with ad libitum access to food and water. Animals were that disseminated intravascular coagulation (DIC) was ob- cared for according to the Guide for the Care and Use of served in 19% of CM patients and correlated with poor Laboratory Animals under Institutional Animal Care outcomes [48]. However, the role of coagulation in neuro- and Use Committee-approved protocols. UTMB Animal pathology is obscured by contradictory outcomes in stud- Resource Center facilities operate in compliance with ies of the effect of the anticoagulant, heparin [49, 50]. In the USDA Animal Welfare Act, the Guide for the Care clinical trials, heparin significantly reduced death in a clin- and Use of Laboratory Animals, under OLAW accredit- ical trial in children with CM in Indonesia (from 13/17 to ation, and IACUC-approved protocols. 2/16, [50]) and reduced patient’s coma and hospitalization time [49]. However, it is not currently recommended for Parasite and infection treatment due to the potential of systemic hemorrhagic Frozen stocks of Plasmodium chabaudi chabaudi side effects of this older drug, suggested by work in (AS)-infected RBCs (iRBCs) (Jean Langhorne, Francis non-human primates [51] and case studies of malarious Crick Institute, London, UK) kept at − 80 °C were soldiers in Asia with pulmonary involvement [52], though thawed and injected intraperitoneally (i.p.) into WT not seen in clinical trials. The presence of monocytes and mice. Parasitized blood from these animals was diluted T cells in the brain vasculature [20], but not in the brain in Krebs-Ringer bicarbonate buffer (Sigma-Aldrich, St. parenchyma [34], is also documented. This has often been Louis, MO) and normal saline to deliver 10 iRBCs i.p. interpreted as a “lack of inflammation,” despite strong evi- in 200 μl into experimental WT or IL-10 KO mice. Thin dence, both genetic and serological, that cytokines play a blood smears were collected at regular intervals to critical role in killing parasite and inducing pathology [53]. monitor for peripheral parasitemia by staining with In an attempt to understand the role of adherent intra- Diff-Quik (Siemens Healthcare Diagnostics, Newark, vascular leukocytes and coagulation in promoting neur- DE) or Giemsa stain (Ricca Chemical Company, Arling- onal malfunction, we investigated the contents of ton, TX) and counted on a light microscope. congested vessels and their effects on the brain paren- chyma, as measured by gliosis. Furthermore, we tested Animal body temperature and weight the role of coagulation in pathology by studying the ef- Internal body temperatures were assessed daily during in- fect of anticoagulants on mortality and histological fea- fection using rounded stainless steel rectal probes and a tures of inflammation-driven neuropathology in P. BIO-TK8851 digital rodent model thermometer (Bioseb, chabaudi infection of IL-10 KO mice. We found that Pinellas Park, FL). Probes were sanitized with CaviCide thrombi were prevalent throughout the brain and coin- (Metrex Research Corp., Romulus, MI) between each use. cide with localization of adherent leukocytes. In Animal weights were measured using an OHAUS Scout addition, areas of coagulation and leukocytes Pro SP601 portable balance (OHAUS, Parsippany, NJ). co-localized with parenchymal gliosis. We also found a striking reduction of mortality and a significantly recov- Animal behavior evaluation ered parenchymal histology on elimination of coagula- Beginning on day 5 post-infection, daily assessments tion suggesting a pathological role for thrombi in this were performed on all animals using an abbreviated ver- model. These observations suggest an important role of sion of the modified SmithKline Beecham, Harwell, Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 4 of 17 Imperial College, Royal London Hospital Phenotype As- MA), or rabbit (catalog no. Z0334, Agilent Technologies, sessment (SHIRPA) protocol [54]. This brief behavioral Carpinteria, CA) anti-GFAP, mouse anti-CD11b biotin assessment was developed based on the full assessment (clone M1/70, catalog no. 13-0112-85, eBioscience, San in a previous study [16]. Higher scores were awarded for Diego, CA), and rat anti-CD45 biotin (clone 104, catalog measures showing higher functional ability. The proce- no. 13-0454-85, eBioscience, Sand Diego, CA). Secondary dures were carried out in an open testing environment antibodies used were goat anti-rat AlexaFluor-488 (catalog away from the home cage and took approximately 5 min no. A11006, Thermo Fisher Scientific, Waltham, MA) and per animal. goat anti-rabbit AlexaFluor-568 (catalog no. A11011, The abbreviated SHIRPA used involves a selection of Thermo Fisher Scientific, Waltham, MA). nine semi-quantitative tests for general health and sen- Streptavidin-FITC (catalog no. 11-4317-87, eBioscience, sory function, baseline behaviors, and neurological re- San Diego, CA) was used as a tertiary step for biotinylated flexes. We observed undisturbed behavior with the antibodies. CellTrace Violet (catalog no. C34557, Thermo mouse placed in an inverted beaker on top of a metal Fisher Scientific, Waltham, MA)-labeled CD4 T cells were grid suspended above the home cage for 3 min, during adoptively transferred into IL-10 KO mice for later which body position and spontaneous activity were co-localization with brain vasculature after i.v. perfusion assessed. Body position scores ranged from 0 (com- with DyLight488-labeled tomato lectin (catalog no. pletely flat) to 5 (repeated vertical leaping). Spontaneous DL-1174, Vector Laboratories, Burlingame, CA). Images of activity scores ranged from 0 (none) to 4 (rapid/dart immunohistochemistry (IHC) sections were taken with an movement). At the end of the observation period, palpe- Olympus IX 71 inverted brightfield microscope using a × bral closure, which was scored from 0 (eyes closed) to 2 20 air objective, while the immunofluorescence images (eyes wide open), and qualitative grip strength, scored were taken with a confocal microscope (Olympus FV 1000) from 0 (none) to 4 (unusually strong), are tested by ap- with the DAPI channel for nuclei, Alexa 488 channel for plying a gentle horizontal force on the animal’s tail as it Iba1 tagged with Alexa 488, and Alexa 647 channel for CD grips the metal grid. The animal is then placed in an 31 tagged with Alexa 647. IHC images of Iba1-stained sec- open arena in which several behaviors are measured. tions were contrast-enhanced and segmented by threshold Gait is observed as the animal traverses the arena and is for microglia using ImageJ (NIH, Version 1.48u). These scored from 0 (incapacity) to 3 (normal). During move- were used to create binary images. Individual microglia ment, tail elevation is scored, ranging from 0 (dragging) were identified using a semi-automatic algorithm employ- to 2 (elevated). Touch escape measures the reaction to a ing the particle analysis function on image and average area finger stroke and is scored from 0 (no response) to 3 (es- per microglia; the microglia density and total immunoreac- cape response to approach). Palpation of the animal’s tive area were calculated from the binary images. Area frac- sternum determines heart rate: 0 (slow) to 2 (fast), and tion of small processes is a ratio of immunoreactive area finally, righting reflex is scored by releasing the animal without microglia to total immunoreactive area which from an upside-down position near the surface and ob- indicates the degree of ramification. Transformation index, serving the responding effort to upright itself, scored and indicator of activation, was calculated as T-Index = from 0 (fails to right) to 3 (lands on feet). The expected (Perimeter )/(4π × Area) per microglia. To quantitatively score of a healthy, uninfected IL-10 KO or WT mouse is describe the degree of ramification, we calculated the area 22. A score of 15 was identified as the humane endpoint fraction of small thin processes to total immunoreactive based on the finding that any female animal that drops area. Ramification could be seen in IHC images as glia below that score by day 9 will succumb to infection (see with long and thin processes that appeared segmented due Additional file 1: Figure S1). to branching in and out of the tissue section plane. The astrocyte-thrombus association index was defined in which Histochemistry the ratio of X (the number of astrocytes contacting a Immunofluorescence of cryosections was examined after thrombus divided by the total number of thrombi) was cal- 48 h of post-fixation of mouse brains in 4% PFA and 72 h culated, and values were normalized based on the following of cryoprotection in 30% sucrose. Fixed frozen sagittal sec- equation, (X − X )/(X − X ), where X =1.3 (lower i min max min min tions (30 μm) were made using Tissue Plus® Optimal Cut- limit of astrocyte-thrombi interaction seen in uninfected ting Temperature Compound (Fisher Healthcare, IL-10 KO brains) and X = 3.25 (~ 75% astrocyte/thrombi max Houston, TX) and mounted on glass slides with Fluoro- association) approximated the lower and upper limit of as- mount mounting medium (Novus Biologicals, Littleton, trocytes interacting with thrombi based on our data. CO). Sections were incubated overnight at 4 °C with pri- mary antibodies rabbit anti-fibrinogen (catalog no. A0080, Cell and in vivo labeling Agilent Technologies, Carpinteria, CA), rat (clone 2.2B10, Some infected IL-10 KO and WT animals were injected 6 + catalog no 13-0300, Thermo Fisher Scientific, Waltham, with 2 × 10 CTV CD4 T cells 3.5 h before sacrifice Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 5 of 17 (i.p.) and 40 μg of DyLight488 labeled Lycopersicon escu- between two #1.5 cover glass. To visualize large regions of lentum (tomato) Lectin (catalog no. DL-1174, Vector La- optically cleared brain tissue using two-photon microscopy, boratories, Burlingame, CA) 20 min before sacrifice image stack mosaic and stitching were applied. Image stack (i.v.). CellTrace Violet (catalog no. C34557, Thermo stitching was done with a 10% overlap on a field of view of Fisher Scientific, Waltham, MA) labeling was performed 2327.3 × 237.3 μm providing 232.73 μm of co-registration as previously described [55]. in X and Y coordinates. Images were analyzed using ImageJ (FIJI), Olympus Fluoview FV1000-ASW 2.0 Viewer (con- Anti-TNF antibody treatment focal), Imaris Image Analysis Software (confocal and Mice receiving anti-TNF antibody (clone XT3.11, Bio X two-photon microscopy; Bitplane USA, Concord, MA), and Cell, West Lebanon, NH) were treated with 0.2 μg/day NIS Elements (confocal; Nikon Instruments, Melville, NY). for 5 days starting on day 5 post-infection (days 5–9). Positive fibrinogen and elevated GFAP staining in each field Untreated mice received isotype rat IgG1 as a control. was quantified by applying a signal intensity threshold and the percent area covered was calculated via the outlined CLARITY and optical clearing areas of positive staining that met the signal intensity Fixed brain sections (IL-10 KO and WT) were subjected threshold per field of view. The percentage of total area in- to the passive CLARITY optical clearing method [56] for cluded was calculated using ImageJ software (FIJI, NIH). large-scale labeling and imaging. In brief, mice were anesthetized and perfused transcardially with a mixture Ammonia assay of 4% (wt/vol) PFA, 4% (wt/vol) acrylamide, 0.05% (wt/ Tissue and serum ammonia was quantified using a com- vol) bis-acrylamide, and 0.25% (wt/vol) VA044 (hydrogel mercial colorimetric ammonia assay kit (ab83360, solution) in PBS. Brains were extracted and incubated in Abcam, Cambridge, MA). Briefly, brain and liver sam- hydrogel solution at 4 °C for 3 days. Solution ples were collected from infected IL-10 KO and WT temperature was then increased for 3 h to 37 °C to initi- mice at the peak of behavioral symptoms, washed in ate polymerization. Hydrogel-embedded brains were sec- cold PBS, resuspended in 100 μl assay buffer, and ho- tioned into 2-mm-thick sagittal sections and placed in mogenized using a Dounce homogenizer to produce clearing solution (sodium borate buffer, 200 mM, pH single-cell suspensions. After 2–5 min of centrifugation 8.5) containing 4% (wt/vol) SDS) for 3 weeks at 40 °C at 4 °C, cells were counted via hemocytometer and under gentle agitation. Samples were immunostained for seeded into a 96-well plate to provide 1–5×10 cells/ GFAP to assess astrogliosis. After immunostaining, sam- well. Serum samples were counted and seeded directly ples were optically cleared using increasing serial con- into plates without processing (5–10 μl/well). The col- centrations (10–100%) of 2,2′-thiodiethanol (TDE) in orimetric assay was conducted using OxiRed probe. Milli-Q water (EMD Millipore, Darmstadt, Germany) to Color change was recorded at OD 570 nm using a spec- achieve optimal refractive index matching with tissue. trophotometer microplate reader and compared to an ammonium chloride standard curve (detects 0–10 nmol/ Microscopy well) after 60 min of incubation at 37 °C. Fixed cryosections (30 μm thickness, fluorescent or con- focal microscopy) were imaged with a Nikon Eclipse 80i Statistics epifluorescence microscope and a Fluoview 1000MPE sys- Where indicated, groups were compared by t test (2 tem configured with an upright BX61 microscope (Olym- groups) or one-way ANOVA (3 or more groups), pus, Center Valley, PA). Fixed, CLARITY-processed followed by post hoc Bonferroni method or Tukey’s test sections (2 mm thickness, two-photon confocal micros- to identify significance between individual groups. Each copy) were imaged using a Prairie Ultima IV (Prairie point represents the average value per animal after ana- Technologies/Bruker, Middleton, WI) upright multipho- lysis of 10 fields, unless otherwise specified. Statistical ton microscope. For two-photon fluorescence microscopy, analysis was performed in Prism (GraphPad, La Jolla, a × 10 0.3 N.A. objective (UPLFL10X, Olympus) and a × CA), *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001. Error bars 25 1.05 N.A. super-objective (XLSLPLN25XGMP, Olym- represent ± SEM. pus) were used for image collection. Illumination for exci- tation of fluorescence was provided by a femtosecond Results laser (Mai Tai, SpectraPhysics, Santa Clara, CA) tuned to Congestion of brain blood vessels with thrombi + + + 800 nm. Fluorescence was collected using a two-photon containing CD45 , CD11b , and CD4 leukocytes in P. standard M filter set including filters with bandwidth 604 chabaudi-infected IL-10 KO mice ± 45 nm, a filter with bandwidth 525 ± 70 nm, and di- To investigate vascular abnormalities in P. chabaudi-in- chroic mirror cutoff at 575 nm. Samples were mounted fected IL-10 KO mice, we examined sagittal sections of on a 30-mm cage plate (CP06, ThorLabs, Newton, NJ) perfused and fixed brain tissue for evidence of vascular Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 6 of 17 leakage as indicated by extravascular fibrinogen at the systemic production of fibrinogen is a risk factor for co- peak of infection (day 8 post-infection). Brains from agulation, it does not lead to clotting by itself [58]. How- infection-matched, disease-resistant WT mice were used ever, an increase in liver fibrinogen production is not as controls (Fig. 1a). In addition to the expected sites of sufficient for accumulation of fibrin, which is triggered by perivascular fibrinogen (evidence of fibrinogen leakage), the coagulation cascade [57, 58]. we also found foci of fibrin(ogen) staining within the Studies of both human CM and murine experimental vascular lumen of brain blood vessels in IL-10 KO mice. cerebral malaria (ECM) have documented congestion of As we had performed transcardial perfusion prior to sac- the brain and retinal vasculature, but the role of thrombi in rifice, this data is suggestive of intravascular thrombi. reduced blood flow is not clear. By imaging through Quantification of fibrin(ogen) staining in the IL-10 KO 200 μm of tissue, we found that both large and small vessels mice showed an increase in the area of the brain with retain intravascular fibrin(ogen) (Fig. 2a), often to the point bright fibrinogen immunoreactivity (percent area of Alexa of complete occlusion of the vascular lumen (Fig. 2b), rem- Fluor 568 pixels, 10 fields/mouse) compared to infected iniscent of thrombosis. The coagulation cascade leads to WT, or uninfected, which were indistinguishable from cleavage of fibrinogen into fibrin during the formation of a each other (Fig. 1b). There was also a large increase in clot [59]. The polyclonal antiserum used to detect fibrino- staining of fibrinogen in the livers of infected IL-10 KO gen here also detects fibrin and other degradation products compared to WT, which had some lighter staining as well of fibrinogen [60, 61]. Therefore, we interpret this staining that was not quantifiable over background levels in unin- pattern to represent fibrin clots. The appearance of sherical fected mice (Fig. 1c). This could potentially be due to an gaps in fibrin staining led us to hypothesize that in addition increase in fibrinogen production by the IL-10 KO mouse to red blood cells and platelets, immune cells could be downstream of inflammation, as fibrinogen is an acute retained within the thrombi of congested vessels. In order phase response protein [57]. However, while increased to identify them, we stained IL-10 KO brains for the Fig. 1 IL-10 KO mice have residual fibrin deposition in and around brain vasculature and increased liver fibrinogen. a Confocal images (× 20) showing immunofluorescent staining of fixed, frozen brain sections (30 μm) from P. chabaudi-infected IL-10 KO and WT mice (day 8 p.i., n =4 mice/group). Fibrin (red) and tomato lectin (green, vascular endothelium). b Fibrin (red) was quantified by surveying 10 fields per brain section (× 10). Graph showing average percent area of fibrin-positive staining above threshold in each field. c Immunofluorescent staining (× 10) and quantitation of fibrinogen (red) in liver from infected IL-10 KO, WT, and uninfected controls (n = 4 mice/group). One-way ANOVA, followed by post hoc Bonferroni method, was used to determine statistical significance. *p < 0.05, **p < 0.01. Scale bar represents 100 μm Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 7 of 17 Fig. 2 Vascular congestion in IL-10 KO mice with malaria includes thrombi-containing monocytes and T cells. Immunofluorescent staining of fixed, frozen brain sections (30 μm) from P. chabaudi-infected IL-10 KO mice (day 8 p.i., n =4 mice). a Confocal images (× 40) of IL-10 KO brain stained for fibrin (red). b Successive single-plane confocal images (× 40) of a 30-μm z-stack showing complete occlusion of a large vessel with residual fibrinogen (red). c Immunofluorescence staining of IL-10 KO brains showing fibrin staining of blood vessels (red) and leukocytes expressing CD45 (green, × 60) and d CD11b (green, × 40). e CTV CD4 T cells (blue) from infected IL-10 KO mice were adoptively transferred into infection-matched IL-10 KO (day 7 p.i.) recipients 3.5 h before sacrifice. Frozen brain sections (day 7 p.i.) were stained for fibrin (red). Max intensity projection of a 30-μm z-stack (× 240) displayed from brain tissue of IL-10 KO mice co-stained with WT control samples (n =3–4 mice per group). Scale bars represent 30 μm(a), 50 μm(b–d), or 10 μm(d) pan-leukocyte marker, CD45 (Fig. 2c), and the monocyte experimental cerebral malaria [62], we analyzed the ex- marker, CD11b (Fig. 2d). Staining showed that many, but tent of astrocyte activation in IL-10 KO mice infected + + not all, CD45 and CD11b leukocytes are contained within with P. chabaudi. In order to visualize extensive activa- areas of residual fibrinogen staining. We previously quanti- tion of astrocytes, we utilized CLARITY followed by op- fied CD11b cells within the brains of P. chabaudi-infected tical clearing, a tissue processing technique which IL-10 KO mice using flow cytometry. In that analysis, we removes relatively opaque lipids, transforming thick sa- + + showed that the CD11b cells were also Ly6C , indicating gittal brain sections (2 mm) to make them optically that they are inflammatory monocytes [16]. There was a transparent. This process diminishes excess light scatter- hi large and significant increase in cerebral Ly6C inflamma- ing during image acquisition by confocal or two-photon tory monocytes in IL-10 KO compared to that in infected microscopy, allowing for increased imaging depth be- int WT mice, while a Ly6C population of resident macro- yond that possible in unprocessed tissue. The ability to phages was not increased. obtain image stacks over the full 2 mm thickness com- We were also interested to see if CD4 T cells, the pri- bined with image stitching allowed for image acquisition mary producers of IL-10 in this infection, were also found of the entire thick sagittal section. Whole brain sections localized with fibrin(ogen) in the vessels. Therefore, CD4 stained for glial fibrillary acidic protein (GFAP), which is T cells (CellTrace Violet ) from IL-10 KO mice 7 days upregulated on activated astrocytes, were imaged to de- post-infection (p.i.) were adoptively transferred into termine the extent of astrocyte activation in susceptible infection-matched IL-10 KO recipients, which underwent IL-10 KO mice (Fig. 3a, c, e) and resistant WT animals transcardial perfusion and brain tissue collection 3.5 h (Fig. 3b, d, f). A higher GFAP signal was observed in later. Transferred CD4 T cells were indeed identified in multiple areas of the IL-10 KO brain compared to WT, the brain, and often within a fibrin(ogen) clot (Fig. 2e). including the hippocampus, thalamus, and caudate puta- While the number of leukocytes is not large, activated leu- men, suggesting astrocyte activation via increased pro- kocytes have the potential to promote activation of the duction of inflammatory cytokines (Fig. 3a, b). While neuroglial cells surrounding the vasculature, namely, as- GFAP is expressed on most astrocytes, even in unin- trocytes. Therefore, we next tested brain sections from in- fected animals, the level of expression is significantly fected IL-10 KO animals for astrogliosis. lower than on activated astrocytes [63]. Interestingly, there was little GFAP signal in the cortex, a result that is Inflammatory cytokine TNF induces astrocyte activation in consistent with findings in human CM autopsy [20]. For clusters near thrombotic cerebral vasculature in IL-10 KO quantitation of astrogliosis, we focused our analysis on mice with malaria the hippocampal formation (Fig. 3c, d), as a representa- As astrocytes play an important role in maintaining the tive region in which astrogliosis was evident. This region integrity of the BBB, including in the context of is amenable to being isolated from other regions by Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 8 of 17 Fig. 3 Increased astrocyte activation in IL-10 KO mice with malaria. Mice were infected with P. chabaudi and sacrificed 8 days post-infection. Thick sagittal brain sections (2 mm) were CLARITY-processed, optically cleared, stained with GFAP (red), and imaged by two-photon confocal microscopy. a, c, e IL-10 KO and b, d, f WT brains from the peak of P. chabaudi infection (day 8 p.i., n = 5 mice/group). a, b Single fields of the entire tissue section (× 10) stitched together. c, d Hippocampus of the thick brain section is masked for increased resolution and quantitation in c IL-10 KO and d WT animals (n = 3 mice/group). e, f Representative high-resolution image (× 25) of astrocytes from the hippocampus showing the e IL-10 KO and f WT control brains. g Quantification of the percent area of astrocyte staining above threshold in the hippocampal formation of the P. chabaudi-infected IL-10 KO and WT brains. Number of fields for IL-10 KO (n = 15) and WT (n = 9). Scale bars represent 1 mm (a, b), 200 μm (c, d), and 50 μm(e, f). Student’s t test was used to determine statistical significance. **p < 0.01 image processing due to its well-defined margin and thus congestion characterized by fibrin staining and astrocyte bright allowed for comparison of GFAP cells in the full vol- activation in this hyper-inflammatory response, we next ume of the hippocampal region in each section. As shown sought to determine the role that inflammatory cyto- in high-resolution 3D micrographs (Fig. 3e, f), in addition kines play in this process. to upregulation of GFAP, astrocytes in IL-10 KO mice Immunopathology in IL-10 KO mice infected with P. showed distinct morphological changes, appearing hyper- chabaudi is generated by the hyper-inflammatory cyto- trophied and with more processes compared to infected kine response generated in the absence of this regulatory bright WT. The GFAP fraction of the hippocampal forma- cytokine primarily made by T cells [32]. Neutralizing tion in infected IL-10 KO mice was significantly increased TNF is known to improve survival and also improve all compared to WT mice (Fig. 3g). While elevated serum measures of symptomatic pathology, while Ifngr1 defi- ammonia from potential liver damage can activate astro- ciency in IL-10 KO mice improves survival [14, 31]. Im- cytes [64], there was no significant difference in ammonia portantly, neutralizing the other major regulatory production between WT and IL-10 KO mice (Add- cytokine, transforming growth factor-β, increases mor- itional file 2: Figure S2). As inflammation, or vascular tality of the IL-10 KO to 100%, suggesting that the bal- damage, can also lead to astrocyte activation, we next in- ance of inflammatory and regulatory cytokines in the vestigated whether vascular congestion and astrocyte acti- immune response to malaria infection determines the le- vation occurred in close proximity. thality of P. chabaudi in IL-10 KO mice [14]. However, To investigate the potential connection between vas- the role of TNF in brain pathology, including its behav- cular congestion and astrocyte activation, we performed ioral results, has not yet been investigated in this model. immunofluorescent staining of peak-infected (day 7 p.i.) As an indication of brain pathology, we used a and uninfected IL-10 KO brains for fibrin(ogen) and semi-quantitative P. chabaudi-specific SHIRPA health astrocyte activation. In the hippocampal formation, we assessment abbreviated from one we previously de- observed an increase in residual fibrin(ogen) staining in scribed [16]. We have now identified a smaller set of be- the infected IL-10 KO brains compared to WT (Fig. 4). havioral symptoms, described in the “Methods” section, Interestingly, the astrocytes showed an increase in GFAP that specifically change at the time that IL-10 KO mice staining and polarity and were more frequently found in begin to succumb to infection. The SHIRPA screen was contact with fibrin-containing vessels in infected IL-10 highly predictive of outcome, as the SHIRPA scores of KO brains compared to infected WT and uninfected mice that died during infection were significantly lower IL-10 KO controls (Additional file 3: Figure S3). How- than those of mice that survived (Additional file 1: Fig- ever, it was noted that not all areas with residual fibrin ure S1). In addition, we were able to use the abbreviated staining were located near highly activated astrocytes. SHIRPA to identify animals predicted to succumb to Uninfected mice showed neither residual fibrinogen de- hyper-inflammatory experimental cerebral malarial dis- position, nor an increase in GFAP immunoreactivity. ease. Any P. chabaudi-infected IL-10 KO mouse that Having established a link between microvascular scored below 17, out of a maximum of 22, on the Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 9 of 17 Fig. 4 Activated astrocytes cluster along thrombus-containing brain vasculature. IL-10 KO mice were either infected with P. chabaudi and sacrificed 8 days post-infection or used as uninfected controls. Representative epifluorescence images (× 20) of the hippocampal formation in cryosections (30 μm) from infected (day 8 p.i.) IL-10 KO brains (left, middle) and uninfected IL-10 KO brains (right) immunostained for GFAP (green), fibrinogen (red), and DAPI (blue). IL-10 KO mice were co-stained with WT control samples (n =5–6 mice per group). Scale bars represent 50 μm abbreviated SHIRPA screen before day 9 post-infection had with P. chabaudi, we treated infected IL-10 KO mice a statistically significant chance of succumbing to infection, with the anticoagulant drug, enoxaparin sodium, a with an odds ratio of 23.7 (95% CI 4.0–126.0, χ test), low-molecular weight heparin (LMWH), starting on day meaning they had almost 24 times more probability to suc- 4 post-infection through to the end of peak illness at day cumb to disease. However, two out of 49 mice (4.1%) that 12 post-infection, when all control animals had died. were predicted to die actually survived. In addition, due to Mice were treated twice per day and monitored using the speed of progression from undetectable morbidity to the abbreviated SHIRPA screen. Blood smears were also mortality, some animals (11/28, 39%) will die naturally collected on day 9 post-infection to monitor parasite without ever exhibiting a low SHIRPA score. burden. Strikingly, LMWH treatment of IL-10 KO mice To test the role of TNF in neuroimmunopathology and rescued them from fatal neurologic disease before day 9 astrocyte activation in this infection, we treated IL-10 KO post-infection (Fig. 6a). However, LMWH-treated IL-10 mice with neutralizing anti-TNF antibody or isotype control KO mice were still susceptible to delayed mortality, as antibody for 5 days (days 5–9p.i.) [14]. To monitor for fi- two out of four ENO-treated mice (50%) died after day 9 brinogen accumulation and astrocyte activation, mice were post-infection. This may represent death from severe sacrificed at day 8 p.i., at the onset of severe disease, and anemia that typically presents after the peak of P. cha- brain tissue was stained for confocal microscopy. We ob- baudi infection [65]. The differential mortality between served an increase in astrocyte activation and increased re- treatment groups was not due to differences in parasit- sidual fibrinogen in isotype-treated IL-10 KO animals emia at the peak of infection on day 9 p.i., while behav- (Fig. 5a), but neither of these changes were observed in the ioral scores were significantly improved with LMWH IL-10 KO group treated with neutralizing anti-TNF anti- treatment (Fig. 6b). As a control to assure the quality of bodies (Fig. 5b), similar to isotype-treated WT mice the treatment, we quantitated fibrinogen deposition in (Fig. 5c). These changes were significant, with a complete the brains of treated animals and confirmed that LMWH reductioninfibrinogen accumulation (Fig. 5d) and astro- eliminated thrombi completely (Fig. 6c). Strikingly, we cyte activation (Fig. 5e). Furthermore, animals were pro- found that astrogliosis was significantly reduced by anti- tected from behavioral symptoms during anti-TNF coagulant treatment, though not to the levels seen in un- treatment (Fig. 5f). Behavioral symptoms declined after infected animals (Fig. 6d). In conclusion, LMWH treatment stopped, but we did not observe any late mortal- treatment decreased astrocyte activation and intravascu- ity. As expected, excess production of fibrinogen in the liver lar fibrin clotting, suggesting that thrombi in cerebral was also reduced by anti-TNF treatment (Fig. 5g). As vasculature play a critical role in astrogliosis and lethal anti-TNF blocks many components of the acute phase re- pathology from malaria without affecting parasitemia. action besides coagulation, we proceeded with more spe- Microglia are important sentinels and potent ampli- cific tests for the importance of coagulation to fiers of inflammation within the CNS. In response to en- hyper-inflammatory experimental cerebral malaria. vironmental cues and inflammatory stimuli, microglia become activated and undergo characteristic morpho- Anticoagulant treatment eliminates early mortality and logical changes. Therefore, we quantified both upregula- reduces glial cell activation in IL-10 KO mice with malaria tion of Iba1, a marker of activation, and morphological To test the hypothesis that thrombi contribute to the changes characteristic of microglial activation in brain fatal neurological phenotype of IL-10 KO mice infected sections from either uninfected or P. chabaudi-infected Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 10 of 17 Fig. 5 Anti-TNF antibody treatment prevents astrocyte activation and mortality in IL-10 KO mice with malaria. Mice were infected with P. chabaudi and followed throughout the acute phase of infection (day 12 p.i.) or sacrificed 8 days post-infection for immunofluorescent staining. One group of IL-10 KO mice received anti-TNF IgG treatment (n = 5), while another group of IL-10 KO mice (n = 5) and a group of WT mice received isotype IgG as control (n =5). a Representative confocal images (× 20) of cryosections stained for astrocytes (GFAP; green) and fibrinogen (red) with DAPI (blue) in sagittal brain sections in anti-TNF antibody-treated IL-10 KO mice, b isotype IgG-treated IL-10 KO mice, c and isotype IgG-treated WT mice. d Brain fibrinogen and e GFAP staining for reactive astrocytes in the hippocampus were quantified by calculating the percent area per field of immunostaining above signal threshold. Ten fields per animal were assessed, with the graph showing the mean value per animal. f General behavior as measured by the abbreviated SHIRPA screen of anti-TNF antibody-treated (IL-10 KO, n = 5) and isotype IgG-treated (IL-10 KO, n =5; WT, n = 5) mice infected with P. chabaudi. Green arrows represent the dosing schedule of either anti-TNF IgG or isotype control IgG. g Liver fibrinogen quantitation. Data shown is representative of two independent experiments (n = 9 total mice/group). One-way ANOVA, followed by post hoc Bonferroni method, was used to determine statistical significance. *p <0.05, **p < 0.01, ***p < 0.001. Scale bars represent 50 μm mice on day 8 p.i. (Fig. 7a). We observed dramatic four quantitative assessments: (1) total immunoreactive changes in the microglia in the IL-10 KO compared to area (% of total Iba1-positive pixels in a field); (2) aver- WT, and we observed further changes in the age immunoreactive area per microglia; (3) transform- anticoagulant-treated animals. To interpret these ation index, a measure of microglial ramification; and changes, we quantitated the extent of microglial activa- (4) area fraction of small processes, which is normalized tion in these images based on morphology. We used to total immunoreactive area. The latter was done to Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 11 of 17 Fig. 6 IL-10 KO mice are rescued from fatal neurologic disease with LMWH treatment. a Two groups of IL-10 KO mice (n = 4) were either treated with 1000 IU/kg (20 IU/dose) enoxaparin Na (ENO) i.p. twice a day (12 h apart) or given saline starting at day 4 post-infection until the middle of the anemic period of disease (day 12 post-infection). b Survival was monitored daily, and blood smears were collected on day 9 post-infection. Behavior was monitored daily using the abbreviated SHIRPA screen (n = 4 mice/group). c Fibrinogen quantification in the brains of untreated and LMWH-treated mice at the peak of infection (day 9 post-infection, n = 4 mice/group). d 30-μm brain hippocampus cryosections stained for astrocytes (GFAP, green). GFAP staining quantified by calculating the percent area per field of immunostaining above signal threshold. One-way ANOVA, followed by post hoc Bonferroni method was used to determine statistical significance. *p < 0.05, **p < 0.01. Scale bars represent 50 μm capture differences in small/fragmented processes, as The morphological changes in infected IL-10 KO mice small processes were not observed in the IL-10 KO group, show significant changes in microglial activation state, while they were present in the LMWH group, although suggestive of increased intracranial inflammation. Interest- not as numerous as the WT group (Fig. 7b). The last ingly, all features of activation show significant improve- graph, therefore, shows how much Iba1-reactive area each ment towards homeostasis after clearance of thrombi group has with respect to the area occupied by microglia following LMWH treatment. Therefore, these findings soma, which was significantly lower in the untreated IL-10 demonstrate a critical role of inflammation-driven coagu- KO group. We interpret this to mean that activated lation in experimental cerebral malaria pathology. microglia retract their dendrites, which then appear thicker, as opposed to the thinner processes that cover Discussion more three-dimensional area in homeostasis. All of these The presence of peripheral immune cells adherent measures suggest that microglial activation is reduced, but within the vasculature in mouse models of CM and in not back to homeostatic levels, by LMWH treatment, brain vessels on autopsy of cerebral malaria patients [66] similar to what we found for astrogliosis above. suggest that such cells play an important role in mediat- In order to determine the relative localization of acti- ing neuropathology [67]. Current paradigms to explain vated microglia and cerebral vasculature, immunofluor- CM pathogenesis support an important role for inflam- escent staining was performed on microglia (Iba1) and mation in the generation and amplification of neuro- CD31 blood vessels (Fig. 7c). We observed increasing pathology but do not explain the derivation of these microglial polarity and thickening of dendrites in IL-10 cytokines in the brain. The derivation and contribution KO animals, with decreased numbers of small processes of cerebral thrombi to CM pathology is also poorly in the microglia of untreated IL-10 KO mice. The understood. The vascular findings in this study suggest- localization of microglia near vessels in infected animals ive of pervasive (Fig. 1) and complete (Fig. 2) blockade is clearly seen when viewed as a 3D stack. Enumeration of the vasculature by inflammation-induced thrombi are of the number of microglia that interacted with a blood striking. These abnormalities have not been described in vessel, defined as either body or process on the blood P. chabaudi infection before. Coagulation is clearly of vessel, indicated 79% of glia interacted with a vessel in major relevance for our understanding of pathological the KO group vs. 54% in the WT (p < 0.05), and while mechanisms in cerebral malaria [21, 58, 68]. Potentially mean value for LMWH-treated IL-0 KO mice was 69%, pathogenic serum levels of both pro- and anticoagula- it was not statistically significant from either KO or WT. tion proteins have been documented in human CM [69, Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 12 of 17 Fig. 7 Microglia changes in IL-10 KO mice infected with P. chabaudi. a Representative images of day 8 p.i. WT, IL-10 KO, and LMWH-treated IL-10 KO mice (n = 4 mice/group) 30-μm brain cryosections stained with anti-Iba-1 antibodies and visualized using DAB. b Quantitative analysis of microglia morphology in WT, IL-10 KO, and LMWH-treated IL-10 KO mice using ImageJ software. c Immunofluoresence imaging of microglia (Iba-1-Alexa 488, green), endothelial cells (CD31-Alexa 567, red), and nuclei (DAPI, blue) in 30-μm brain cryosections from WT, IL-10 KO, and LMWH-treated IL-10 KO mice during the peak of infection. Right, 3D reconstruction showing the spatial orientation of microglia cells in relationship to microvasculature in a P. chabaudi-infected IL-10 KO mouse. One-way ANOVA, followed by post hoc Tukey’s test, was used to determine statistical significance. **p < 0.01. Scale bars represent 20 and 50 μm 70]. Systemic inflammation has also recently been shown formation and activation of cells in the brain paren- to contribute to intravascular clotting via mechanisms chyma in the absence of local parasite adhesion. Studies involving neutrophils and monocyte interaction with of Plasmodium berghei (ANKA) (PbA) infection have platelets in CM [71, 72], linking inflammation and clot- established the importance of the inflammatory response ting, which in turn promote sequestration. Recent stud- in the development of neurocognitive dysfunction [74– ies also show that the anticoagulation endothelial 76]. PbA infection shows pathogenic immune cell accu- protein C receptor (EPCR) may bind the parasite and be mulation in cerebral blood vessels as a result of inflam- downregulated, thus promoting clotting and suggesting matory TNF and IP-10 secretion [77, 78] and a mechanism for the induction of coagulation by P. fal- intercellular adhesion molecule-1 (ICAM-1) on the vas- ciparum sequestration [45, 73]. Interestingly, studies cular endothelium [79]. PbA infection has also been point to the bi-directional amplification of the clotting shown to induce astrocyte activation and degeneration cascade and inflammation suggesting an important inter- near sites of monocyte vascular adhesion [62, 80]. How- section that is likely to be crucial to pathology in CM ever, the signals leading to the breakdown of local astro- [58]. cyte barrier function in malaria have not yet been The data presented here confirm that inflammatory defined. The activation of astrocytes is a feature of many cells within the vasculature can drive both clot neurological diseases, including cerebral malaria [81, 82]. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 13 of 17 Our results demonstrate a causal link between 91]. Selection of treatments with relatively moderate hyper-inflammation, hyper-coagulation, glial cell activa- anticoagulation activity is likely essential to achieving tion, and mortality (Figs. 3 and 4). Gliosis across mul- therapeutic goals while avoiding hemorrhagic complica- tiple areas of the brain was observed in infected IL-10 tions. LMWH, as the name implies, involves only the ac- KO mice, with astrocytes and microglia associating tivity of the smaller heparin proteins, which act with highly with the vasculature compared to the WT higher specificity on factor Xa, exhibit less thrombin in- group—yet both microglial and astrocyte gliosis were hibition, and produce a more reliable therapeutic profile. significantly reduced upon LMWH treatment, indicating Our studies show that LMWH treatment is protective this direct link. within the context of hyper-inflammatory cerebral mal- This is important because resolution of CM in African aria and prevents intravascular thrombi formation in the children and Asian adults can be resistant to brains of mice exhibiting behavioral dysfunction (Fig. 6). anti-malarial drug treatment, suggesting that parasite This is particularly important in that both astrocyte and alone does not cause the full cerebral malaria syndrome. microglial activation were dependent on this coagulation Furthermore, it is not yet clear how parasite adhesion event to some degree (Figs. 6 and 7). Activation of alone drives the neuropathology evident from patient microglia has been shown to be an important compo- symptoms [83]. However, because of the overlap of in- nent of neuroinflammation and behavioral dysfunction flammation with parasite-dependent factors, determining associated with PbA infection [92–94]. Widespread the independent contributions of each presents an on- microglial activation, not always restricted to areas of going challenge to investigators. The impact of parasite parasite sequestration, has also been identified in cases adhesion to the vascular endothelium on coagulation, of human CM [95, 96]. However, these findings are vascular integrity, and congestion has been shown in in novel in the context of P. chabaudi infection. Further- vitro endothelial cultures and animal models of cerebral more, the spatial relationship of intravascular coagula- malaria [19, 43, 67, 84, 85]. Sequestration is seen in most tion with glial cell activation is also previously unknown fatal pediatric and adult CM cases [20, 21] and is used in any malaria infection and should be examined in hu- as a critical hallmark of disease. We chose to study the man CM autopsy samples. role of inflammatory cytokines in isolation from the po- Efforts to manipulate the inflammatory response and tential contribution of sequestration using an clotting cascade have provided mixed results in clinical inflammation-induced cerebral malaria model. The re- trials to date [97–99], highlighting the importance of un- sults confirm that inflammation can cause many of the derstanding the interactions between various arms of the pathological changes seen in CM, though not all. host response within the pathogenesis of cerebral mal- In this study, we show that both the congestion aria. In summary, our experiments support the import- phenotype associated with intravascular clotting and ance of intravascular coagulation and leukocytes astrocyte activation can be reversed via neutralization of producing inflammatory cytokines in malaria-induced TNF (Fig. 5), or anticoagulant therapy (Fig. 6). Serum cerebral pathology. The activation of surveilling microglia TNF concentration correlates with severity of human and vascular/neuronal-supportive astrocytes downstream malaria [86]. However, TNF blockade has thus far of systemic inflammation could promote the generation of proven ineffective in preventing death in childhood cere- neuropathology secondary to malaria infection. Identifica- bral malaria [87, 88]. As different reagents displayed dif- tion of both T cells and monocytes within fibrin clots sug- ferential effects, the timing, dose, or precise antigenic gests a new working model where inflammatory cells specificity of treatments may yet be improved for adju- promote cerebral damage even from their localization vant therapy. Strikingly, these data also show that fatal within the cerebral vasculature. It is possible that leuko- neurological disease in IL-10 KO mice is dependent on cytes within the structure of intravascular thrombi serve intravascular coagulation, as it can be prevented by to amplify pathological inflammatory cytokines leading to LMWH treatment (Fig. 6). This demonstrates a central immunopathology in the brain. These data demonstrate role for thrombi in driving the disease mortality and the interaction of the anti-parasitic and hemostatic ele- promoting neuropathology in P. chabaudi infection of ments of host defense, promoting a new appreciation of IL-10 KO mice. As anti-TNF and anticoagulants have the interplay between mechanisms important for develop- similar effects in this model, it is likely that cytokines ment of fatal cerebral malaria. and the coagulation cascade promote each other, as in other systems. Despite the WHO recommendation Conclusions against the use of heparin since 1984, citing excessive Our study has identified intravascular thrombi within bleeding [89], there are several clinical trials showing the cerebral vasculature during severe P. chabaudi infec- significant beneficial effects of anticoagulant usage on tion and showed that they contribute to lethal immuno- mortality and length of coma in human CM [49, 50, 90, pathology. Furthermore, vascular congestion with an Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 14 of 17 accumulation of leukocytes is spatially associated with Acknowledgements We would like to acknowledge Andrew S. Mendiola for suggesting the astrocyte and microglial activation in this model, with fibrinogen stain and for the immunofluorescence protocols from the the former being driven by TNF. The most striking find- Cardona laboratory. We appreciate Ping Wu, Tiffany Dunn, Yongquan Jiang, ing is that dissipation of these inflammatory foci within Paula Villarreal, and the Biomedical Imaging Network at UTMB for the technical assistance in the sample prep and confocal microscopy. We fibrin-rich thrombi by LMWH treatment leads to a sig- appreciate the feedback from the UTMB Joint Immunology Lab Meeting and nificant decrease in early lethal pathology. These find- the Neuro-Infectious Diseases working group organized by the UTMB ings begin to define the parameters of inflammation in Institute for Human Infection and Immunity, particularly Kathryn Cunningham and Kelly Dineley. We are also indebted to Linsey Yeager for the brain during cerebral malaria, and the downstream the critical review of the manuscript. We are grateful for the excellent animal pathology linked to hyper-inflammation. Previously, care by the UTMB Animal Resource Center and the training from the Rodent findings of cytokine gene linkage to CM were under- In Vivo Assessment Core. stood in terms of increasing parasite binding within the Funding capillary bed. Our findings demonstrate that inflamma- Funding for these studies was from the UTMB Institute for Human Infections tory cytokines contribute both pathogenic coagulation and Immunity (IHII) and the John S. Dunn Foundation (RS, KDW), McLaughlin and activation of sentinel glia in the brain parenchyma, Endowment Predoctoral Grant and Jeane B. Kempner Award (KDW), and NIH (R01AI363327 (RS), R01NS078501 (AEC), and T32AI363327 (KDW)). This study which are capable of causing neurological sequelae, even was also supported by a grant from the University of Texas System in the absence of localized sequestration, although to a Neuroscience and Neurotechnology Research Institute (RS, GV, #363327). The lesser degree than more virulent parasites. These find- funders had no role in the study design, data collection and interpretation, or the decision to submit the work for publication. ings, therefore, contribute to the current understanding of the etiologies of cerebral pathology and neurovascular Availability of data and materials abnormalities in malaria infection. While the effective- All data generated or analyzed during this study are included in this ness and safety of this approach must be validated, the published article and its supplementary information files. positive effect of anticoagulants could inform develop- ment of future adjunctive therapy for CM patients. Authors’ contributions Epifluorescent and confocal imaging studies were designed by KDW, GV, RS, conducted by KDW, with assistance from LFO and ODS, and analyzed by KDW, GV, and RS. Two-photon imaging studies were designed by GV and RS, Additional files conducted by LFO and ODS, and analyzed by LFO, ODS, and GV. Fibrinogen and microglia imaging studies were designed by KDW, RS, and AEC, con- ducted by KDW and SMC, and analyzed by KDW, RS, GV, LFO, and RP. Adop- Additional file 1: Figure S1. IL-10 KO mouse behavioral scores are tive transfer, antibody treatment, and animal behavior studies were designed predictive of outcome during P. chabaudi Infection. Left: representative by KDW and RS, conducted by KDW, and analyzed by KDW and RS. Tissue experiment showing SHIRPA scores of infected male IL-10 KO mice ammonia quantitation were performed and analyzed by VHC and ODS. grouped by eventual outcome (survived = blue, n =6; died = black, n =5). Statistical analyses and figures were generated by KDW, LFO, ODS, VHC, and Right: graph of the lowest abbreviated SHIRPA score in individual mice before PHK. KDW and RS drafted the manuscript, with essential input from GV, LFO, day 9 post-infection with infected IL-10 KO mice grouped according to ODS, and AEC. All authors read and approved the final manuscript. outcome. Showing concatenated data from multiple experiments (n =48). Error bar represents SEM, ***p < 0.001, Wilcoxon signed rank test. (TIF 326 kb) Additional file 2: Figure S2. IL-10 KO mice ammonia levels are not Ethics approval elevated above WT during P. chabaudi infection. WT and IL-10 KO mice Animals were cared for according to the Guide for the Care and Use of were infected with P. chabaudi and monitored during the peak of Laboratory Animals under the Institutional Animal Care and Use Committee- infection. WT mice were sacrificed at the peak of infection (day 10 p.i.) approved protocol #1006031. UTMB Animal Resource Center facilities operate in and IL-10 KO mice upon severe morbidity as determined via SHIRPA compliance with the USDA Animal Welfare Act, the Guide for the Care and Use of score. Organ and plasma ammonia levels were measured using a Laboratory Animals, under OLAW accreditation, and IACUC-approved protocols. colorimetric ammonia assay. (TIF 218 kb) Additional file 3: Figure S3. P. chabaudi-infected IL-10 KO mice show Competing interests astrocyte association with thrombi. Left, normalized astrocyte-thrombus The authors declare that they have no competing interests. association ratio. Right, representative confocal images of experimental groups stained for astrocytes (green) and fibrinogen (red). N =3–5 mice/ group. Error bar represents 30 μm. (TIF 12464 kb) Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details Abbreviations Department of Microbiology and Immunology, University of Texas Medical 3D: Three-dimensional; BBB: Blood-brain-barrier; CM: Cerebral malaria; Branch, 301 University Boulevard, Galveston, TX 77555, USA. Center for CNS: Central nervous system; CXCR3: C-X chemokine receptor 3; Biomedical Engineering, University of Texas Medical Branch, 301 University DIC: Disseminated intravascular coagulation; ECM: Experimental cerebral Boulevard, Galveston, TX 77555, USA. Department of Biology, One UTSA malaria; GFAP: Glial fibrillary acidic protein; i.p.: Intraperitoneal; ICAM- Circle, University of Texas at San Antonio, San Antonio, TX 78249, USA. 1: Intracellular adhesion molecule-1; IFN-γ: Interferon gamma; Department of Neuroscience and Cell Biology, University of Texas Medical IHC: Immunohistochemistry; IL-10 KO: IL-10-deficient; iRBCs: Infected red Branch, 301 University Boulevard, Galveston, TX 77555, USA. Department of blood cells; MHC-II: Major histocompatibility complex class II; Internal Medicine, Division of Infectious Diseases, University of Texas Medical PbA: Plasmodium berghei (ANKA); SHIRPA: SmithKline Beecham, Harwell, Branch, 301 University Boulevard, Galveston, TX 77555-0435, USA. Institute Imperial College, Royal London Hospital Phenotype Assessment; TNF: Tumor for Human Infections and Immunity, University of Texas Medical Branch, 301 necrosis factor; WT: Wild-type, C57Bl/6J University Boulevard, Galveston, TX 77555, USA. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 15 of 17 Received: 14 February 2018 Accepted: 17 May 2018 congestion are associated with coma in human cerebral malaria. J Infect Dis. 2012;205:663–71. 22. Jensen AT, Magistrado P, Sharp S, Joergensen L, Lavstsen T, Chiucchiuini A, Salanti A, Vestergaard LS, Lusingu JP, Hermsen R, et al. Plasmodium References falciparum associated with severe childhood malaria preferentially expresses 1. World Malaria Report 2016. Geneva: World Health Organization; 2016. http:// PfEMP1 encoded by group A var genes. J Exp Med. 2004;199:1179–90. www.who.int/malaria/publications/world-malaria-report-2016/report/en/. 23. Bernabeu M, Danziger SA, Avril M, Vaz M, Babar PH, Brazier AJ, Herricks T, Accessed 13 March 2018. Maki JN, Pereira L, Mascarenhas A, et al. Severe adult malaria is associated 2. Seydel KB, Kampondeni SD, Valim C, Potchen MJ, Milner DA, Muwalo FW, with specific PfEMP1 adhesion types and high parasite biomass. Proc Natl Birbeck GL, Bradley WG, Fox LL, Glover SJ, et al. Brain swelling and death in Acad Sci U S A. 2016;113:E3270–9. children with cerebral malaria. N Engl J Med. 2015;372:1126–37. 24. Lavstsen T, Turner L, Saguti F, Magistrado P, Rask TS, Jespersen JS, Wang CW, 3. Idro R, Jenkins NE, Newton CR. Pathogenesis, clinical features, and Berger SS, Baraka V, Marquard AM, et al. Plasmodium falciparum erythrocyte neurological outcome of cerebral malaria. Lancet Neurol. 2005;4:827–40. membrane protein 1 domain cassettes 8 and 13 are associated with severe 4. Idro R, Kakooza-Mwesige A, Balyejjussa S, Mirembe G, Mugasha C, Tugumisirize J, malaria in children. Proc Natl Acad Sci U S A. 2012;109:E1791–800. Byarugaba J. Severe neurological sequelae and behaviour problems after cerebral 25. Avril M, Tripathi AK, Brazier AJ, Andisi C, Janes JH, Soma VL, Sullivan DJ Jr, malaria in Ugandan children. BMC Res Notes. 2010;3:104. Bull PC, Stins MF, Smith JD. A restricted subset of var genes mediates 5. Armah HB, Wilson NO, Sarfo BY, Powell MD, Bond VC, Anderson W, Adjei AA, adherence of Plasmodium falciparum-infected erythrocytes to brain Gyasi RK, Tettey Y, Wiredu EK, et al. Cerebrospinal fluid and serum biomarkers endothelial cells. Proc Natl Acad Sci U S A. 2012;109:E1782–90. of cerebral malaria mortality in Ghanaian children. Malar J. 2007;6:147. 26. Grau GE, Piguet PF, Vassalli P, Lambert PH. Tumor-necrosis factor and other 6. McGuire W, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D. Variation in cytokines in cerebral malaria: experimental and clinical data. Immunol Rev. the TNF-alpha promoter region associated with susceptibility to cerebral 1989;112:49–70. malaria. Nature. 1994;371:508–10. 27. Kwiatkowski DP, Luoni G. Host genetic factors in resistance and 7. Ouma C, Davenport GC, Were T, Otieno MF, Hittner JB, Vulule JM, Martinson susceptibility to malaria. Parassitologia. 2006;48:450–67. J, Ong’echa JM, Ferrell RE, Perkins DJ. Haplotypes of IL-10 promoter variants 28. May J, Lell B, Luty AJ, Meyer CG, Kremsner PG. Plasma interleukin-10: tumor are associated with susceptibility to severe malarial anemia and functional necrosis factor (TNF)-alpha ratio is associated with TNF promoter variants changes in IL-10 production. Hum Genet. 2008;124:515–24. and predicts malarial complications. J Infect Dis. 2000;182:1570–3. 8. Burt RA, Baldwin TM, Marshall VM, Foote SJ. Temporal expression of an H2-linked 29. Ho M, Schollaardt T, Snape S, Looareesuwan S, Suntharasamai P, White NJ. locus inhost responsetomouse malaria. Immunogenetics. 1999;50:278–85. Endogenous interleukin-10 modulates proinflammatory response in 9. Knight JC, Udalova I, Hill AVS, Greenwood BM, Peshu N, Marsh K, Plasmodium falciparum malaria. J Infect Dis. 1998;178:520–5. Kwiatkowski D. A polymorphism that affects OCT-1 binding to the TNF 30. Freitas do Rosario AP, Langhorne J. T cell-derived IL-10 and its impact on the promoter region is associated with severe malaria. Nat Genet. 1999;22:145. regulation of host responses during malaria. Int J Parasitol. 2012;42:549–55. 10. Fortin A, Stevenson MM, Gros P. Complex genetic control of susceptibility 31. Li C, Corraliza I, Langhorne J. A defect in interleukin-10 leads to enhanced to malaria in mice. Genes Immun. 2002;3:177–86. malarial disease in Plasmodium chabaudi chabaudi infection in mice. Infect 11. Amante FH, Haque A, Stanley AC, Rivera Fde L, Randall LM, Wilson YA, Yeo Immun. 1999;67:4435–42. G, Pieper C, Crabb BS, de Koning-Ward TF, et al. Immune-mediated 32. Freitas do Rosario AP, Lamb T, Spence P, Stephens R, Lang A, Roers A, mechanisms of parasite tissue sequestration during experimental cerebral Muller W, O'Garra A, Langhorne J. IL-27 promotes IL-10 production by malaria. J Immunol. 2010;185:3632–42. effector Th1 CD4+ T cells: a critical mechanism for protection from severe 12. Chakravorty SJ, Carret C, Nash GB, Ivens A, Szestak T, Craig AG. Altered immunopathology during malaria infection. J Immunol. 2012;188:1178–90. phenotype and gene transcription in endothelial cells, induced by 33. Couper KN, Blount DG, Wilson MS, Hafalla JC, Belkaid Y, Kamanaka M, Flavell Plasmodium falciparum-infected red blood cells: pathogenic or protective? RA, de Souza JB, Riley EM. IL-10 from CD4CD25Foxp3CD127 adaptive Int J Parasitol. 2007;37:975–87. regulatory T cells modulates parasite clearance and pathology during 13. Palomo J, Reverchon F, Piotet J, Besnard AG, Couturier-Maillard A, Maillet I, malaria infection. PLoS Pathog. 2008;4:e1000004. Tefit M, Erard F, Mazier D, Ryffel B, Quesniaux VF. Critical role of IL-33 34. Turner GD, Morrison H, Jones M, Davis TM, Looareesuwan S, Buley ID, Gatter receptor ST2 in experimental cerebral malaria development. Eur J Immunol. KC, Newbold CI, Pukritayakamee S, Nagachinta B, et al. An 2015;45:1354–65. immunohistochemical study of the pathology of fatal malaria. Evidence for 14. Li C, Sanni LA, Omer F, Riley E, Langhorne J. Pathology of Plasmodium widespread endothelial activation and a potential role for intercellular chabaudi chabaudi infection and mortality in interleukin-10-deficient mice are adhesion molecule-1 in cerebral sequestration. Am J Pathol. 1994;145:1057–69. ameliorated by anti-tumor necrosis factor alpha and exacerbated by anti- 35. MacPherson GG, Warrell MJ, White NJ, Looareesuwan S, Warrell DA. Human transforming growth factor beta antibodies. Infect Immun. 2003;71:4850–6. cerebral malaria. A quantitative ultrastructural analysis of parasitized 15. Sanni LA, Jarra W, Li C, Langhorne J. Cerebral edema and cerebral erythrocyte sequestration. Am J Pathol. 1985;119:385–401. hemorrhages in interleukin-10-deficient mice infected with Plasmodium 36. Turner G. Cerebral malaria. Brain Pathol. 1997;7:569–82. chabaudi. Infect Immun. 2004;72:3054–8. 37. Spitz S. The pathology of acute falciparum malaria. Mil Surg. 1946;99:555–72. 16. Wilson KD, Stutz SJ, Ochoa LF, Valbuena GA, Cravens PD, Dineley KT, Vargas 38. Berendt AR, Tumer GDH, Newbold CI. Cerebral malaria: the sequestration G, Stephens R. Behavioural and neurological symptoms accompanied by hypothesis. Parasitol Today. 1994;10:412–4. cellular neuroinflammation in IL-10-deficient mice infected with 39. Pongponratn E, Turner GD, Day NP, Phu NH, Simpson JA, Stepniewska K, Plasmodium chabaudi. Malar J. 2016;15:428. Mai NT, Viriyavejakul P, Looareesuwan S, Hien TT, et al. An ultrastructural 17. Mota MM, Jarra W, Hirst E, Patnaik PK, Holder AA. Plasmodium chabaudi- study of the brain in fatal Plasmodium falciparum malaria. Am J Trop Med infected erythrocytes adhere to CD36 and bind to microvascular endothelial Hyg. 2003;69:345–59. cells in an organ-specific way. Infect Immun. 2000;68:4135–44. 40. White NJ, Turner GD, Medana IM, Dondorp AM, Day NP. The murine 18. Brugat T, Cunningham D, Sodenkamp J, Coomes S, Wilson M, Spence PJ, Jarra cerebral malaria phenomenon. Trends Parasitol. 2010;26:11–5. W, Thompson J, Scudamore C, Langhorne J. Sequestration and histopathology 41. Lewallen S, Bakker H, Taylor TE, Wills BA, Courtright P, Molyneux ME. Retinal in Plasmodium chabaudi malaria are influenced by the immune response in findings predictive of outcome in cerebral malaria. Trans R Soc Trop Med an organ-specific manner. Cell Microbiol. 2014;16:687–700. Hyg. 1996;90:144–6. 19. Cunningham DA, Lin JW, Brugat T, Jarra W, Tumwine I, Kushinga G, Ramesar J, Franke-Fayard B, Langhorne J. ICAM-1 is a key receptor mediating 42. Patnaik JK, Das BS, Mishra SK, Mohanty S, Satpathy SK, Mohanty D. Vascular cytoadherence and pathology in the Plasmodium chabaudi malaria model. clogging, mononuclear cell margination, and enhanced vascular Malar J. 2017;16:185. permeability in the pathogenesis of human cerebral malaria. Am J Trop 20. Dorovini-Zis K, Schmidt K, Huynh H, Fu W, Whitten RO, Milner D, Kamiza S, Med Hyg. 1994;51:642–7. Molyneux M, Taylor TE. The neuropathology of fatal cerebral malaria in 43. Franke-Fayard B, Janse CJ, Cunha-Rodrigues M, Ramesar J, Buscher P, Que I, Malawian children. Am J Pathol. 2011;178:2146–58. Lowik C, Voshol PJ, den Boer MA, van Duinen SG, et al. Murine malaria 21. Ponsford MJ, Medana IM, Prapansilp P, Hien TT, Lee SJ, Dondorp AM, Esiri parasite sequestration: CD36 is the major receptor, but cerebral pathology is MM, Day NP, White NJ, Turner GD. Sequestration and microvascular unlinked to sequestration. Proc Natl Acad Sci U S A. 2005;102:11468–73. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 16 of 17 44. Bridges DJ, Bunn J, van Mourik JA, Grau G, Preston RJ, Molyneux M, Combes 70. Reid HA, Nkrumah FK. Fibrin-degradation products in cerebral malaria. V, O'Donnell JS, de Laat B, Craig A. Rapid activation of endothelial cells Lancet. 1972;1:218–21. enables Plasmodium falciparum adhesion to platelet-decorated von 71. Feintuch CM, Saidi A, Seydel K, Chen G, Goldman-Yassen A, Mita-Mendoza Willebrand factor strings. Blood. 2010;115:1472–4. NK, Kim RS, Frenette PS, Taylor T, Daily JP. Activated neutrophils are 45. Moxon CA, Wassmer SC, Milner DA, Jr., Chisala NV, Taylor TE, Seydel KB, associated with pediatric cerebral malaria vasculopathy in Malawian Molyneux ME, Faragher B, Esmon CT, Downey C, et al: Loss of endothelial children. MBio. 2016;7:e01300–15. protein C receptors links coagulation and inflammation to parasite 72. Hochman SE, Madaline TF, Wassmer SC, Mbale E, Choi N, Seydel KB, Whitten sequestration in cerebral malaria in African children. Blood 2013, 122:842–851. RO, Varughese J, Grau GE, Kamiza S, et al. Fatal pediatric cerebral malaria is 46. O'Regan N, Gegenbauer K, O'Sullivan JM, Maleki S, Brophy TM, Dalton N, Chion associated with intravascular monocytes and platelets that are increased A, Fallon PG, Grau GE, Budde U, et al. A novel role for von Willebrand factor in with HIV coinfection. MBio. 2015;6:e01390–15. the pathogenesis of experimental cerebral malaria. Blood. 2016;127:1192–201. 73. Sampath S, Brazier AJ, Avril M, Bernabeu M, Vigdorovich V, Mascarenhas A, 47. Sun G, Chang WL, Li J, Berney SM, Kimpel D, van der Heyde HC. Inhibition Gomes E, Sather DN, Esmon CT, Smith JD. Plasmodium falciparum adhesion of platelet adherence to brain microvasculature protects against severe domains linked to severe malaria differ in blockade of endothelial protein C Plasmodium berghei malaria. Infect Immun. 2003;71:6553–61. receptor. Cell Microbiol. 2015;17:1868–82. 48. Moxon CA, Chisala NV, Mzikamanda R, MacCormick I, Harding S, Downey C, 74. Desruisseaux MS, Gulinello M, Smith DN, Lee SC, Tsuji M, Weiss LM, Spray Molyneux M, Seydel KB, Taylor TE, Heyderman RS, Toh CH. Laboratory DC, Tanowitz HB. Cognitive dysfunction in mice infected with Plasmodium evidence of disseminated intravascular coagulation is associated with a fatal berghei strain ANKA. J Infect Dis. 2008;197:1621–7. outcome in children with cerebral malaria despite an absence of clinically 75. Lackner P, Beer R, Heussler V, Goebel G, Rudzki D, Helbok R, Tannich E, evident thrombosis or bleeding. J Thromb Haemost. 2015;13:1653–64. Schmutzhard E. Behavioural and histopathological alterations in mice with 49. Rampengan TH. Cerebral malaria in children. Comparative study between cerebral malaria. Neuropathol Appl Neurobiol. 2006;32:177–88. heparin, dexamethasone and placebo. Paediatrica Indonesiana. 1991;31: 76. Lacerda-Queiroz N, Rodrigues DH, Vilela MC, Miranda AS, Amaral DC, 59–66. Camargos ER, Carvalho LJ, Howe CL, Teixeira MM, Teixeira AL. Inflammatory 50. Munir M, Tjandra H, Rampengan TH, Mustadjab I, Wulur FH. Heparin in the changes in the central nervous system are associated with behavioral treatment of cerebral malaria. Paediatr Indones. 1980;20:47–50. impairment in Plasmodium berghei (strain ANKA)-infected mice. Exp 51. Reid HA, Sucharit P. Ancrod, heparin, and -aminocaproic acid in simian Parasitol. 2010;125:271–8. Knowlesi malaria. Lancet. 1972;2:1110–2. 77. Nie CQ, Bernard NJ, Norman MU, Amante FH, Lundie RJ, Crabb BS, Heath 52. Stone WJ, Hanchett JE, Knepshield JH. Acute renal insufficiency due to WR, Engwerda CR, Hickey MJ, Schofield L, Hansen DS. IP-10-mediated T cell falciparum malaria. Review of 42 cases. Arch Intern Med. 1972;129:620–8. homing promotes cerebral inflammation over splenic immunity to malaria 53. Grau GE, Craig AG. Cerebral malaria pathogenesis: revisiting parasite and infection. PLoS Pathog. 2009;5:e1000369. host contributions. Future Microbiol. 2012;7:291–302. 78. Campanella GS, Tager AM, El Khoury JK, Thomas SY, Abrazinski TA, Manice 54. Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE. Behavioral LA, Colvin RA, Luster AD. Chemokine receptor CXCR3 and its ligands CXCL9 and functional analysis of mouse phenotype: SHIRPA, a proposed protocol and CXCL10 are required for the development of murine cerebral malaria. for comprehensive phenotype assessment. Mamm Genome. 1997;8:711–3. Proc Natl Acad Sci U S A. 2008;105:4814–9. 55. Stephens R, Langhorne J. Effector memory Th1 CD4 T cells are maintained 79. Rudin W, Eugster HP, Bordmann G, Bonato J, Müller M, Yamage M, Ryffel B. in a mouse model of chronic malaria. PLoS Pathog. 2010;6:e1001208. Resistance to cerebral malaria in tumor necrosis factor-alpha/beta-deficient 56. Chung K, Wallace J, Kim S-Y, Kalyanasundaram S, Andalman AS, Davidson TJ, mice is associated with a reduction of intercellular adhesion molecule-1 up- Mirzabekov JJ, Zalocusky KA, Mattis J, Denisin AK, et al. Structural and regulation and T helper type 1 response. Am J Pathol. 1997;150:257–66. molecular interrogation of intact biological systems. Nature. 2013;497:332–7. 80. Ampawong S, Chaisri U, Viriyavejakul P, Nontprasert A, Grau GE, 57. Bertini R, Bianchi M, Villa P, Ghezzi P. Depression of liver drug metabolism Pongponratn E. Electron microscopic features of brain edema in rodent and increase in plasma fibrinogen by interleukin 1 and tumor necrosis cerebral malaria in relation to glial fibrillary acidic protein expression. Int J factor: a comparison with lymphotoxin and interferon. Int J Clin Exp Pathol. 2014;7:2056–67. Immunopharmacol. 1988;10:525–30. 81. Ma N, Madigan MC, Chan-Ling T, Hunt NH. Compromised blood-nerve 58. Francischetti IM, Seydel KB, Monteiro RQ. Blood coagulation, inflammation, barrier, astrogliosis, and myelin disruption in optic nerves during fatal and malaria. Microcirculation. 2008;15:81–107. murine cerebral malaria. Glia. 1997;19:135–51. 59. Blombäck B, Hessel B, Hogg D, Therkildsen L. A two-step fibrinogen—fibrin 82. Medana IM, Turner GDH. Human cerebral malaria and the blood–brain transition in blood coagulation. Nature. 1978;275:501–5. barrier. Int J Parasitol. 2006;36:555–68. 60. Raut S, McEvoy F, Gaffney PJ. Development of an ELISA for the 83. O'Sullivan JM, Preston RJ, O'Regan N, O'Donnell JS. Emerging roles for quantification of fibrin in canine tumours. Thromb Res. 1999;96:11–7. hemostatic dysfunction in malaria pathogenesis. Blood. 2016;127:2281–8. 61. Clemmensen I. Three new E-antigenic fibrinogen fractions found in a 84. Hearn J, Rayment N, Landon DN, Katz DR, de Souza JB. Immunopathology commercial plasmin preparation. Science Tools LKB Instr J. 1973;20:7–8. of cerebral malaria: morphological evidence of parasite sequestration in 62. Medana IM, Chaudhri G, Chan-Ling T, Hunt NH. Central nervous system in murine brain microvasculature. Infect Immun. 2000;68:5364–76. cerebral malaria: ‘innocent bystander’ or active participant in the induction 85. Franke-Fayard B, Fonager J, Braks A, Khan SM, Janse CJ. Sequestration and of immunopathology? Immunol Cell Biol. 2001;79:101–20. tissue accumulation of human malaria parasites: can we learn anything from rodent models of malaria? PLoS Pathog. 2010;6:e1001032. 63. Ridet JL, Privat A, Malhotra SK, Gage FH. Reactive astrocytes: cellular and molecular cues to biological function. Trends Neurosci. 1997;20:570–7. 86. Kwiatkowski D, Sambou I, Twumasi P, Greenwood BM, Hill AVS, Manogue 64. Haussinger D, Schliess F. Pathogenetic mechanisms of hepatic KR, Cerami A, Castracane J, Brewster DR. TNF concentration in fatal cerebral, encephalopathy. Gut. 2008;57:1156–65. non-fatal cerebral, and uncomplicated Plasmodium falciparum malaria. 65. Yap GS, Stevenson MM. Blood transfusion alters the course and outcome of Lancet. 1990;336:1201–4. Plasmodium chabaudi AS infection in mice. Infect Immun. 1994;62:3761–5. 87. van Hensbroek MB, Palmer A, Onyiorah E, Schneider G, Jaffar S, Dolan G, 66. Brown H,Hien TT, DayN,Mai NT,ChuongLV, Chau TT, LocPP, Phu Memming H, Frenkel J, Enwere G, Bennett S, et al. The effect of a NH, Bethell D, Farrar J, et al. Evidence of blood-brain barrier monoclonal antibody to tumor necrosis factor on survival from childhood dysfunction in human cerebral malaria. Neuropathol Appl Neurobiol. cerebral malaria. J Infect Dis. 1996;174:1091–7. 1999;25:331–40. 88. Kwiatkowski D, Molyneux ME, Stephens S, Curtis N, Klein N, Pointaire P, Smit 67. Nacer A, Movila A, Baer K, Mikolajczak SA, Kappe SH, Frevert U. M, Allan R, Brewster DR, Grau GE, et al. Anti-TNF therapy inhibits fever in Neuroimmunological blood brain barrier opening in experimental cerebral cerebral malaria. Q J Med. 1993;86:91–8. malaria. PLoS Pathog. 2012;8:e1002982. 89. Severe and complicated malaria. World Health Organization Malaria Action 68. Milner DA, Whitten RO, Kamiza S, Carr R, Liomba G, Dzamalala C, Seydel KB, Programme. Trans R Soc Trop Med Hyg. 1986;80 Suppl:3–50. Molyneux ME, Taylor TE. The systemic pathology of cerebral malaria in 90. Havlik I, Looareesuwan S, Vannaphan S, Wilairatana P, Krudsood S, Thuma African children. Front Cell Infect Microbiol. 2014;4:104. PE, Kozbor D, Watanabe N, Kaneko Y. Curdlan sulphate in human severe/ 69. Borochovitz D, Crosley AL, Metz J. Disseminated intravascular coagulation cerebral Plasmodium falciparum malaria. Trans R Soc Trop Med Hyg. 2005; with fatal haemorrhage in cerebral malaria. Br Med J. 1970;2:710. 99:333–40. Wilson et al. Journal of Neuroinflammation (2018) 15:173 Page 17 of 17 91. Jaroonvesama N. Intravascular coagulation in falciparum malaria. Lancet. 1972;1:221–3. 92. Sarkar S, Keswani T, Sengupta A, Mitra S, Bhattacharyya A. Differential modulation of glial cell mediated neuroinflammation in Plasmodium berghei ANKA infection by TGF beta and IL 6. Cytokine. 2017;99:249–59. 93. Reverchon F, Mortaud S, Sivoyon M, Maillet I, Laugeray A, Palomo J, Montecot C, Herzine A, Meme S, Meme W, et al. IL-33 receptor ST2 regulates the cognitive impairments associated with experimental cerebral malaria. PLoS Pathog. 2017;13:e1006322. 94. Capuccini B, Lin J, Talavera-Lopez C, Khan SM, Sodenkamp J, Spaccapelo R, Langhorne J. Transcriptomic profiling of microglia reveals signatures of cell activation and immune response, during experimental cerebral malaria. Sci Rep. 2016;6:39258. 95. Schluesener HJ, Kremsner PG, Meyermann R. Widespread expression of MRP8 and MRP14 in human cerebral malaria by microglial cells. Acta Neuropathol. 1998;96:575–80. 96. Deininger MH, Kremsner PG, Meyermann R, Schluesener H. Macrophages/ microglial cells in patients with cerebral malaria. Eur Cytokine Netw. 2002; 13:173–85. 97. Warrell DA, Looareesuwan S, Warrell MJ, Kasemsarn P, Intaraprasert R, Bunnag D, Harinasuta T. Dexamethasone proves deleterious in cerebral malaria. N Engl J Med. 1982;306:313–9. 98. Di Perri G, Di Perri HG, Monteiro GB, Bonora S, Hennig C, Cassatella M, Micciolo R, Vento S, Dusi S, Bassetti D, Concia E. Pentoxifylline as a supportive agent in the treatment of cerebral malaria in children. J Infect Dis. 1995;171:1317–22. 99. Hemmer CJ, Kern P, Holst FGE, Nawroth PP, Dietrich M. Neither heparin nor acetylsalicylic acid influence the clinical course in human Plasmodium falciparum malaria: a prospective randomized study. Am J Trop Med Hyg. 1991;45:608–12.

Journal

Journal of NeuroinflammationSpringer Journals

Published: Jun 4, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off