Antibodies

Armenian hamster anti- mouse IL-1β was purchased (Leinco, I-437) and used for in vivo studies along with an Armenian hamster IgG isotype control (Leinco, I-140), at 300 μg/mouse. Dilutions were made in PBS.

HRPII purification

The coding sequence for the mature form of HRPII was cloned into pET-15b (Novagen) without a tag, expressed and purified from E. coli lysate using nickel-affinity chromatography as described (27). Protein was exchanged into 20 mM Tris, 500 mM NaCl, 50 mM imidazole and loaded on a 5 ml nickel column (GE Healthcare). After washing with 60 column volumes of 20 mM Tris, 10 mM NaCl, 0.1% Triton X-114 to remove residual LPS, the column was washed with 20 column volumes of loading buffer and eluted with loading buffer with 1 M imidazole. All preparations of HRPII were tested for residual LPS using a LAL endotoxin test (Charles Rivers, R1708K); levels administered to mice contained less than 5 EU/kg. Fully active preparations of the protein were used for experiments. Activity was measured using a Factor Xa assay [12]. Protein concentration was determined by BCA assay (Fisher).

Mouse model of cerebral malaria

Four-week old female C57BL/6 mice were purchased from Taconic. Animals were housed under pathogen-free conditions. All experiments were approved by and performed in compliance with Animal Studies guidelines at Washington University and Johns Hopkins University. Mice were given retro-orbital intravenous injections (50 μg of recombinant HRPII or BSA, in 100 μl of PBS) approximately 12 hours prior to infection. The mice were infected by retro-orbital inoculation of P. berghei ANKA parasites (10⁵ parasites/100 μl) derived from stock mice (Swiss-Webster, Taconic) with parasitemias of <3%. All efforts were made to minimize animal suffering. We needed to use early death (6–9 days) as an endpoint rather than euthanization of ill-looking animals because the mice are capable of recovering from infection and it was important to distinguish those that recover from those that do not. Animals that survived the early period (30–70%) were euthanized for high parasitemia or sick appearance, using ketamine/xyazine overdose.

In vivo assessment of BBB permeability

Two doses of recombinant HRPII (200 μg in 100 ul) were given 24 hours apart by retro-orbital injection to 4-week old female C57BL/6J mice from Jackson Labs. 48 hours post initial injection, the mice were evaluated for sodium fluorescein extravasation as previously described [13]. In some mice, anti-IL-1β antibody or isotype control (300 ug each) was administered with the first HRPII infusion.

Statistical analysis

All data were analyzed using Graph Pad Prism 6.0 (Graph Pad Software, La Jolla, CA, USA). A p<0.05 was designated as significant. Pair-wise comparisons were analyzed by two-tailed t-test. Log rank test was used to compare survival between the groups in the Kaplan-Meier survival curves.

HRPII promotes increased BBB permeability in vivo

In previous work, we showed that HRPII can disrupt a human cerebral endothelial cell barrier in vitro [11]. At concentrations found in patients with cerebral malaria [14], HRPII triggered inflammasome activation, resulting in junctional protein rearrangement and barrier leakage. To determine whether the HRPII could induce a compromise in barrier integrity of the brain endothelium in vivo, uninfected mice were administered two 200 μg doses of HRPII 24 hours apart by intravenous (IV) injection, and fluorescein extravasation was measured in the brain parenchyma (Fig 1A). We observed an increase in vascular leakage of fluorescein into the cortex and cerebellum of mice infused with HRPII compared to control animals (Fig 1B and 1C). Peak serum HRPII levels at one-hour post infusion were 300–400 ng/ml. At the time of harvest, HRPII levels were 150–200 ng/ml.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. HRPII causes vascular leakage in vivo.

(A) Scheme of experimental design. Two doses of HRPII or BSA (200 μg) were injected into 4-week old female C57Bl/6 mice at 0 and 24 hours. At 48 hours, fluorescein levels in the cortex (B) and cerebellum (C) of the mice was measured. HRPII treatment was significantly different from control by two-tailed t-test, p = 0.01 (cortex) and p = 0.02 (cerebellum). Data are mean values +/-SEM for 8–16 mice per group accumulated over 3 independent experiments.

https://doi.org/10.1371/journal.pone.0177142.g001

Our in vitro BBB model suggested that HRPII-mediated permeability was inflammasome dependent. To assess this effect in vivo, we infused uninfected mice with a neutralizing antibody to IL-1β or an isotype control and then treated them with either HRPII or a control protein (Fig 2). IL-1β-specific antibody blocked HRPII-induced sodium fluorescein leakage but did not affect sodium fluorescein extravasation in animals not receiving HRPII. These data confirm the in vivo relevance of IL-1β-mediated signaling for the actions of HRPII on BBB permeability.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. HRPII-mediated vascular leakage is blocked by antibody to IL-1β.

Mice were infused as in Fig 1, with HRPII plus an isotype antibody (Iso, positive control), HRPII plus anti-IL-1β antibody (experimental condition) or anti-IL-1β antibody alone (negative control). Untreated mice (no HRPII, no antibody) served as a further control. Vascular leakage in mice infused with HRPII/isotype is statistically significantly different from mice infused with HRPII/ anti-IL-1β, p = 0.01 (cerebellum) and p = 0.01 (cortex), by two-tailed t-test; p = 0.003 (cerebellum) and p = 0.06 (cortex) by ANOVA one-way variance with significance between HRPII/isotype and HRPII/ anti-IL-1β. Data are mean values +/-SEM for 6–12 mice per group accumulated over 3 independent experiments.

https://doi.org/10.1371/journal.pone.0177142.g002

HRPII reduces host survival in an experimental cerebral malaria model

We next determined whether the compromise in vascular integrity observed with purified HRPII had clinical consequences during malaria infection in mice. We infused 6-week old female C57BL/6 mice with 50 μg of BSA or HRPII prior to infection with 2 x10⁵ P. berghei ANKA infected erythrocytes. Experimental cerebral malaria in mice infected with P. berghei ANKA has variable penetrance, with a 40 to 100% lethality rate from cerebral malaria [8,15] defined as neurological symptoms and death at or below 10% parasitemia, by day 10 post infection. Mice infused with HRPII had early lethality compared to control mice (Fig 3; mean time to death for HRPII = 10 days and for BSA = 16 days, P = 0.018). The parasitemia of mice dying from cerebral malaria-like symptoms was low, as expected, and importantly, no differences in parasitemia were observed between mice infused with HRPII or control protein.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. HRPII reduces survival time in an experimental cerebral malaria model.

(A) Survival curves of 4-week old female mice infused with 50 μg of BSA or HRPII prior to infection with P. berghei ANKA (10⁵ parasites). Shown are the means for n = 24 to 27 mice pooled from four independent experiments. Curves are significantly different, p = 0.03, by the log-rank (Mantel-Cox) test. Mean time to death for HRPII = 11.5 days and for BSA = 16 days, p = 0.018, by two tailed t-test. (B) Mice displaying cerebral malaria-like symptoms died at low parasitemia by day 10, yet parasitemias between HRPII-infused mice and controls were closely matched on each day. Representative data from one of three experiments shown in panel A, 10 mice per group.

https://doi.org/10.1371/journal.pone.0177142.g003