Female C57BL/6 mice at the age of 8 weeks were purchased from KOATECH (Pyeongtaek, Korea) and were bred at the Institute for Laboratory Animals of Kosin University College of Medicine. The P. berghei ANKA strain was originally obtained from the Department of Parasitology and Tropical Medicine, Kyungpook National University School of Medicine, Republic of Korea. The parasite was maintained at Kosin University College of Medicine by passage in C57BL/6 mice and by frozen stock. Serial passage of P. berghei was initiated after storage at −70°C by intra-peritoneal (i.p.) injection of 0.2 ml blood containing parasites into mice. Subsequent passages for maintaining the parasites were carried out by i.p. inoculation of normal mice with 0.2 ml blood diluted to contain 1×10⁶ parasitized red blood cells from a donor mouse.

A drop of blood was collected from each mouse by venesection of the tail and transferred onto the edge of a microscope slide. The blood was evenly spread across the slide to make a thin-film blood smear, which was allowed to dry before staining with Giemsa stain. Slides were viewed using light microscopy with oil immersion (1,000×magnification). Four fields of approximately 200 cells each were counted, and parasitemia was calculated as the percentage of infected red blood cells in total blood cells.

To evaluate the protein expression levels of HIF1α and VEGF in malaria-infected tissues, Western blot analysis was conducted as previously described [17]. Briefly, the samples were homogenized using a sterile glass homogenizer. The homogenized tissues were lysed using PRO PREP protein extraction solution (Intron Biotechnology, Seongnam, Korea). Protein extracts (30 μg) of various tissues from the control and infected (10% or 30% parasitemia) groups were electrophoresed with a Novex 4–12% Bis-Tris gel (Invitrogen, Carlsbad, California, USA) and transferred to nitrocellulose membranes (Thermo Fisher Scientific, Waltham, Massachusetts, USA). The membranes were pre-incubated for 2 hr in Tris-buffered saline (TBS) containing 5% skim milk and 0.05% Tween 20 (TBS-T). The membranes were incubated overnight at 4°C in TBS-T containing antibodies against GAPDH (1:5,000, Abcam, Cambridge, UK), HIF1α (1:1,000, Novus Biologicals, Littleton, Colorado, USA), or VEGF (1:1,000, Abcam). Each membrane was washed 5 times with TBS-T and then incubated with a secondary antibody (GAPDH and HIF1α: anti-mouse; VEGF: anti-rabbit) for 1 hr at room temperature. The membranes were again washed 5 times with TBS-T, and detected with an enhanced chemiluminescence detection kit (Amersham Pharmacia Biotech, Piscataway, New Jersey, USA). Quantification of each protein expression was calculated by band densities by image analysis software (Image J, NIH, Bethesda, Maryland, USA) and normalized using the average value for GAPDH.

Immunofluorescence was as previously described [18] and briefly described below. Paraffin-embedded tissue sections were incubated in PBS containing 2% FBS (for blocking) for 20 min and then incubated with antibody against VEGF (1: 100, Abcam) overnight at 4°C. Samples were washed 5 times in PBS solution and incubated with secondary antibody (Alexa Fluor 488 anti-rabbit, Thermo Fisher Scientific) for 1 hr. Cellular nuclei were counterstained with QnuclearTM Deep Red stain solution (Invitrogen) in PBS.

The hypoxia marker pimonidazole hydrochloride (Hypoxyprobe^TM-1) and mouse monoclonal antibodies were supplied by NPI, Inc. (Belmont, Massachusetts, USA). Slides were permeabilized by incubation in PBS containing 0.1% Triton X-100 for 5 min and treated with 10% normal serum in PBS for 1 hr for blocking of the nonspecific antibody binding. Slides were then incubated with a mixture of mouse anti-HIF1α monoclonal antibody (1:100, Novus Biologicals), and rabbit anti-pimonidazole polyclonal antibody (1:100, NPI Inc.) at 4°C overnight. After primary antibody treatment, slides were washed 3 times in PBS for 5 min and incubated with secondary antibodies (Alexa Fluor 546 anti-mouse and Alexa Fluor 488 anti-rabbit antibody; Invitrogen) for 1 hr. The slides were observed for epifluorescence with a confocal laser-scanning microscope.

CD31 staining for vessel counting was conducted as previously described [19]. Sections were incubated in PBS containing 2% FBS (for blocking) for 20 min and then incubated with antibody against CD31 (1:40, Dako, Carpinteria, California, USA) overnight at 4°C. Anti-mouse and anti-rabbit secondary antibodies were detected with an HRP kit (Dako). Slides were also stained with hematoxylin and eosin for histopathological analysis. The number of blood vessels was counted in 6 fields per 3 randomly chosen sections under microscopy (×200).

Statistical analysis was performed using SPSS (Version 25.0; IBM Corp., Armonk, New York, USA) software. One-way ANOVA and Bonferroni’s post hoc test were conducted for comparison among groups of western blot and IHC. Pearson’s correlation analysis was performed to analyze the correlation between HIF1α and VEGF. The results were considered statistically significant less than P<0.05.

Parasitemia was reported as the percentage of parasitized red blood cells among all red blood cells counted. In malaria-infected mice, parasitemia was increased over the duration of infection and peaked 18 days after infection (Fig. 1A). At 8-day post-infection, 0% mortality and 12.9% parasitemia (95% confidence interval [CI]=10.4% to 15.4%) and were recorded. At 14 days, approximately 23% mortality and 20% parasitemia (95% CI=14.5% to 25.6%) were recorded. At 24 days, 100% mortality was observed (Fig. 1B). We selected experimental time points to coincide with 0%, 10%, and 30% parasitemia for further studies.

We examined the expression of HIF1α and VEGF at 0%, 10%, and 30% parasitemia in various tissues. A Western blot analysis showed that HIF1α protein expression was significantly increased in the brain, heart, kidney, muscle, and spleen (Fig. 2). In comparing levels of parasitemia, HIF1α expression was generally higher in tissues with 30% parasitemia than in tissues with 10% parasitemia. This suggests that HIF1α expression is up-regulated by parasitemia. Similarly, VEGF protein level was meaningfully up-regulated in brain, heart, kidney, liver, and muscle tissues from an early stage of infection (Fig. 2). However, HIF1α and VEGF expressions were changed significantly in certain tissue types. The expression of HIF1α was very low in liver and the expression of VEGF was not changed in spleen of malarial infected mice, respectively.

To identify whether hypoxia occurred in malaria-infected tissues and to explore the relationship between hypoxia and HIF1α protein expression, we double-stained malaria-infected tissues with a marker for hypoxia and an anti-HIF1α antibody. As shown in Fig. 3, both hypoxia and HIF1α protein expression were significantly increased with parasitemia. HIF1α proteins generally co-localized with hypoxic regions, but there were some exceptions. VEGF protein level in malaria-infected tissues was also identified by immunofluorescence staining (Fig. 4) and was significantly increased in tissues with either 10% or 30% parasitemia. Thus, these results were consistent with the Western blot analysis.

Using immunohistochemical staining for CD31 (an endothelium marker), we compared vessel formation in various tissue types showing 0%, 10%, and 30% parasitemia. Blood vessel formation was significantly induced by infection in each tissue type, in a parasitemia-dependent manner (Figs. 5, 6).

Taken together, our data suggest that malaria infection induces tissue hypoxia and angiogenesis in a parasitemia-dependent manner by increasing the expression of HIF1α and VEGF.