Cerebral Malaria Model Applying Human Brain Organoids

Abstract

Neural injuries in cerebral malaria patients are a significant cause of morbidity and mortality. Nevertheless, a comprehensive research approach to study this issue is lacking, so herein we propose an in vitro system to study human cerebral malaria using cellular approaches. Our first goal was to establish a cellular system to identify the molecular alterations in human brain vasculature cells that resemble the blood-brain barrier (BBB) in cerebral malaria (CM). Through transcriptomic analysis, we characterized specific gene expression profiles in human brain microvascular endothelial cells (HBMEC) activated by the Plasmodium falciparum parasites. We also suggest potential new genes related to parasitic activation. Then, we studied its impact at brain level after Plasmodium falciparum endothelial activation to gain a deeper understanding of the physiological mechanisms underlying CM. For that, the impact of HBMEC-P. falciparum-activated secretomes was evaluated in human brain organoids. Our results support the reliability of in vitro cellular models developed to mimic CM in several aspects. These systems can be of extreme importance to investigate the factors (parasitological and host) influencing CM, contributing to a molecular understanding of pathogenesis, brain injury, and dysfunction.

Keywords: HBMEC activation; brain organoids; cerebral malaria; human iPSCs; secretome; transcriptome.

Figure 1

Data quality and data analysis description of HBMEC transcriptome. (A) Multidimensional Scaling Analysis: Using each sample’s normalized value, a 2D plot illustrates the variability of the total data. (B) Heatmap for DEG-Hierarchical Clustering: The similarity of genes and samples by expression level using Z-score for normalized value (log2 based) are clustered. Up to 3000 of 8055 genes from the significant list satisfied the |fc| ≥ 2. Each column represents a condition. Genes that were down-regulated are shown in blue, and up-regulated in yellow. (C) Significant UP and DOWN count by fold change shows up-regulated and down-regulated genes based on fold change of comparison pair. (D) Volume Plot between pooled samples for sample versus negative control (NC). Expression volume is the geometric mean of two groups’ expression levels. A volume plot is drawn to confirm the genes that show higher expression differences compared to the control according to the expression volume (X-axis: Volume, Y-axis: log2 Fold Change).

Figure 1

Data quality and data analysis description of HBMEC transcriptome. (A) Multidimensional Scaling Analysis: Using each sample’s normalized value, a 2D plot illustrates the variability of the total data. (B) Heatmap for DEG-Hierarchical Clustering: The similarity of genes and samples by expression level using Z-score for normalized value (log2 based) are clustered. Up to 3000 of 8055 genes from the significant list satisfied the |fc| ≥ 2. Each column represents a condition. Genes that were down-regulated are shown in blue, and up-regulated in yellow. (C) Significant UP and DOWN count by fold change shows up-regulated and down-regulated genes based on fold change of comparison pair. (D) Volume Plot between pooled samples for sample versus negative control (NC). Expression volume is the geometric mean of two groups’ expression levels. A volume plot is drawn to confirm the genes that show higher expression differences compared to the control according to the expression volume (X-axis: Volume, Y-axis: log2 Fold Change).

Figure 2

Cell receptor expression levels. Flow cytometry levels of receptors PECAM-1, VCAM-1, ICAM-1, EPCR, CD36, N-cadherin, JAM-A, Integrin alpha V beta 3, and ICAM-2 in HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of P. falciparum parasites. Levels of surface expression of (A) PECAM-1, (B) VCAM-1, (C) ICAM-1, (D) ICAM-2, (E) CD36, (F) EPCR, (G) JAM-A, (H) N-cadherin, and (I) Integrin αVβ3, and expression on endothelial cells when in the presence of stimulation with TNF-α and P. falciparum parasites strains for 4 h and 24 h of activation, respectively. The ratio of receptor expression is in comparison to the negative control (NC) levels. The bar graphs show mean values and corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; ***, p < 0.001, **, p = 0.01; *, p < 0.05 (One-way ANOVA with Tukey’s posttest). Compilation of valuation of the levels of PECAM-1 (J), CD36 (K), JAM-A (L), and Integrin alpha V beta 3 (αVβ3) (M) receptor expression of HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of Plasmodium falciparum parasites, and in the presence of TNF-α, between the time points 4 h and 24 h. The bar graphs show mean values and the corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; **, p = 0.01; *, p < 0.05 (Two-way ANOVA with Bonferroni test).

Figure 2

Cell receptor expression levels. Flow cytometry levels of receptors PECAM-1, VCAM-1, ICAM-1, EPCR, CD36, N-cadherin, JAM-A, Integrin alpha V beta 3, and ICAM-2 in HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of P. falciparum parasites. Levels of surface expression of (A) PECAM-1, (B) VCAM-1, (C) ICAM-1, (D) ICAM-2, (E) CD36, (F) EPCR, (G) JAM-A, (H) N-cadherin, and (I) Integrin αVβ3, and expression on endothelial cells when in the presence of stimulation with TNF-α and P. falciparum parasites strains for 4 h and 24 h of activation, respectively. The ratio of receptor expression is in comparison to the negative control (NC) levels. The bar graphs show mean values and corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; ***, p < 0.001, **, p = 0.01; *, p < 0.05 (One-way ANOVA with Tukey’s posttest). Compilation of valuation of the levels of PECAM-1 (J), CD36 (K), JAM-A (L), and Integrin alpha V beta 3 (αVβ3) (M) receptor expression of HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of Plasmodium falciparum parasites, and in the presence of TNF-α, between the time points 4 h and 24 h. The bar graphs show mean values and the corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; **, p = 0.01; *, p < 0.05 (Two-way ANOVA with Bonferroni test).

Figure 2

Cell receptor expression levels. Flow cytometry levels of receptors PECAM-1, VCAM-1, ICAM-1, EPCR, CD36, N-cadherin, JAM-A, Integrin alpha V beta 3, and ICAM-2 in HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of P. falciparum parasites. Levels of surface expression of (A) PECAM-1, (B) VCAM-1, (C) ICAM-1, (D) ICAM-2, (E) CD36, (F) EPCR, (G) JAM-A, (H) N-cadherin, and (I) Integrin αVβ3, and expression on endothelial cells when in the presence of stimulation with TNF-α and P. falciparum parasites strains for 4 h and 24 h of activation, respectively. The ratio of receptor expression is in comparison to the negative control (NC) levels. The bar graphs show mean values and corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; ***, p < 0.001, **, p = 0.01; *, p < 0.05 (One-way ANOVA with Tukey’s posttest). Compilation of valuation of the levels of PECAM-1 (J), CD36 (K), JAM-A (L), and Integrin alpha V beta 3 (αVβ3) (M) receptor expression of HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of Plasmodium falciparum parasites, and in the presence of TNF-α, between the time points 4 h and 24 h. The bar graphs show mean values and the corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; **, p = 0.01; *, p < 0.05 (Two-way ANOVA with Bonferroni test).

Figure 2

Cell receptor expression levels. Flow cytometry levels of receptors PECAM-1, VCAM-1, ICAM-1, EPCR, CD36, N-cadherin, JAM-A, Integrin alpha V beta 3, and ICAM-2 in HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of P. falciparum parasites. Levels of surface expression of (A) PECAM-1, (B) VCAM-1, (C) ICAM-1, (D) ICAM-2, (E) CD36, (F) EPCR, (G) JAM-A, (H) N-cadherin, and (I) Integrin αVβ3, and expression on endothelial cells when in the presence of stimulation with TNF-α and P. falciparum parasites strains for 4 h and 24 h of activation, respectively. The ratio of receptor expression is in comparison to the negative control (NC) levels. The bar graphs show mean values and corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; ***, p < 0.001, **, p = 0.01; *, p < 0.05 (One-way ANOVA with Tukey’s posttest). Compilation of valuation of the levels of PECAM-1 (J), CD36 (K), JAM-A (L), and Integrin alpha V beta 3 (αVβ3) (M) receptor expression of HBMEC, when stimulated with Dd2, HB3, and 3D7 strains of Plasmodium falciparum parasites, and in the presence of TNF-α, between the time points 4 h and 24 h. The bar graphs show mean values and the corresponding ± standard error of the mean (SEM) (n ≥ 3 independent experiments and 3 replicates per experiment). ****, p < 0.0001; **, p = 0.01; *, p < 0.05 (Two-way ANOVA with Bonferroni test).

Figure 3

Differentiation of human cerebral organoids in vitro. Representative immunostaining of human cerebral organoids with 45 days post-differentiation, in cryosections with 10–15 μm. Scale bars: 200 μm. Nuclear staining with DAPI (blue). (A) Staining for the neuronal marker MAP2 (green) enlightens a superficial preplate, and for newborn neurons, marker doublecortin (DCX) (red). (B) Staining for cortical-layer neurons by tubulin beta 3 (TUBB3) (green) marker, showing a large cortical region. The pre-plate marker Tbr1 (red) for early-born neurons did not show any presence, thus revealing that early-born neurons were scarce and the brain organoids were mature. (C) Immunostaining of human cerebral organoids derived from iPSCs expressed for a neural stem cell marker (NESTIN) (green) at a low level and some glial fibrillary acidic protein (GFAP, human astrocytic marker) (red). (D) Immunostaining with neural progenitor cells (NPC) marker SOX2 for the ventricular zone (VZ) and apoptosis marker Caspase 3 (CASP3).

Figure 4

Data quality and data analysis of cerebral organoids transcriptome. (A) Reproducibility between total samples: The similarity between samples obtained through Pearson’s coefficient of the sample’s normalized value, for range: −1 ≤ r ≤ 1. (B) Heatmap for DEG-Hierarchical Clustering: Clusters the similarity of genes and samples by expression level using Z-score for normalized value (log2 based). A total of 2045 genes satisfy |F| ≥ 2 and p-value (p < 0.05) from the significant list. Each column represents a condition. Genes down-regulated are in blue and up-regulated are in yellow. (C) Multidimensional Scaling Analysis: Using each sample’s normalized value, the similarity between samples is graphically shown in a 2D plot to show the variability of the total data. (D) Significant UP and DOWN count by fold change (|F| ≥ 2) and p-value (p < 0.05) shows the number of up and down-regulated genes based on fold change of comparison pair.

Figure 4

Data quality and data analysis of cerebral organoids transcriptome. (A) Reproducibility between total samples: The similarity between samples obtained through Pearson’s coefficient of the sample’s normalized value, for range: −1 ≤ r ≤ 1. (B) Heatmap for DEG-Hierarchical Clustering: Clusters the similarity of genes and samples by expression level using Z-score for normalized value (log2 based). A total of 2045 genes satisfy |F| ≥ 2 and p-value (p < 0.05) from the significant list. Each column represents a condition. Genes down-regulated are in blue and up-regulated are in yellow. (C) Multidimensional Scaling Analysis: Using each sample’s normalized value, the similarity between samples is graphically shown in a 2D plot to show the variability of the total data. (D) Significant UP and DOWN count by fold change (|F| ≥ 2) and p-value (p < 0.05) shows the number of up and down-regulated genes based on fold change of comparison pair.

Figure 5

Gene Ontology (GO) from cerebral organoids transcriptomic. Dot plot shows significant Gene Ontology (GO). GO terms in each comparison were chosen if their adjusted p-value (adj. p-value) was less than 0.05. The top 10 GO terms are shown. The x-axis and y-axis were Gene Ratio (=intersection size/query size) and GO terms, respectively. Dot size was intersection size. Single-cell RNA sequencing (scRNA-seq) of cerebral organoids was performed after stimulation with secretome from P. falciparum-HBMEC. The transcriptomes analyzed were under several conditions, such as with TFN-α stimulation at time point from 4 h, which works as a positive control (BO_TNF), three different wild-type strains, 3D7, Dd2, HB3 of P. falciparum parasites induction, from 4 h (BO_3D7_4H, BO_Dd2_4H, BO_HB3_4H) and 24 h (BO_3D7_24H, BO_Dd2_24H, BO_HB3_24H), respectively. Cerebral organoids stimulated with the secretome from HBMEC without any stimulation named modified media (BO_MM). (A) Biological Processes (B) Cellular Component (C) Molecular function.

Figure 5

Gene Ontology (GO) from cerebral organoids transcriptomic. Dot plot shows significant Gene Ontology (GO). GO terms in each comparison were chosen if their adjusted p-value (adj. p-value) was less than 0.05. The top 10 GO terms are shown. The x-axis and y-axis were Gene Ratio (=intersection size/query size) and GO terms, respectively. Dot size was intersection size. Single-cell RNA sequencing (scRNA-seq) of cerebral organoids was performed after stimulation with secretome from P. falciparum-HBMEC. The transcriptomes analyzed were under several conditions, such as with TFN-α stimulation at time point from 4 h, which works as a positive control (BO_TNF), three different wild-type strains, 3D7, Dd2, HB3 of P. falciparum parasites induction, from 4 h (BO_3D7_4H, BO_Dd2_4H, BO_HB3_4H) and 24 h (BO_3D7_24H, BO_Dd2_24H, BO_HB3_24H), respectively. Cerebral organoids stimulated with the secretome from HBMEC without any stimulation named modified media (BO_MM). (A) Biological Processes (B) Cellular Component (C) Molecular function.