Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 May 16;19(7):1365-1377.
doi: 10.1016/j.celrep.2017.04.021.

Huntington's Disease iPSC-Derived Brain Microvascular Endothelial Cells Reveal WNT-Mediated Angiogenic and Blood-Brain Barrier Deficits

Affiliations
Free PMC article

Huntington's Disease iPSC-Derived Brain Microvascular Endothelial Cells Reveal WNT-Mediated Angiogenic and Blood-Brain Barrier Deficits

Ryan G Lim et al. Cell Rep. .
Free PMC article

Abstract

Brain microvascular endothelial cells (BMECs) are an essential component of the blood-brain barrier (BBB) that shields the brain against toxins and immune cells. While BBB dysfunction exists in neurological disorders, including Huntington's disease (HD), it is not known if BMECs themselves are functionally compromised to promote BBB dysfunction. Further, the underlying mechanisms of BBB dysfunction remain elusive given limitations with mouse models and post-mortem tissue to identify primary deficits. We undertook a transcriptome and functional analysis of human induced pluripotent stem cell (iPSC)-derived BMECs (iBMEC) from HD patients or unaffected controls. We demonstrate that HD iBMECs have intrinsic abnormalities in angiogenesis and barrier properties, as well as in signaling pathways governing these processes. Thus, our findings provide an iPSC-derived BBB model for a neurodegenerative disease and demonstrate autonomous neurovascular deficits that may underlie HD pathology with implications for therapeutics and drug delivery.

Keywords: BMEC; Huntington’s disease; RNA sequencing; WNT signaling; angiogenesis; blood-brain barrier; brain microvascular endothelial cell; epigenetics; induced pluripotent stem cell; neurodegeneration; transcriptome.

Figures

Figure 1
Figure 1. iBMECs from Healthy Control Patients Stain for BBB Markers, Show Functional Barrier Properties, and Provide Insight into Novel Regulators of BBB Genes
(A) Diagram of the human BBB. Paracellular transport is prevented by TJs formed by CLAUDINS (CLDN; blue), OCCLUDIN (OCLN; red), and ZONA OCCLUDENS (ZO; purple oval). The low levels of transcytosis are controlled by a small number of caveolae expressing CAVEOLIN-1 (orange circles and light blue). Lastly, the transport and efflux of molecules are regulated by solute carriers, ATP-binding cassette genes, and other ion channels. (B) Representative images for control iBMEC stained for PECAM-1 (CD31) (28Q), GLUT-1 (SLC2A1) (33Q), CLDN-5 (28Q), OCLN (33Q), and ZO-1 (33Q). The scale bars represent 100 μm. (C) Flow cytometry quantification of % CD31+ (594) and GLUT1+ (488) double positive control and HD cells. The bar graph shows greater than 90% pure populations of iBMECs and no statistical difference between each sample using a one-way ANOVA [n = 3 (28Q), 5 (33Q), 7 (66Q), 5 (71Q), and 7 (109Q), independent experiments/differentiations with a minimum of two technical replicates). The dot plot is shown for 33Q and 66Q iBMECs unstained, FMO, and fully stained cells. (D) Scatterplot of TEER values from control iBMECs (blue, 33Q and 28Q) and control iNPCs (red, 33Q) lines over 120 hr. The TEER values for the iBMECs are shown as average between two control iPSC lines (28Q and 33Q) and one iNPC control line over three individual readings taken from triplicate wells. There was no statistical difference in TEER values between two control iBMEC samples, but a significant difference was seen between two control iBMECs with the iNPCs [(n) = 14 (33Q BMEC), 16 (28Q BMEC), 3 (NPCs, adjusted p values = 1.94 × 10−4 (24 hr), 1.44 × 10−3 (48 hr), 2.74 × 10−4 (72 hr), and 9.37 × 10−3 (96 h), n.s. (120 hr) independent experiments/differentiations per sample; ANOVA with Bonferroni post hoc correction]. (E) List of uniquely expressed CLDNs found in RNA-seq data from control iBMECs. (F) A Venn diagram of shared SLC- and ABC- transporters between control iBMECs data and previously published BMEC transcriptomic data. (G) Selected results from motif analysis on all SLC- and ABC- transporter genes expressed in control iBMECs. p values represent the likelihood of finding the calculated enrichment of that motif in random sequences with similar GC content. See also Tables S1–S5. # (*p < 0.05; **p < 0.01; and ***p < 0.001). For all of the error bars (mean ± SEM).
Figure 2
Figure 2. HD iBMECs Show Increased Migration
(A) Wound-healing assay shows HD iBMECs have increased migration into wound. The images are 0 and 6 hr time points. The scale bars represent 200 μm. (B) Plot shows change in area over time [(n) = 4 (all lines) independent experiments/differentiations; adjusted p value = (66Q 4.69 × 10−4) and (71Q 9.02 × 10−3) with ANOVA and Bonferroni post hoc correction]. The lines used were 28Q, 33Q, 66Q, and 71Q. (C) Transwell migration assay showing increased migration after 24 hr of HD compared to control iBMECs treated with 0, 50, or 100 ng/mL VEGF [(n = 3 (28Q), 3 (33Q), 3 (66Q), 2 (71Q), and 3 (109Q) independent experiment/differentiations with five images counted per growth replicate; adjusted p value = (66Q 2.22 × 10−2, 3.22 × 10−2, and 7.15 × 10−3), (71Q 2.83 × 10−2, 2.31 × 10−2, and 3.21 × 10−2), and (109Q 9.06 × 10−3, 4.67 × 10−4, and 8.85 × 10−5); two-way ANOVA and Bonferroni post hoc correction]. (D) Quantification of proliferating cells in HD and control iBMECs by % Ki67 positive shows no statistical difference (n = 3 independent experiments/differentiations with five images counted per two growth replicates each; one-way ANOVA). (E and F) Quantification of relative number of dead (E) and apoptotic (F) cells in HD and control iBMECs by flow cytometry. Only 109Q HD iBMECs have a significant increase in the number of apoptotic and dead cells [(n = 10 (28Q), 8 (33Q), 11 (66Q), 9 (71Q), and 8 (109Q) independent experiment/differentiations comparing controls to each individual HD line (adjusted p value = 109Q 1.99 × 10−2 and 5.26 × 10−4); one-way ANOVA and Bonferroni post hoc correction]. # (*p < 0.05; **p < 0.01; and ***p < 0.001). For all of the error bars (mean ± SEM).
Figure 3
Figure 3. HD iBMECs Have Abnormal BBB Function
(A) Representative immunofluorescence micrographs for CLDN5 (33Q and 109Q) and OCLN (33Q and 71Q) in control and HD iBMECs (red arrows mark larger puncta). The scale bars represent 10 μm. The bar graph shows relative CLDN5 intracellular puncta per cell. There is an increase in intracellular CLDN5 protein inside HD cells, [(33Q n = 6), (28Q n = 3), (66Q n = 3), (71Q n = 6), and (109Q n = 3); p value = 7.76 × 10−3, Student’s t test]. There were at least three images at 100× per experiment that were used. (B) Western blots with bar graph quantitation of CLDN5 and OCLN levels with the LICOR system. The quantitation was normalized to β-ACTIN, and the bar graph values are shown relative to 28Q line. No difference between any lines was found (ANOVA and Dunnett’s multiple comparison test). (C) A bar graph of TEER values for control (black) and HD (gray) lines at 72 hr post subculture. The TEER is decreased in all HD iBMECs. The resistance values from each triplicate measurement/well were averaged in each experiment containing three replicate wells per condition/sample [n and adjusted p value; (n) = 14 (33Q) and 16 (28Q), (60Q n = 20 and 4 × 10−2), (66Q n = 11 and 1 × 10−3), (71Q n = 8 and 1 × 10−6), and (109Q n = 16 and 1 × 10−9); ANOVA with Bonferroni correction]. (D) Histograms and bar graphs of the puromycin survival assay. The bar graph values are relative to untreated for each cell line (n = 3 independent experiments/differentiations with triplicate wells for each condition and five images per well were counted comparing controls to each individual HD line [adjusted p value (60Q 9 × 10−3 and 1.05 × 10−3) and (109Q 1 × 10−2 and 1.08 × 10−2; ANOVA and Bonferroni post hoc correction]. (E) Flow cytometry quantification of Rhodamine 123 efflux/uptake in control and HD iBMECs. For the R123 uptake assay, controls were compared to each individual HD line and a positive control iNPC [n and adjusted p value; (n) = 8 (33Q) and 17 (28Q), (66Q n = 13 and 1.23 × 10−2), (71Q n = 5 and 4.5 × 10−4), (109Q n = 10 and 1.96 × 10−6), and (33Q-iNPCs n = 3 and 5.59 × 10−7); ANOVA and Bonferroni post hoc correction]. (F) Histograms shown for 33Q and 109Q and flow cytometry quantification of PGP protein levels showing an increase in HD iBMECs compared to controls [n and adjusted p value; (n) = 4 (33Q) and 8 (28Q), (66Q n = 6 and 4.59 × 10−5), (71Q n = 3 and 1.77 × 10−3), and (109Q n = 3 and 2.89 × 10−6); ANOVA and Bonferroni post hoc correction]. The histograms are shown for 28Q and 109Q. (G) Flow cytometry quantification of CAV1 protein levels showing an increase in HD iBMECs [n and adjusted p value; (n) = 4 (33Q) and 11 (28Q), (66Q n = 8 and 4 × 10−3), (71Q n = 11 and 5 × 10−3), and (109Q n = 9 and 3 × 10−2); one-way ANOVA and Bonferroni post hoc correction]. The histograms are shown for 28Q and 66Q. There is reduced MDR1 function in HD iBMECs (R123) and less cell survival due to decreased efflux of puromycin. (H) Histograms and bar graph of the albumin uptake assay show increased uptake of albumin-A594 in HD iBMECs and positive control 33Q iNPCs compared to control iBMECs [n and adjusted p values; (n) = 12 (33Q) and 17 (28Q), (66Q n = 13 and 3.04 × 10−3), (71Q n = 5 and 9.18 × 10−6), (109Q n = 10 and 1.01 × 10−8), and (33Q-iNPCs n = 3 and 1.59 × 10−9); by one-way ANOVA and Bonferroni post hoc]. The histogram is shown for 28Q and 109Q. # (*p < 0.05; **p < 0.01; and ***p < 0.001). For all of the error bars (mean ± SEM). See also Figure S2.
Figure 4
Figure 4. Exploratory, Pathway, and Motif Analysis of Transcriptomic Data from Control and HD iBMECs Reveals mHTT Dysregulation of BBB Genes and Related Pathways
(A) PCA of log2 normalized count data on global expression demonstrates grouping within individual control or HD samples and separation between both groups along PC#2. (B) Hierarchical clustering of DEGs in signaling pathways regulating BBB function using log2 normalized count data. The heatmap shows expression values normalized to row min (blue) and max (orange), with genes grouped by pathways indicated by the color bar. The FDR and fold change are displayed for each gene. (C) IPA upstream regulator analysis showing transcriptional regulators predicted to be activated by calculation of activation Z scores. The p value was calculated by Fisher’s exact test from expected and observed genes overlapping with our DEG list and all genes regulated by each transcription factor. (D) Scatterplots of log2 normalized count data for NOTCH3, TCF3, GLI2, and SMAD3, plotted by increasing CAG length, with best fit line and R2 values displayed. See also Tables S6 and S7.
Figure 5
Figure 5. HD iBMECs and Patient Tissue Show Enrichment for Angiogenesis, WNT, and PRC2 Signaling and Histone Methylation
(A) Network mapping of BiNGO biological process analysis showing enrichment for genes that regulate WNT signaling, angiogenesis, and vascular development. Node size = number of genes and node color = decreasing FDR blue to purple. All enrichment nodes have an adjusted p value < 0.05. Adjusted p values were calculated based on overrepresentation of categories over a background sample using a hypergeometric test and adjusted using a Benjamini-Hochberg FDR. (B) Gene network showing differentially expressed genes that are involved in angiogenesis and have a direct protein-protein interaction connection (edges). The orange denotes genes upregulated and blue downregulated in HD. (C) Combined ChEA and ENCODE analysis to determine transcription factors that regulate the DEGs. Adjusted p value by Fisher’s exact test and Benjamini-Hochberg correction log (adjusted p value). (D) Enrichment analysis of ENCODE histone modifications relevant to DEGs. Adjusted p value by Fisher’s exact test and Benjamini-Hochberg correction. (E) PRC2 genes/effectors and histone methyltransferases that are dysregulated in HD iBMECs. (F) Western blotting in human HD cortical tissue for the WNT/β-catenin transcriptional target SOX17. SOX17 is 2.5-fold higher in HD patient samples (p < 0.01, Student’s t test). (G) Immunohistochemistry for SOX17 in control or HD human cortex. The SOX17 (brown) is expressed in blood vessels (hematoxylin, blue). The scale bars represent 20 μm. See also Table S8 and Figure S1. # (*p < 0.05; **p < 0.01; and ***p < 0.001). For all of the error bars (mean ± SEM).
Figure 6
Figure 6. WNT Inhibition Restores Angiogenic Deficits in HD iBMECs and Model of NVU Impairment in HD
(A) Wound-healing assay shows HD iBMECs treated with 20 μM XAV939 have decreased migration into the wound. The plot shows change in area overtime. [(n) = 3 (all lines) independent experiments/differentiations and adjusted p value; (33Qv66Q 4.43 × 10−2), (33Qv71Q 2.27 × 10−6), (33Qv109Q 5.31 × 10−5), (66Qv66Q-XAV 5.70 × 10−3), (71Qv71Q-XAV 1.34 × 10−4), and (109Qv109Q-XAV 2.96 × 10−3); two-way ANOVA and bonferroni post hoc]. (B) Model of EC barrier dysfunction in HD. Schematic diagram of NVU in HD and selected genes (yellow denotes genes upregulated and blue downregulated in HD) that are changed in HD iBMECs and HD astrocytes. These genes may contribute to neuronal dysfunction and death. # (*p < 0.05; **p < 0.01; and ***p < 0.001). For all of the error bars (mean ± SEM).

Similar articles

See all similar articles

Cited by 51 articles

See all "Cited by" articles

References

    1. Bindea G, Mlecnik B, Hackl H, Charoentong P, Tosolini M, Kirilovsky A, Fridman WH, Pagès F, Trajanoski Z, Galon J. ClueGO: a Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks. Bioinformatics. 2009;25:1091–1093. - PMC - PubMed
    1. Chamorro-Jorganes A, Lee MY, Araldi E, Landskroner-Eiger S, Fernández-Fuertes M, Sahraei M, Quiles Del Rey M, van Solingen C, Yu J, Fernández-Hernando C, et al. VEGF-induced expression of miR-17–92 cluster in endothelial cells is mediated by ERK/ELK1 activation and regulates angiogenesis. Circ Res. 2016;118:38–47. - PMC - PubMed
    1. Cheranov SY, Karpurapu M, Wang D, Zhang B, Venema RC, Rao GN. An essential role for SRC-activated STAT-3 in 14,15-EET-induced VEGF expression and angiogenesis. Blood. 2008;111:5581–5591. - PMC - PubMed
    1. Dong X, Tsuji J, Labadorf A, Roussos P, Chen JF, Myers RH, Akbarian S, Weng Z. The role of H3K4me3 in transcriptional regulation is altered in Huntington’s disease. PLoS ONE. 2015;10:e0144398. - PMC - PubMed
    1. Drouin-Ouellet J, Sawiak SJ, Cisbani G, Lagacé M, Kuan WL, Saint-Pierre M, Dury RJ, Alata W, St-Amour I, Mason SL, et al. Cerebrovascular and blood-brain barrier impairments in Huntington’s disease: Potential implications for its pathophysiology. Ann Neurol. 2015;78:160–177. - PubMed

Publication types

MeSH terms

Feedback