Variants of WNT7A and GPR124 are associated with hemorrhagic transformation following intravenous thrombolysis in ischemic stroke

Abstract Aims The canonical Wnt signaling pathway plays an essential role in blood‐brain barrier integrity and intracerebral hemorrhage in preclinical stroke models. Here, we sought to explore the association between canonical Wnt signaling and hemorrhagic transformation (HT) following intravenous thrombolysis (IVT) in acute ischemic stroke (AIS) patients as well as to determine the underlying cellular mechanisms. Methods 355 consecutive AIS patients receiving IVT were included. Blood samples were collected on admission, and HT was detected at 24 hours after IVT. 117 single‐nucleotide polymorphisms (SNPs) of 28 Wnt signaling genes and exon sequences of 4 core cerebrovascular Wnt signaling components (GPR124, RECK, FZD4, and CTNNB1) were determined using a customized sequencing chip. The impact of identified genetic variants was further studied in HEK 293T cells using cellular and biochemical assays. Results During the study period, 80 patients experienced HT with 27 parenchymal hematoma (PH). Compared to the non‐PH patients, WNT7A SNPs (rs2163910, P = .001, OR 2.727; rs1124480, P = .002, OR 2.404) and GPR124 SNPs (rs61738775, P = .012, OR 4.883; rs146016051, P < .001, OR 7.607; rs75336000, P = .044, OR 2.503) were selectively enriched in the PH patients. Interestingly, a missense variant of GPR124 (rs75336000, c.3587G>A) identified in the PH patients resulted in a single amino acid alteration (p.Cys1196Tyr) in the intracellular domain of GPR124. This variant substantially reduced the activity of WNT7B‐induced canonical Wnt signaling by decreasing the ability of GPR124 to recruit cytoplasmic DVL1 to the cellular membrane. Conclusion Variants of WNT7A and GPR124 are associated with increased risk of PH in patients with AIS after intravenous thrombolysis, likely through regulating the activity of canonical Wnt signaling.


Variant Discovery
Germline short variant discovery proceeds from analysis-ready BAM files and produces variant calls. GATK(v3.3) HaplotypeCaller was used to call variants per sample in targeted and flanking regions for each individual in order to produce a file in GVCF format. We then perform joint genotyping, to combine the multisample GVCF. Next, we perform GenotypeGVCFs to get multisample genotype for all sites, and finally, hard-filter was applied to produce the final multisample callset with the desired balance of precision and sensitivity.

Annotation
SnpEff was used to separate single-nucleotide variations (SNVs) into different functional categories according to their genic location and their expected effect on encoded gene products, based on information from the RefSeq database. All variants were further annotated by the control population of the 1000 Genomes Project (2014 Oct release, http://www.1000genomes.org), ExAC (http://exac.broadinstitute.org), EVS (http://evs.gs.washington.edu/EVS), Disease databases of ClinVar (http://www.ncbi.nlm.nih.gov/clinvar), OMIM (http://www.omim.org). In addition, we categorized the SNVs into known or novel according to whether they were present in dbSNP (version 150).

Data Analysis
We used GATK CombineGVCFs to combine target sequencing dataset with ethnically matched and unrelated subjects in HT cases (HI and PH) and non-HT controls. Then, the qualified datasets were employed for statistical analysis. PLINK1.9 was performed to check the dataset: principal components analysis (--PCA) extracts the top 10 principal components of the variance-standardized relationship matrix for Population stratification and Hardy-Weinberg equilibrium exact deviation was calculated(--hardy). After filtered unqualified variants (--biallelic-only --geno 0.2 --hwe 0.0001), singlevariant association analysis for SNVs and inDel were performed to compute the the odds ratios(ORs) and P values by case-control association analysis with fisher model using PLINK 1.9.

Western blotting
Western blotting was performed using standard method. Briefly, samples (or eluted from magnetic beads) were run on polyacrylamide gels, transferred to PVDF membranes (EMD Millipore, Hayward, CA), blocked with 5% BSA (Sigma-Aldrich), and incubated with the indicated primary antibodies (1:1000) overnight at 4℃. After washed with TBST, the membranes were incubated with secondary antibodies (1:10000) at room temperature for 2 hours. The signals were detected using a chemiluminescence (ECL) kit, scanned using GelView 6000M system and analyzed by densitometric evaluation using the Image J.
Immunofluorescence HEK 293T cells were seeded into PDL-coated 8-chamber glass slides and incubated for 24 hrs. Cells were transfected with 5 ng of various expression plasmids for 48 hrs and fixed with 4% formaldehyde in PBS for 20 min. Fixed cells were washed 3 times with PBS, incubated with 0.2% Triton X-100 for 10 min, blocked with 5% normal goat serum in PBS for 1 hr and incubated with anti-3xFLAG or anti-His antibody in 1% normal goat serum overnight at 4℃. Cells were washed as above and incubated with Alexa Fluor 488-conjugated goat anti-mouse or Alexa Fluor 594-conjugated goat antirabbit antibody (Jackson ImmunoResearch, 1:500) in 1% normal goat serum for 1 hr. Cells were washed as before and covered with anti-fade medium with DAPI, followed by imaging using a fluorescence microscope (CKX53, Olympus, Germany).

Immunoprecipitation
For co-immunoprecipitation assays, HEK 293T were collected 48 hrs after transfection by using RIPA buffer. The supernatant was incubated with anti-3xFLAG antibody (Sigma) overnight at 4℃. Protein G Magnetic beads (Cell Signaling Technology) were incubated with the supernatant for 2 hrs at 4℃, followed by washing five times with the lysis buffer and boiled in 2x Laemmli Sample buffer.

Quantitative PCR (qPCR)
Total RNA was extracted using Direct-zol™RNA MiniPrep (Zymo research), and reverse transcribed using Hiscript II Q RT SuperMix for qPCR (Vazyme). The qPCR was performed by using ChamQ™ Universal SYBR qPCR Master Mix(Vazyme) for amplifying certain PCR products. Expressions of the target genes were analyzed by relative expression ratio between the genes and GAPDH. All reactions were performed in triplicate. The primers used are shown as follows: Human AXIN2 Forward: CAACACCAGGCGGAACGAA; Human AXIN2 Reverse: GCCCAATAAGGAGTGTAAGGACT; Human GAPDH Forward: TGTGGGCATCAATGGATTTGG; Human GAPDH Reverse: ACACCATGTATTCCGGGTCAAT.

TOP-Flash assay
The Firefly/Renilla dual luciferase reporter assay (Promega) was used to determine the transcriptional activity of Wnt/β-catenin signaling pathway. HEK 293T cells stably expressing the reporter plasmids (firefly luciferase and renilla luciferase) were seeded in a 96-well plate and transient transfection was performed with Fugene HD (Promega) for various GPR124 and DVL1/2/3 expressing plasmids. After 48 hrs, cells were collected and luciferase activity was measured using the Promega Dual-Luciferase Reporter Assay System. The b-catenin responsive firefly luciferase activity of each sample was normalized to Renilla luciferase activity and presented as ratio to empty plasmid controls.

Supplemental Tables
Supplemental Table 1