Impact of disrupted cyclic stretch in intracranial aneurysms: Insights from endothelial cell transcriptomic dataset

Intracranial aneurysm (IA) rupture is a common cause of hemorrhagic stroke. The treatment of unruptured IAs is a challenging decision that requires delicate risk stratification. The rate of poor clinical outcomes after surgical intervention (aneurysm clipping) or endovascular coiling remains elevated (6.7% and 4.8%, respectively), and they do not provide an absolute guarantee to prevent IA growth and rupture. Currently, there is no pharmaceutical treatment to cure or stabilize IAs. Improving the current or developing new treatments for IA disease would require a better understanding of the cellular and molecular mechanisms occurring in the different stages of the disease. Hemodynamic forces play a critical role in IA disease. While the role of wall shear stress in IAs is well-established, the influence of cyclic circumferential stretch (CCS) still needs clarification. IAs are generally characterized by a lack of CCS. In this investigation, we sought to understand the effect of aneurysmal CCS on endothelial cell (EC) function and its potential significance in IA disease, hypothesizing that CCS can influence IA wall remodelling. RNA-seq data were generated from human umbilical vein ECs (HUVECs) exposed to physiological (6%) or aneurysmal CCS (static). We performed differential gene expression and pathway enrichment analysis. Additionally, we highlighted cell junction gene expression between static and 6% CCS to contribute to the debate about how cell junctions affect endothelium stability and integrity. Researchers in the vascular biology field may benefit from this transcriptomic profile to understand the effect of mechanical stretch on EC biology and its potential significance in vascular disease development.

Intracranial aneurysm (IA) rupture is a common cause of hemorrhagic stroke.The treatment of unruptured IAs is a challenging decision that requires delicate risk stratification.The rate of poor clinical outcomes after surgical intervention (aneurysm clipping) or endovascular coiling remains elevated (6.7% and 4.8%, respectively), and they do not provide an absolute guarantee to prevent IA growth and rupture.Currently, there is no pharmaceutical treatment to cure or stabilize IAs.Improving the current or developing new treatments for IA disease would require a better understanding of the cellular and molecular mechanisms occurring in the different stages of the disease.Hemodynamic forces play a critical role in IA disease.While the role of wall shear stress in IAs is well-established, the influence of cyclic circumferential stretch (CCS) still needs clarification.IAs are generally characterized by a lack of CCS.In this investigation, we sought to understand the effect of aneurysmal CCS on endothelial cell (EC) function and its potential significance in IA disease, hypothesizing that CCS can influence IA wall remodelling.RNA-seq data were generated from human umbilical vein ECs (HUVECs) exposed to physiological (6%) or aneurysmal CCS (static).We performed differential gene expression and pathway enrichment analysis.Additionally, we highlighted cell junction gene expression between static and 6% CCS to contribute to the debate about how cell junctions affect endothelium stability and integrity.Researchers in the vascular biology field may benefit from this transcriptomic profile to understand the effect of mechanical stretch on EC biology and its potential significance in vascular disease development. ©

Value of the Data
• Our data provides insights into the potential mechanisms involved in the pathophysiology of intracranial aneurysms.Understanding the genetic and pathway alterations in response to aneurysmal cyclic circumferential stretch contributes to the knowledge base of intracranial aneurysm disease.• We provide an initial framework for delving into specific genes or molecular pathways associated with intracranial aneurysm disease.Other researchers can build up on this work for further research and experimentation.• Endothelial cells play a crucial role in vascular health, and understanding how mechanical stretch influences their behaviour can have broader implications for vascular biology and disease.
• The focus on cell junction gene expression in the context of physiological and aneurysmal cyclic stretch contributes to ongoing debates about how hemodynamic forces impact cell junctions and influence endothelium stability and integrity.

Background
The pathogenesis of intracranial aneurysm (IA) is highly complex [1] .Several studies emphasized the role of wall shear stress (WSS) in endothelial cell (EC) dysfunction and IA initiation and growth [2][3][4][5][6] .Recently, we showed that disturbed endothelial junction organization and subsequent increased permeability in primary cilia-deficient cells may explain the severity of IA disease observed in polycystic kidney disease patients [7] .This work highlighted the importance of endothelium integrity in IA disease.Besides WSS exposure, ECs are also influenced by cyclic circumferential stretch (CCS), the level of which varies along the vascular tree.The elastic human aorta diameter undergoes about 10% increase under physiological conditions, whereas muscular peripheral arteries undergo only approximatively 5% cyclic stretch [8] .Physiological CCS participates in vascular maintenance by regulating processes like cell proliferation, morphology [9][10][11] and extracellular matrix (ECM) formation and orientation [12 , 13] .The levels of CCS in saccular IAs are unknown but generally considered low or even absent.Earlier in vitro studies reported that in comparison to no-stretch control condition, the combination of CCS and oscillatory low WSS markedly increased the expression of the gap junction connexin43 in bEnd.3 cells [14] and HUVECs [15] , suggesting a relationship between CCS and junction genes.Unravelling the role of cyclic stretch in IA disease requires further research.

RNA sequencing, data quality assessment and differential gene expression
In the present study we compared by next generation RNA-sequencing the transcriptome profile of human umbilical vein ECs (HUVECs) exposed to 48 h of physiological (6%) or aneurysmal (0%) CCS ( Fig. 1 ) using a FlexCell device.Six biological replicates were used to produce high confidence data.As a quality check of stretch assay, we tested whether we could observe in our experiments a well-known response of ECs to CSS, i.e. phosphorylation of endothelial nitric oxide synthase (eNOS) on Serine 1177 (Ser1177) [16] .By Western blot, we confirmed a higher level of eNOS phosphorylation on Serine 1177 in HUVECs exposed to physiological CCS (6%) in comparison to cells exposed to aneurysmal CCS (0%), thus validating our methodology ( Fig. 2A -2B ).Next, we performed RNA sequencing and Fastq, and Raw count data have been submitted to GEO database under accession number GSE222834.The sequencing quality of all samples was assessed with FastQC ( Fig. 3A ).Table 1 summarizes the number of reads mapping and assigned status.All samples were of high quality.Variation between samples and experimental groups were measured using a multi-dimensional scaling plot ( Fig. 3B ).We observed some variation between samples under aneurysmal (0%) CCS that lessens considerably when exposed to physiological (6%) CCS.

Junction genes expression
The RNA-seq analysis revealed that the expression of GJA4 and GJA5 genes coding for the gap junction proteins connexin37 (Cx37) and Cx40, respectively, was down-regulated by aneurysmal CCS (LogFC = −2.477and −2.032, Supplementary Table 1 ( https://data.mendeley.com/datasets/wkxnrfbzzc/1 )).This down-regulation was confirmed by quantitative PCR (qPCR) in independent experiments ( Fig. 4A -3B).Cx43 mRNA expression levels ( GJA1 , Fig. 4C ) as well as Cx43 protein expression ( Fig. 4D -4E ) were similar under physiological and aneurysmal CCS.However, Western blot analysis showed that the electrophoretic mobility of Cx43 was consequently different between 6% and 0% CSS.Indeed, while multiple bands were observed between 41 kDa and 45 kDa under physiological CCS (6%), the signal coalesced into 1 band at 41 kDa under aneurysmal CCS RNA-seq raw data and mapping metrics.The reads that are mapped only once to the genome were considered for the read allocation to genomic features.This step was done using featureCounts which counts spliced reads only once and removes reads overlapping two exons from different genes (ambiguous reads).Reads without overlap with Unique Gene Model items are not counted (noFeatures reads).

Counts
Mapping (0%) ( Fig. 4D ).It is thus likely that physiological phosphorylation of Cx43 is disrupted by absence of CSS.Connexins are crucial proteins for endothelial homeostasis, vascular function, EC cycle regulation and inflammation.Further investigations may help to better define the role of connexins and other junction proteins in IA disease, and to understand the significance of absence of CCS for IA disease progression.Along these lines, and based on the list of junction genes from the HUGO Gene Nomenclature Committee (HGNC), we generated a heatmap highlighting junction gene expression in presence or not of cyclic stretch ( Fig. 5 ).

Cyclic stretch exposure
HUVECs were seeded at 30 0 0 0 0 cells/well on pre-coated Collagen I BioFlex culture plates® (Flexcell International) and grown until confluence.Then, the cells were submitted to cyclic stretch of 6% or not (0%) for 48 h using the Flexcell strain unit FX-50 0 0T (Flexcell International) according to the manufacturers' instructions ( Fig. 1A ).Thus, the cells were stretched over a 25 mm loading station or maintained under static conditions (0%) by placing a Flexstop beneath the well of the BioFlex culture plate.The mechanical stretch was applied following arterial pulse waveform with a frequency of 1 Hz.After 48 h, we collected the cells that received a uniform strain, i.e. those in the middle of the well.The area of uniform strain was determined using the following formula: Diameter = (Diameter of loading station)/(1 + (Max% elongation/100)).

RNA extraction, library preparation, sequencing, read mapping to the reference genome and gene coverage reporting
RNA was isolated from HUVECs after exposure to cyclic stretch of 6% or not (0%).Total RNA was extracted using the NucleoSpin RNA II kit (Macherey-Nagel) according to the manufacturers' instructions.cDNA libraries were constructed by the genomic platform of the University of Geneva ( https://www.ige3.unige.ch/ ) using the Illumina TruSeq RNA sample preparation kit according to the manufacturers' protocol.Libraries were sequenced using single-end (50 nt-long) on Illumina HiSeq20 0 0 ( Fig. 1B ).FastQ reads were mapped to the ENSEMBL reference genome (GRCh38.96)using STAR version 2.4.0j[17] with standard settings, except that any reads mapping to more than one location of the genome (ambiguous reads) were discarded ( m = 1).

RNA-seq data analysis
A unique gene model was used to quantify reads per gene.Briefly, the model considers all annotated exons of all annotated protein coding isoforms of a gene to create a unique gene where the genomic region of all exons are considered coming from the same RNA molecule and merged together.All reads overlapping the exons of each unique gene model were reported using featureCounts version 1.4.6-p1[18] .Gene expression levels were reported as raw counts and in parallel normalized in Read per kilobase of exon per million reads mapped (RPKM) in order to filter out genes with low expression value (1 RPKM) before calling for differentially expressed genes.Library size normalizations and differential gene expression calculations have been performed using the package edgeR [19] designed for the R software [20] .Only genes having a significant fold change ≥2 and the Benjamini-Hochberg corrected p -value < 0.05 were considered for the differentially expressed genes analysis ( Fig. 1B ).

Gene set enrichment analysis (GSEA)
All annotated pathways for Homo sapiens, Mus musculus, Rattus norvegicus, Danio rerio, Sus scrofa and Saccharomyces cerevisiae available on WikiPathways database ( http://www.wikipathways.org/index.php/WikiPathways ) were used to generate gene sets, as well as the Kyoto Encyclopedia of Genes and Genomes metabolic pathways (KEGG http://www.genome.jp/kegg/ ) relative to GRCh38.96.Genes were ranked by their calculated fold changes (decreasing ranking).A gene set analysis using the GSEA package Version 2.2 [21 , 22] from the Broad Institute (MIT, Cambridge, MA) was used to analyse the pattern of differential gene expression between the two groups.Gene set permutations were performed 10 0 0 times for each analysis.The normalized enrichment score (NES) was calculated for each gene set.GSEA results with a nominal false discovery rate (FDR) < 0.05 and absolute normalized enrichment score (abs(NES)) > 1 were considered significant.

Statistical analysis
qPCR and Western blot analyses were done with GraphPad Prism 9.4.1 software.Results are shown in mean ±SEM.Comparisons of means have been performed using a ratio paired t -test.Data were considered statistically significant at p < 0.05.

Limitations
We specifically examined the influence cyclic circumferential stretch on endothelial cell function.While our investigation contributes greatly to our understanding of the role of CCS, it may not fully replicate the EC response to the complex physiological environment present in vivo within intracranial aneurysms.Another limitation of our study is the use of HUVECs, which may not exhibit identical responses to cyclic circumferential stretch compared to endothelial cells in cerebral arteries.

Ethics Statement
We confirm that the authors have read and followed the ethical requirements for publication in Data in Brief and that the current work does not involve human subjects, animal experiments, or any data collected from social media platforms.

Fig. 1 .
Fig. 1.Experimental design and workflow.(A) Human umbilical vein endothelial cells (HUVECs) from 6 different donors were used.Cells were seeded on BioFlex culture plates (left) and submitted to cyclic stretch of 6% or not (0%) for 48 h using the Flexcell strain unit FX-50 0 0T (middle).Protein lysates were collected for Western blot validation experiments.Total RNA was collected for sequencing and qPCR validation.(B) RNA, libraries and sequencing reads quality control have been performed prior to differential expression of genes (DEGs) and pathway analyses.

Fig. 2 .
Fig. 2. Total and phosphorylated eNOS (Serine 1177) expression.(A) Phosphorylated eNOS (Serine 1177) and total eNOS expression in HUVECs exposed to physiological (6%) or aneurysmal (0%) CCS was assessed by Western blotting.Beta-actin was used as a loading control.(B) Graph shows the levels of phosphorylated eNOS normalized to total eNOS.Differences between groups were analysed using a ratio paired t -test.* p < 0.05.

Fig. 3 .
Fig. 3. Quality assessment and differential expression analysis.(A) Base quality along the length of the reads in replicate 1 submitted to 0% stretch (left) and 6% stretch (right) as a representative sample.(B) Multidimensional scaling plot representing similarities of gene expression between HUVECs exposed to aneurysmal (0%, red dots) or physiological CCS (6%, blue dots).N = 6 for each condition.(C) Volcano plot displaying differential expressed genes between HUVECs exposed to aneurysmal or physiological CCS.The y-axis corresponds to the mean expression value of -log 10 ( p -value), and the x -axis displays the log 2 (fold change) value.The red and blue dots represent respectively the up-and downregulated genes (fold change ≥2; p < 0.05).(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5 .
Fig. 5. Heatmap showing the expression of Tight, Adherens and Gap Junction genes.The list of junction genes has been retrieved from the HUGO Gene Nomenclature Committee resource.Shown are genes with more than 1 Read per kilobase of exon per million reads mapped.Expression levels are represented as the ratio to the maximal value for each gene.
Human umbilical vein endothelial cells were subjected to 48 h of physiological or aneurysmal cyclic stretch using the Flexcell strain unit FX-50 0 0T.The success of the stretch was assessed by checking eNOS phosphorylation on serine 1177 through Western Blot analysis.RNA was isolated, and sequencing was performed using Illumina HiSeq 40 0 0. Validation experiments were conducted using Western Blot and quantitative RT-PCR analysis.Statistical analyses were performed using GraphPad Prism and R software.Data source location • Institution : Department of Pathology and Immunology, Faculty of Medicine, University of Geneva • City : Geneva • Country : Switzerland • Geographical coordinates : 46.1935 °N, 6.1512 °E Data accessibility Repository name: GEO database, Mendeley Data identification number: GSE222834, doi:10.17632/wkxnrfbzzc.2Direct URL to data: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE222834 https://data.mendeley.com/datasets/wkxnrfbzzc/1For peer review only Token for accessing the data: 1.The following secure token has been created to allow review of record GSE222834 while it remains in private status: odmbkiwqrdqntkl.2. https://data.mendeley.com/datasets/wkxnrfbzzc/1