Vascular Transcriptomics: Investigating Endothelial Activation and Vascular Dysfunction Using Blood Outgrowth Endothelial Cells, Organ-Chips, and RNA Sequencing

Abstract

Vascular organ-chip or vessel-chip technology has significantly impacted our ability to model microphysiological vasculature. These biomimetic platforms have garnered significant interest from scientists and pharmaceutical companies as drug screening models. However, these models still lack the inclusion of patient-specific vasculature in the form of patient-derived endothelial cells. Blood outgrowth endothelial cells are patient blood-derived endothelial progenitors that have gained interest from the vascular biology community as an autologous endothelial cell alternative and have also been incorporated with the vessel-chip model. Next-generation sequencing techniques like RNA sequencing can further unlock the potential of personalized vessel-chips in discerning patient-specific hallmarks of endothelial dysfunction. Here we present a detailed protocol for (1) isolating blood outgrowth endothelial cells from patient blood samples, (2) culturing them in microfluidic vessel-chips, (3) isolating and preparing RNA from individual vessel-chips for sequencing, and (4) performing differential gene expression and bioinformatics analyses of vascular dysfunction and endothelial activation pathways. This method focuses specifically on identification of pathways and genes involved in vascular homeostasis and pathology, but can easily be adapted for the requirements of other systems. © 2022 Wiley Periodicals LLC. Basic Protocol 1: Isolation of blood outgrowth endothelial cells from patient blood Basic Protocol 2: Culture of blood outgrowth endothelial cells in microfluidic vessel-chips Basic Protocol 3: Isolation of RNA from autologous vessel-chips Basic Protocol 4: Differential gene expression and bioinformatics analyses of endothelial activation pathways.

Keywords: BOEC; bioinformatics; blood outgrowth endothelial cells; endothelial activation; organ-chips; vascular transcriptomics.

Figure 1:. Timeline of BOEC isolation, autologous vessel-chip formation, and RNA isolation.

(A) Autologous BOEC isolation takes around 4 hours till plating. (B) Outgrowth colonies appear within 2–3 weeks of plating. (C) Vessel-chip formation with patient-derived BOECs takes 24 hours to form a confluent lumen. (D) RNA isolation from confluent vessel-chips requires ~45 minutes, followed by RNA sequencing and bioinformatics which can take variable time depending on sequencing timeline of the sequencing core and availability of resources.

Figure 2:. Density gradient centrifugation of diluted blood samples.

(A) Layering blood over the density gradient centrifugation at a ~20° angle. (B) Separation of whole blood components after centrifugation. The buffy layer is present as a thin white layer between the aqueous plasma and density gradient medium.

Figure 3:. Different orientations while seeding the microfluidic device.

(A) Seeded microfluidic devices incubated while upright. (B) Second round of BOEC incubation while upside-down.

Figure 4:. Connections for setting up perfusion in microfluidic devices.

(A) Syringe with the spin lock to barb connector. (B) Tubing connected to the male end of the spin lock to barb connector attached to the syringe. (C) Curved dispensing tip connected to the female end of the female luer lock to barb connector. (D) Assembled syringe-tubing-connector setup. (E) Media reservoir made of a cut 5 mL attached to the curved dispensing tip. (F) Assembled pump – chip – reservoir setup.

Figure 5:. Updating the setup file.

The users might need to change the versions in the “Mathur.RNASeq.AlignmentScript.Setup.sh” file according to their systems. Latest versions can be found at https://repo.anaconda.com/miniconda.

Figure 6:

The users need to update the file paths to the location where the analysis will take place in the “Mathur.RNASeq.AlignmentScript.Main.sh” file. This location will be used to make intermediary folders where output files will be saved.

Figure 7:

The users need to update the file paths to the location where the raw, sequenced FASTQ files are stored in the “Mathur.RNASeq.AlignmentScript.Main.sh” file. The script will move these files from this location to the “FASTQ” folder created after the analysis begins.

Figure 8:

Users will need to update the file paths to the location where the analysis will take place in the “Mathur.RNASeq.BioinformaticsScript.Main.R” file. This will be the same as the path added in Fig. 6.

Figure 9:

Users will need to define tables for treatment-wise analysis of differential gene expression in the “Mathur.RNASeq.BioinformaticsScript.Main.R” file.

Figure 10:. Updating the KEGG pathways to be investigated.

Users will need to update the KEGG IDs in the format shown in the “Mathur.RNASeq.BioinformaticsScript.Main.R” script to investigate specific pathways.

Figure 11:. Expected results of the bioinformatics pipeline.

(A) An example heatmap of genes belonging to specific KEGG endothelial activation and vascular dysfunction pathways for three endothelial cell types as predicted by the pipeline. (B) Volcano plot of differentially expressed genes between patient #2 and #1 BOECs. (C) Volcano plot of differentially expressed genes between patient #3 and #1 BOECs.