RNA-seq Analysis of Transcriptomes in Thrombin-treated and Control Human Pulmonary Microvascular Endothelial Cells

Summary

This protocol presents a complete and detailed procedure to apply RNA-seq, a powerful next-generation DNA sequencing technology, to profile transcriptomes in human pulmonary microvascular endothelial cells with or without thrombin treatment. This protocol is generalizable to various cells or tissues affected by different reagents or disease states.

Abstract

The characterization of gene expression in cells via measurement of mRNA levels is a useful tool in determining how the transcriptional machinery of the cell is affected by external signals (e.g. drug treatment), or how cells differ between a healthy state and a diseased state. With the advent and continuous refinement of next-generation DNA sequencing technology, RNA-sequencing (RNA-seq) has become an increasingly popular method of transcriptome analysis to catalog all species of transcripts, to determine the transcriptional structure of all expressed genes and to quantify the changing expression levels of the total set of transcripts in a given cell, tissue or organism^1,2 . RNA-seq is gradually replacing DNA microarrays as a preferred method for transcriptome analysis because it has the advantages of profiling a complete transcriptome, providing a digital type datum (copy number of any transcript) and not relying on any known genomic sequence³.

Here, we present a complete and detailed protocol to apply RNA-seq to profile transcriptomes in human pulmonary microvascular endothelial cells with or without thrombin treatment. This protocol is based on our recent published study entitled "RNA-seq Reveals Novel Transcriptome of Genes and Their Isoforms in Human Pulmonary Microvascular Endothelial Cells Treated with Thrombin,"⁴ in which we successfully performed the first complete transcriptome analysis of human pulmonary microvascular endothelial cells treated with thrombin using RNA-seq. It yielded unprecedented resources for further experimentation to gain insights into molecular mechanisms underlying thrombin-mediated endothelial dysfunction in the pathogenesis of inflammatory conditions, cancer, diabetes, and coronary heart disease, and provides potential new leads for therapeutic targets to those diseases.

The descriptive text of this protocol is divided into four parts. The first part describes the treatment of human pulmonary microvascular endothelial cells with thrombin and RNA isolation, quality analysis and quantification. The second part describes library construction and sequencing. The third part describes the data analysis. The fourth part describes an RT-PCR validation assay. Representative results of several key steps are displayed. Useful tips or precautions to boost success in key steps are provided in the Discussion section. Although this protocol uses human pulmonary microvascular endothelial cells treated with thrombin, it can be generalized to profile transcriptomes in both mammalian and non-mammalian cells and in tissues treated with different stimuli or inhibitors, or to compare transcriptomes in cells or tissues between a healthy state and a disease state.

Protocol

A flowchart outlining this protocol is displayed in Figure 1.

1. Treatment of Cells with Thrombin, RNA Isolation, Quality Assessment and Quantification of RNA

1. Culture Human Lung Microvascular Endothelial Cells (HMVEC-LBl) to 90-100% confluence in 6-well plates in EGM-2 medium with 5% FBS, growth factors and antibiotics (Lonza, cat# CC-3202).
2. Change media to the starvation media (0% FBS) 30 min prior to treatment with thrombin.
3. Treat the cells with 0.05 U/ml thrombin or leave untreated as a control for 6 hr at 37 °C and 5% CO₂.
4. Isolate total RNA from the treated and control cells using the Ambion mirVana kit according to manufacturer's instructions.
5. Assess the quality of the RNA with an Experion StdSense Eukaryotic RNA chip according to the standard protocol on the Experion Automated Electrophoresis Station (www.bio-rad.com).
6. Quantify the RNA using a standard spectrophotometric method.

2. Library Construction and Sequencing

1. Use 1 μg of high quality total RNA per sample as starting material.
2. To construct the library, follow the standard procedure from Illumina (protocol # 15008136 Rev. A). In this protocol, two rounds of poly(A) containing mRNA selections are performed to remove rRNA to minimize the rRNA sequencing.
3. Assess the quality of the libraries using an Experion DNA 1K chip according to the standard protocol on the Experion Automated Electrophoresis Station (www.bio-rad.com).
4. Quantify the library using qPCR: Use a library that has previously been sequenced as a standard curve and primers specific for the ligated adapters. Use a range of dilutions of the unknown libraries (i.e. 1:100, 1:500 and 1:1,000). Run the qPCR according to the SyberGreen MM protocol and calculate the original stock concentration of each library.
5. Dilute the library stocks to 10 nM and store at -20 °C until ready to cluster a flow cell.
6. When ready to cluster a flow cell, thaw the cBot reagent plate in a water bath. cBot is an Illumina instrument used to streamline the cluster generation process.
7. Wash the cBot instrument.
8. Denature the libraries: Combine 13 μl 1x TE and 6 μl 10 μM library and, to the side of the tube, add 1 μl 1 N NaOH (provided by Illumina). Vortex, spin down, incubate at room temperature for 5 min and place the denatured libraries on ice.
9. Dilute the libraries: Dilute the denatured libraries with pre-chilled hybridization buffer (HT1, provided by Illumina) by combining 996 μl HT1 and 4 μl denatured library for a final concentration of 12 pM. Place the denatured and diluted libraries on ice.
10. Invert each row of tubes of the cBot plate, ensuring that all the reagents are thawed. Spin down the plate, remove/puncture the foil seals and load onto the cBot.
11. Aliquot 120 μl of the diluted, denatured libraries to a strip tube, labeled 1-8. Add 1.2 μl diluted, denatured PhiX control library (from Illumina) into each tube as a spike-in control. Vortex and spin down the tubes and load them on the cBot in the correct orientation (tube #1 to the right).
12. Load a flow cell and manifold onto the cBot.
13. Complete the flow check and begin the clustering run.
14. After the run is complete, check reagent delivery across all lanes. Make note of any abnormalities. Either start the sequencing run immediately or store the flow cell in the provided tube at 4 °C.
15. Thaw the sequencing-by-synthesis (SBS, Illumina) reagents.
16. Load the reagents to the appropriate spots on the reagent trays, making sure not to touch the other reagents after touching the cleavage mix.
17. Using a non-sequencing flow cell (i.e. one that was sequenced previously), prime the reagent lines twice.
18. Thoroughly clean the sequencing flow cell with 70% EtOH and Kimwipes, followed by 70% EtOH and lens paper. Inspect the flow cell for any streaks. Re-clean it if necessary.
19. Load the flow cell onto the sequencer and perform a flow check to ensure that the seal between the manifolds and the flow cell is tight.
20. Start the sequencing run.
21. Assess the quality metrics (e.g. the cluster density, clusters passing filter, Q30, intensity) as they become available during the run.
22. Monitor intensity throughout the run.
23. After 101 cycles are completed, perform turnaround chemistry to complete the second read: Thaw the paired end reagents and the second read Incorporation Buffer (ICB, a component of the SBS reagents, Illumina) and load the reagents.
24. Continue the sequencing run, assessing 2^nd read intensity, Q30 and other quality metrics as the run progresses.

3. Data Analysis

1. Use the latest version of CASAVA (Illumina, currently 1.8.2) to convert the base call files (.bcl) files to .fastq files, setting fastq-cluster-count to 0 to ensure the creation of a single fastq file for each sample. Unzip the fastq files for downstream analysis.
2. Perform paired end alignment using the latest versions of TopHat (1.4.1)⁵, which aligns RNA-seq reads to mammalian-sized genomes using the ultra high-throughput short read aligner(Bowtie,0.12.7)⁶ and SAMtools (0.1.17)⁷. SAMtools implements various utilities for post-processing alignments in the SAM format. The reference human transcriptome can be downloaded from iGenomes (www.illumina.com). In running TopHat, we used all default parameter settings including the library type option as fr-unstranded (default).
3. Using the program CuffDiff, part of the CuffLinks (1.3.0)⁸ software package, compare the thrombin-treated cells to the controls cells to screen out the differentially expressed gene transcripts in the former based on the human reference transcriptome. This comparison detects the differential expression of known transcripts. Use Microsoft Excel to visualize the result in table form. In running Cufflinks program, we used all default parameter settings. Those gene transcripts with FPKM<0.05 and p>0.05 are filtered out.
4. To detect novel isoforms, run Cufflinks without a reference transcriptome. Compare the sample transcript files to the reference genome using Cuffcompare and test the differential expression with Cuffdiff using the combined thrombin transcript files as the reference genome for one analysis and the combined control transcript files as the reference genome for a second analysis. Use Microsoft Excel to visualize the result in tabular format. Again, those gene transcripts with FPKM<0.05 and p>0.05 are filtered out. After this step, investigators may opt to upload a list of newly reported transcripts to the UCSC Genome Browser website (http://genome.ucsc.edu/) to verify their validity by a manual inspection.
5. Submit lists of differentially expressed genes to Ingenuity Pathway Analysis (IPA, www.ingenuity.com) for characterization of the genes and pathways affected by the thrombin treatment. In this step, investigators may opt to use CummeRbund (http://compbio.mit.edu/cummeRbund/), an R package that is designed to aid and simplify the task of analyzing Cufflinks RNA-seq output, to help manage, visualize and integrate all of data produced by a Cuffdiff analysis.

4. Validation of the RNA-seq Results by Quantitative Real-time-Polymerase Chain Reaction (qRT-PCR)

1. Perform total RNA isolation from control and thrombin-treated HMVEC-LBl cells, RNA quality assessment and RNA quantification described in Steps 1.4 to 1.6.
2. Generate complementary DNA from 1 μg total RNA of each sample with SuperScript III First-Strand Synthesis System RT Kits, following the manufacturer's instructions (Invitrogen, 18080-051).
3. Perform qRT-PCR analysis on a Applied Biosystems ViiA 7 Real-Time PCR System using Taqman Assay-on-Demand designed oligonucleotides for the detection of CUGBP, Elav-likefamilymember1(CELF1, Hs00198069_m1), Fanconianemia,complementationgroupD2 (FANDCD2, Hs00276992_m1), TNFreceptor-associated factor 1(TRAF1, Hs01090170_m1),and β-actin (ACTB,Hs99999903_m1). Each sample had a template equivalent to 5 ng of total RNA. Measure quantitation using the DDCt method and normalize to β-actin. Each assay was performed across at least three biological replicates.

Access restricted. Please log in or start a trial to view this content.

Results

For Step 1: The 28s:18s ratio is traditionally used as an indicator of RNA degradation. Ideally, the 28s peak should have approximately twice the area of the 18s band (a ratio of 2), however this ideal ratio is often not seen in practice. Furthermore, 28s:18s ratios obtained from spectrophotometric methods can underestimate the amount of degradation of the RNA. To more accurately quantify the degradation, and therefore the quality of the RNA sample, the Experion system calculates an RNA Quality I...

Access restricted. Please log in or start a trial to view this content.

Discussion

Key steps

RNA Handling: RNases will degrade even the most high-quality RNA, therefore care must be taken during the isolation, storage and use of RNA¹⁰. Gloves are always worn to prevent contamination by RNases found on human hands. Gloves should be changed often, particularly after touching skin, doorknobs or other common surfaces. A set of pipettes should be dedicated solely to RNA work and all tips and tubes should be RNase-free. RNA isolation and downstream applica...

Access restricted. Please log in or start a trial to view this content.

Materials

Name	Company	Catalog Number	Comments
Human Lung Microvascular Endothelial Cells	Lonza	CC-2815
Lonza, Bullet Kit	Lonza	CC-3202	Contains EGM-2, FBS, growth factors and antibiotics
Thrombin	Sigma	T4393
Ambion mirVana Kit	Life Technologies	AM 1560
RNase-Zap	Life Technologies	AM9782
Experion StdSens RNA	Bio-Rad	700-7103
TruSeq RNA Preparation Kit	Illumina	FC-122-1001
AMPureXP Beads	Beckman Coulter	A63881
Superscript Reverse Transcriptase II	Life Technologies	18064-014
Experion DNA 1K	Bio-Rad	700-7107
QuantiTect SyberGreen	Qiagen	204163
PE Cluster Generation Kit	Illumina	PE-401-3001
PhiX Control Kit	Illumina	FC110-301
200 Cycle SBS Kit	Illumina	FC-401-3001
HiScanSQ*	Illumina	SY-103-2001
cBot	Illumina	SY-301-2002
qPCR machine - Viia7	Life Technologies	Model #VIIA7 / Equipment #10631261	Or equivalent
Experion System	Bio Rad	7007001	Bioanalyzer is an alternative system
Spectrophotometer	Bio-Tek	Epoch Microplate Spectrophotometer	Or equivalent
Centrifuge - Sorvall Legend XTR	Thermo Scientific	75004521	Or equivalent
Magnetic stand	Life Technologies	AM10027
96-well thermocycler	General Lab Supplier
Table 3. List of Key Reagents and Major Equipment. *, In the video, HiSeq1000 instead of HiScanSQ was demonstrated.