SAHARA: Suppression before Amplification of Highly Abundant mRNA to optimize RNAseq for gene expression profiling

Patrice Martin, GABI, INRA, AgroParisTech, Université Paris-Saclay, Jouy-en-Josas, France

C. Bevilacqua1, N. Crapart1,2, C. Hue-Beauvais1, B. Brandao1, S. Lemoine3, F. Coulpier3, P. Martin1
1. GABI, INRA, AgroParisTech, Université Paris-Saclay, Jouy-en-Josas, France ;
2. EXCILONE, Elancourt, France ;
3. Plateforme de Génomique, Institut de Biologie de l’ENS, IBENS, Paris, France
Corresponding author: Claudia.bevilacqua

Introduction: Amplification of RNA obtained from microdissected cells remains a critical step that may impact RNAseq profiles, mainly regarding weakly expressed genes. The quality (RIN value) and quantity of RNA play an important role in the choice of amplification kit to use. Moreover, RNAseq approaches on mammary epithelial cells (MEC) isolated by laser capture microdissection (LCM) showed that, at the peak of lactation, 2/3 of the transcripts correspond to 6 genes encoding the main milk proteins: 4 caseins, α-lactalbumin and Β-lactoglobulin (Canovas et al., 2014). The objectives of our study were: To amplify a few amount of RNA (less than 1 ng) extracted from cells isolated by LCM with a RIN value ranging between 6.5 and 7 without loss of transcripts. To eliminate transcripts coding for the 6 main milk proteins prior amplification of cDNA for a RNAseq analysis and thereby increase the depth of analysis to reach less expressed genes.

Materials and Methods: We compared 2 kits: Ovation Human FFPE RNAseq Library Systems (Nugen) based on the Inda-C technology to deplete the 6 most abundant mRNA and ribosomal RNA, and Smarter V4 kit (Clontech) which is designed to generate high-quality, full-length cDNA directly from few hundred of cells (10 pg to10 ng of total RNA), based on the SMART system and unbiased amplification of cDNA transcripts.

The first step was to design 12-15 primers for each of the 6 targeted genes to add to the Nugen kit. MECs were microdissected from the mammary tissue of 6 lactating goats using the Arcturus system (Life technologies). Total RNA extracted from each microdissected MEC sample was used to prepare 3 libraries: 1 library with Smarter V4 kit (Clontech), 1 library with depletion of the 6 targeted genes and ribosomal RNA (Nugen C) and 1 library with depletion of Ribosomal RNA (Nugen R) to validate the impact of milk protein cDNA transcripts depletion on the transcripts profile. Sequencing of libraries thus prepared was carried out on a NextSeq system (Illumina) and differential analysis performed between the 3 protocols.

Results and Discussion: Both kits (Nugen and Clontech) show good quality sequencing however unexpected differences were observed between protocols (high level of intronic reads in Nugen libraries). The abundance of transcripts encoding the 6 major milk proteins in Nugen libraries depleted for both ribosomal RNA and milk protein transcripts was lowered to 3-4% of reads aligned on exons whereas it reached 75-80% in the Smarter library. The Smarter Kit V4 provides a largernumber of reads exploitable for expression analyses while a lot of Nugen reads are intronic. In addition, depletion does not seem to impact the transcriptomic profile of samples.

Conclusions: The strategy implemented to deplete over represented transcripts proved to be very effective, but generates a significant number of intronic reads. Regarding the amount of initial material required to build a library for RNAseq, the Smarter kit from Clontech seems best suited for this type of approach, starting from microdissected material. One ng is enough for the Smarter kit whereas 25 ng are required for Nugen libraries.

References: Cánovas A, Rincón G, Bevilacqua C, Islas-Trejo A, Brenaut P, Hovey RC, Boutinaud M, Morgenthaler C, VanKlompenberg MK, Medrano JF, Martin P (2014) Comparison of five different RNA sources to examine the lactating bovine mammary gland transcriptome using RNA-Sequencing. Sci Rep. 4:5297

Acknowledgements: This study was funded by the National Agency for Research (MilkChEST, ANR-12-BSV6-0013-04) and by the incentive funds of the INRA Animal Genetics Division.

Download PDF

Leave a Reply

Meet Our Sponsors