Work performed during this project can be divided in three main phases
1. Generation and validation of new Dox-inducible ESC lines:
To avoid technical issues related to poor antibody quality and variations in the levels of Dox-inducible transgene, we have engineer new ESCs lines containing a different epitope tag in each TF and a fluorescent reporter (mVenus) in the Dox-inducible locus. These new ESCs lines were characterized and validated in terms of their ability to induce HC and neuronal differentiation, as previously reported (Costa et. al. Development 2015). ESCs lines containing in their inducible locus just one TF (Atoh1, Pou4f3 or Gfi1), combination of these 2 TFs and the 3 TFs combination were generated and validated.
Main results achieved:
-Epitope tagging was successful for Atoh1 and Pou4f3 but not for Gfi1. Therefore, the new Dox-inducible ESCs lines were re-engineered not to contain any tag for Gfi1 only.
-HC differentiation is more efficient when all 3 TFs are induced together, however, up-regulation of several HC markers was detected when Atoh1 was combined with Gfi1 and when Pou4f3 was combined with Gfi1.
-Up-regulation of neuronal but not HC markers was efficiently promoted by overexpression of Atoh1 alone or when Atoh1 was combined with Pou4f3.
2. RNA-seq and ChIP-seq protocol optimization, sample preparation and highthroughput sequencing:
The first step was the optimizations of RNA collection followed by fluorescence activated cell sorting (FACS), as well as, ChIP protocol optimization for each TF (Atoh1, Pou4f3 and Gfi1) assessed by ChIP-qPCR. The second step involved the optimization of the libraries preparation protocol for ChIP-seq and RNA-seq experiments. Finally, these optimized procedures were applied in all the Dox-inducible ESC lines validated in the previous task and highthroughput sequencing was performed for the RNA-seq and ChIP-seq libraries.
Main results achieved:
-RNA-seq revealed that Gfi1 induces a significant transcriptional change in the genes regulated by Atoh1 and Pou4f3.
-ChIP-seq for Gfi1 requires a superior antibody quality to achieve a good signal/noise ratio similar to those obtained in Atoh1 and Pou4f3 ChIP-seq.
3. Bioinformatics training and analyses:
The bioinformatics analyses of the next generation sequencing experiments performed in the previous task were carried out by the beneficiary (Dr. Aida Costa). The beneficiary carried out extensive training which included Unix, Python and R programming and Bioinformatics methods for the whole workflow, ranging from data processing to functional analyses.
Main results achieved:
-During HC differentiation Atoh1 and Pou4f3 regulate distinct enhancers and promoters showing only 20% of binding overlap.
-Despite being a known transcriptional repressor in haematopoiesis, Gfi1 enables the activation of HC genes mediated by Atoh1 and Pou4f3 during HC differentiation.
