Cell fate conversion by direct reprogramming makes use of transcription factors (TFs) that induce the identity of specific cell types.When ectopically expressed, these TFs are usually restricted in most cell types by inhibitory mechanisms, which act as barriers for cell fate conversion. In order to identify these barriers in C. elegans we are using the Zn-finger TF CHE-1 that induces the glutamatergic ASE neuron fate. Upon ectopic expression of CHE-1 and removal of barrier genes by RNAi, induction of the ASE neuronal fate marker can be seen in a variety of cell types as shown recently (Kolundzic et al.,2018).In a whole-genome RNAi screen we identified an unexpected barrier for the reprogramming of germ cells into neurons: the NAD+-dependent mitochondrial isocitrate dehydrogenase. Upon RNAi knockdown of alpha (IDHA-1) or gamma (IDHG-1) subunit of this complex, germ cells acquire neuron-like morphology and express neuronal genes upon CHE-1 overexpression. But how can disturbing mitochondrial function may feed back to the nucleus to alter gene expression and allow reprogramming?We found that the
idha-1 depletion-mediated reprogramming of germ cells to neurons is partially repressed in mutants of hypoxia-induced factor HIF-1. In the context of reprogramming HIF-1 has already been implicated in regulating cellular plasticity and oxidative phosphorylation versus glycolytic metabolism. Importantly changing levels of the metabolite alpha-ketoglutarate (aKG), which are regulated by IDHA-1, affects HIF-1 activity. In addition,we show that knock down of
idha-1 leads to a global loss of the repressive histone mark H3K9me3 and by performing a genetic screen,we identified members of the Jumonji histone demethylases, which are involved in this germ conversion phenotype.To dissect the molecular events priming this cellular conversion we performed tissue-specific ChIP-seq analysis of HIF-1 and Jumonji demethlyase members as well as GC-MS metabolome analysis upon depletion of
idha-1. Strikingly the
idha-1 depletion-mediated germ cell reprogramming is not cell autonomous, as RNAi against
idha-1 only in the germline is insufficient to efficiently reprogram germ cells. Conversely, AID-mediated tissue-specific degradation of CRISPR-tagged IDHA-1 in the soma enhances germ cell conversion. We hypothesize that metabolites such as succinate might mediate such cell non-autonomous signaling and expect that our ongoing work may reveal the exact nature of the inter-tissue crosstalk during germ cell reprogramming to neurons.