Finally, we ask how the soma ensures it produces enough lactate to support the germline. We show it's a very delicate balance and the somatic cells are careful not to use pyruvate in their own mitochondria; in other words, they exclusively dedicate the glycolytic pathway to support the germline.

Excitingly, germ cells consume lactate (but don't produce it) - knocking down the enzyme that interconverts lactate and pyruvate, Ldh, in germ cells, results in lactate accumulation (contrary to the somatic cells which produce lactate)

So what do the somatic cells provide? We used fluorescent reporters for metabolites and we see that they produce lactate. Knocking down the enzyme that produces lactate from the glycolytic product pyruvate, Ldh, in somatic cells causes germ cells to die.

If we block glycolysis in somatic cells, we find that germ cells start dying. There's always some death in the germline, but almost twice as much when we knock down glycolytic genes in somatic cells. Knock down in the germ cells themselves has no effect - they don't break down sugars themselves.

We show that somatic cells of the testis take up circulating sugars and break them down through glycolysis.
But germ cells don't express the genes needed for glycolysis.

Adding translational control on top of transcriptional control means that cells can respond rapidly to changes in their environment - a need to differentiate, or a space becoming available in the niche - without being "confused" about their identity, as they only translate the "right" set of genes.

Finally, we show that this pathway controlling translation is key to self-renewal of stem cells, and can even restore stem cell self-renewal when niche signals are lost.

In other words, eIF3d phosphorylation helps translation initiation through eIF4F, and indeed, in human cells, it modulates the strength of interaction between the two complexes (as shown by Nick Roiuk @mykolaroiuk.bsky.social and Aurelio Teleman @telemanlab.bsky.social )

We focused on one of the regulators we identify in the screen: eIF3d. Previous work by Amy Lee and colleagues showed that eIF3d is regulated by phosphorylation, by a kinase called CK2. Knockdown of CK2 also leads to loss of stem cells, similar to loss of eIF3d and eIF4F.

How do niche signals control translation? We screened translation initiation factors by RNAi and found that different factors are required for self-renewal than for differentiation. This means that there is a mechanism to control how translation initiation happens.

This means that post-transcriptional regulation is important. In fact, we find that translation rates change during differentiation, and that this depends on the niche signal, Upd.

Stem cells rely on signals from their niche to self-renew. These signals are known to regulate transcription of genes required for self-renewal. But we find that transcripts encoding proteins expressed in differentiating cells are also present in stem cells, but the protein is not.