Blood cell fate changes without cell cycle transition.

In recent years, a multitude of groundbreaking papers have demonstrated that cell fates can be altered almost at will by the enforced expression of defined sets of transcription factors. Such cell fate changes are referred to as reprogramming (if they involve the transition from a differentiated cell into a pluripotent stem cell) or as transdifferentiation or direct lineage conversion (if they involve transitions between differentiated cell types). Much of the attention that induced cell fate changes have received focuses on the possibility to generate patient-specific cells for potential tissue replacement applications or disease modeling. Nevertheless, the controlled tempering with cellular identity also represents a powerful tool to probe general mechanisms underlying development and differentiation. In an elegant study published in a recent issue of Cell Cycle, di Tulio and Graf use a transdifferentiation system to investigate the role of the cell cycle during cellular commitment in the blood cell lineage.

The relationship between DNA replication and proliferation, on one hand, and cell cycle arrest and terminal differentiation, on the other, has long intrigued cell biologists. It is generally accepted that terminal differentiation leads to cell cycle exit, and that this is an important regulatory mechanism during organ growth and regeneration. It is less clear how many, or if any, cell divisions are required for cells to change fate or to terminally differentiate. In fact, there is evidence that this may be dependent on cellular context. Thus, while fibroblasts or B cells undergoing reprogramming into induced pluripotent stem (iPS) cells often transit through the cell cycle dozens of times before entering the pluripotent state, the conversion of fibroblasts or hepatocytes into neurons does not require cell division at all. Di Tulio and Graf studied the link between cell division and transdifferentation using a rapidly cycling pre-B cell line that expresses an inducible form of the myelomonocytic transcription factor C/EBPα. These B cells can be triggered to differentiate into macrophage-like cells at essentially 100% efficiency in a matter of a few days. This makes them a unique tool to study transdifferentation and develop frameworks and hypotheses that can then be tested in less accessible experimental systems, such as animal models or primary cell cultures. The authors find that the majority of B cells undergo exactly one cell division before terminally exiting the cell cycle and adopting macrophage morphology, marker gene expression and behavior such as phagocytotic activity. Preventing cell cycle transition significantly reduces the efficiency of transdifferentiation. However, a subset of cells adopts all macrophage characteristics tested, even in the presence of chemical inhibitors of DNA polymerase and without evidence for DNA replication. In fact, time-lapse imaging shows that cells that are not dividing transdifferentiate faster, and that the proportion of non-dividing cells increases with higher levels of C/EBPa. This demonstrates that cell division is not required to turn a B cell into a macrophage and provides further evidence that transdifferentiaton is mechanistically different from iPS cell reprogramming.

So, why can transdifferentiation succeed without cell cycle transition, while reprogramming cells to pluripotency apparently requires it? The answer to this might simply be that reprogramming involves large-scale epigenetic remodeling, while transdifferentiation does not. For example, since B cells and macrophages share a number of master blood cell regulators, C/EBPα partly operates by re-wiring a preexisting transcription factor network by recruiting the transcription factor PU.1 to new target genes. During reprogramming, key components of the pluripotency network such as Nanog or Pou5f1 have to first be reactivated, as they are not expressed in somatic cells. This reactivation entails DNA demethylation, which during iPS cell formation takes more than a week to occur and might require DNA replication. In contrast, no detectable changes in promoter DNA methylation have been observed during B lineage cell into macrophage conversions using the C/EBPa overexpression system, while changes in histone tail modifications do occur (Fig. 1 summarizes differences between transdifferentiation and reprogramming). Many exciting questions remain unanswered. Exactly which molecular remodeling events during iPS cell formation require cell division, and how does this relate to physiological reprogramming events in the early embryo? Does transdifferentiation without cell division generate fully functional, mature cell types? Undoubtedly, further studies with sophisticated in vivo and in vitro cellular conversion models will point toward the answers.

Figure 1. Scheme summarizing important differences between transdifferentiation and iPS cell reprogramming. Transdifferentiation events between somatic cells are rapid and can occur without cell division or apparent changes in promoter DNA methylation. Reprogramming somatic cells to pluripotency is a lengthy process with defined intermediate steps that requires cell division and DNA demethylation.