Biotechnology

Promoter methylation is treated as a single mechanism with a single consequence, silencing, yet the functional output of that silencing differs sharply depending on which cellular process the affected gene serves. The unresolved question is why comparable methylation changes at comparable promoters produce a measurable functional deficit in some processes and no detectable change in others. There is no established framework that predicts which cellular functions are sensitive to promoter methylation and which are buffered against it. Related to this, the quantitative relationship between methylation density at a promoter and the magnitude of the downstream functional effect has not been characterised across processes: it is unclear whether the relationship is graded, threshold driven, or dependent on the redundancy present in the pathway the gene belongs to. Metabolic flux, proliferation control, stress response, secretory output, adhesion and motility have each been examined in isolation, but they have not been compared under a common perturbation and a common readout, so process specific sensitivity cannot be separated from differences in assay, cell type and methylation extent. What is missing is a rule connecting a defined promoter methylation state to a predictable functional consequence.

The gap concerns what happens to promoter methylation inside a developmental checkpoint rather than on either side of it. Most maps of promoter methylation are drawn from cells sampled before commitment and again after a lineage identity is fixed, which yields two stable states and no account of the transition that connects them. It remains unestablished whether the loss or gain of methylation at lineage specifying promoters is instructive, actively permitting or blocking transcription factor access at the moment of commitment, or whether it is consequential, arriving after regulatory factors have already resolved the decision and serving only to make that decision durable. The ordering of events within the checkpoint window is therefore unknown for most promoters, as is the rate at which de novo methylation and active or passive demethylation operate once the window opens. Nor is it established how much of the population level methylation change observed across a checkpoint reflects a graded shift shared by all cells, and how much reflects an all or none switch occurring at different times in different cells. Until timing, direction of causality and cell to cell heterogeneity are resolved together, promoter methylation at checkpoints cannot be interpreted as either a control point or a record.