Transcription is all too often graphically represented by a straight line (of DNA) broken up by boxes representing regulatory elements lying upstream of the promoter. These regulatory elements bind their cognate activator and repressor proteins which are then brought to the promoter to activate (or repress) transcription. This simple graphical, 'black-box' representation, while useful as a schematic, carries an incorrect view of the molecules and mechanisms required to carry out transcription. This view can often carry over into the types of experiments that we do and the interpretations that we draw.
Reality is somewhat different. Each cell contains about 2 m of DNA packaged into a nucleus <10 ^m in diameter. This is brought about by assembling the DNA into chromatin. The chromatin is then packaged into a hierarchy of structures starting with the nucleosome, which is assembled into 30 nm fibres and finally into a metaphase chromosome. Any protein concerned with the regulation or execution of transcription must interact with and penetrate this repressive structure. Activation domains of transcription factors interact with components of the transcriptional complex and/or chromatin modifying complexes. This takes place in a two-stage process. First, chromatin remodelling complexes and/or histone modifying enzymes convert the chromatin from a 'closed' to an 'open' state. Second, as a result of this remodelling, access of proximal activators is increased. Bound activators can then interact with either the RNA polymerase holocomplex or the TFIID complex to stabilize formation of the pre-initiation complex. Activation or repression of transcription can occur by interference or modulation of any of these steps (see Lemon and Tjian 2000). Stated otherwise, the in vivo template for transcription is chromatin, not DNA. This perspective is underlined by the increasing number of transcription factors that are known to interact with histones via recruitment of histone modifying enzymes (see Struhl 1998) and/or chromatin remodelling activities (Flaus and Owen-Hughes 2001; Muller and Leutz 2001). If we embrace this less linear view of transcription then we must rephrase our questions and carry out different experiments if we are ever to answer the type of question: What is responsible for ensuring a hippocampal pyramidal cell expresses Ml but not M5 muscarinic cholinergic receptors? One immediate consequence of this is to examine the experimental paradigm that has, more than any other, been responsible for identifying these regulatory elements—reporter gene analysis using transient transfection.
Was this article helpful?