Transcription of insulin is regulated by the binding of various transcription factors to the ~400 base pairs before the insulin transcription start site, called the "insulin regulatory sequence".[1] This sequence is made up of several distinct regions with different biochemical properties, each of which serve as binding sites for distinct regulatory proteins. First, multiple A/T-rich sequences, called "A elements", each of which contains a "TAAT" reocognized by homeodomain proteins. These regions are primarily bound by PDX-1, but also Cdx2 and Isl-1.[1] Second, two so-called "C elements" – C1 located 107–118 base pairs before the transcription start site; C2 311–317 base pairs before the start site. C1 is bound by RIPE3b1 via MafA. C2 (also called the "pancreatic islet cell enhancer sequence" or "PISCES") is bound by PAX6.[1] Third, an "E element" (two in rodents) with sequence GCCATCTG is 91–100 base pairs before the transcription start site and binds the helix-loop-helix transcription factors NEUROD1.[1] Lastly, several "cyclic AMP response elements" with sequence TGACGTCA that binds CREB.[1]
In humans, a "Z-element" resides 243–292 base pairs before the start site and binds a complex called ZaI, as well as PDX-1 and MafA.[1]
| Par1 | ||
| Pdx1 | ||
| negative regulatory element (NRE)[3] | glucocorticoid receptor, Oct1 | |
| Z (overlapping NRE and C2) | ISF | |
| Pax4, MafA(?) | ||
| USF1/USF2 | ||
| Pdx1 | ||
| CREB, CREM | ||
| – | ||
| CAAT enhancer binding (CEB) (partly overlapping A2 and C1) | – | |
| – | ||
| E2A, NeuroD1, HEB | ||
| Pdx1 | ||
| – |