We have derived a secondary structure model for the c-myc internal ribosome
entry segment (IRES) by using information from chemical probing of the c-m
yc IRES RNA to constrain structure prediction programs. Our data suggest th
at the IRES is modular in nature, and can be divided into two structural do
mains linked by a long unstructured region. Both domains are required for f
ull IRES function. Domain 1 is a complex element that contains a GNNRA apic
al loop and an overlapping double pseudoknot motif that is topologically un
ique amongst published RNA structures. Domain 2, the smaller of the two, co
ntains an apical AUUU loop.
We have located the ribosome landing site and have shown that ribosomes ent
er in a 16 nt region downstream of the pseudoknots in a situation similar t
o that observed in several viral IRESs. To test the structure, several key
regions of the IRES were mutated and, interestingly, it appears that some o
f the structural elements that we have identified function to repress c-myc
IRES function. This has profound implications for de-regulation of c-myc e
xpression by mutations occurring in the IRES. (C) 2001 Academic Press.