Methods for Studying the Interactions of Translation Factors with the Ribosome - Biophysical Approaches to Translational Control of Gene Expression - page 93

subunit (Lomakin et al. 2003 ) as that of IF3 on the bacterial 30S subunit (Dallas and
Noller 2001 ). Remarkably, Pestova and co-authors used directed hydroxyl radical
cleavage (Lomakin et al. 2006 ) to find that bacterial IF3 and human eIF1 can bind
to each other's sites on mammalian and bacterial ribosomes, respectively. eIF1
bound not only to the principal IF3 binding site on the platform, but also to the bind-
ing site on the solvent face of the bacterial 30S ribosomal subunit observed by
crystallography (Pioletti et al. 2001 ), indicating that the IF3 binding site on the sol-
vent face of the 30S subunit is likely a bona fide secondary binding site for IF3.
The docking models obtained from directed radical probing may only be correct
to within a few Å, especially if the protein and/or the ribosome undergo conforma-
tional changes upon binding or if the actual ribosomal structure differs from the
structure used for the docking. For example, in the recently published crystal struc-
ture of the eukaryotic 40S:eIF1 complex (Rabl et al. 2011 ), the position of eIF1
differs by several Å from the model obtained earlier with directed hydroxyl radical
probing (Lomakin et al. 2003 ) and clashes with the position of the P-site tRNA
inferred from the structure of the elongating ribosome. Thus, the 40S:eIF1 structure
confirms earlier speculations that the position of the initiator tRNA in the P-site of
initiation complexes differs from that of P-site tRNA in elongating ribosomes
(Lomakin et al. 2003 ; Marintchev and Wagner 2004 ). More importantly, the eIF1
position found in the crystal structure indicates that eIF1 contacts eIF1A on the
ribosome (Rabl et al. 2011 ). This finding is consistent with the observation that eIF1
and eIF1A bind cooperatively to the 40S subunit (Maag and Lorsch 2003 ; Majumdar
et al. 2003 ), whereas the inability to detect binding between free eIF1 and eIF1A
could be explained with an affinity too low to be observed in solution. The 40S:eIF1
crystal structure raises another intriguing possibility. Since IF3-CTD and eIF1 bind
to the same sites on both the bacterial and eukaryotic small ribosomal subunits
(Lomakin et al. 2006 ), and since eIF1A and its bacterial homolog IF1 bind to the
same site in the ribosomal A-site (Carter et al. 2001 ; Yu et al. 2009 ), it is likely that
if eIF1 and eIF1A contact each other on the eukaryotic 40S subunit, IF3-CTD, and
IF1 may contact each other on the bacterial 30S subunit. Dallas and Noller used the
position of the P-site tRNA as a constraint in docking (Dallas and Noller 2001 ) and
thus IF3-CTD could not have been docked to the position occupied by eIF1, even if
the cleavage data were consistent with it. In support of this possibility, IF1 and IF3
bind cooperatively to the ribosome (Zucker and Hershey 1986 ) , similar to eIF1A
and eIF1. However, cross-linking studies have failed to detect cross-links between
IF1 and IF3 on the ribosome (Boileau et al. 1983 ) .

5.5.7 Outlook

Although directed hydroxyl radical probing is much more challenging and labor-
intensive than cross-linking and footprinting, it can provide structural models for
the ribosomal positions of proteins, comparable to low-resolution Cryo-EM recon-
structions. It is therefore the method of choice for systems where Cryo-EM and