Structure of the Wilms Tumor Suppressor Protein Zinc Finger Domain Bound to DNA

Wilms tumor protein recognizes 5-carboxylcytosine within a specific DNA sequence

“…The first of these does not contact the bases (Stoll et al 2007) and contributes little to specificity (Hamilton et al 1995;Nakagama et al 1995). Consequently, WT1 binds the same consensus sequence as Egr1/Zif268 (Stoll et al 2007) and, in some cases, antagonizes Egr1/Zif268 function (Ritchie et al 2010). For the work described here, we used a construct of WT1 containing only ZnF2 through ZnF4, which was structurally analogous to Egr1/Zif268.…”

“…This result was somewhat surprising. We anticipated that proline in the first turn (the third residue) of the helices might destabilize them (Stoll et al 2007) and abolish binding, but, evidently, they did not (Fig. 5B, C).…”

“…Because the published X-ray structure of WT1 has only 3.15 Å resolution (Stoll et al 2007), which is insufficient to discern the various C modifications, we sought a higher-resolution structure. We purified a construct comprising WT1 ZnF2-4 and determined its structure anew in complex with oligos containing various C modifications to final resolutions between 1.5 and 2.1 Å .…”

The Protein-Binding Potential of C2H2 Zinc Finger Domains

“…There are DNA binding fingers that contain C2C + H4H spacing that have been shown to not bind protein (e.g., finger 4 in the Ikaros family of protein (Honma et al, 1999;Kelley et al, 1998;Molnar and Georgopoulos, 1994;Morgan et al, 1997;Perdomo et al, 2000;Sun et al, 1996)), just as there are fingers with C4C + H3H spacing that bind protein but do not participate in sequence specific DNA binding (e.g., finger 1 from WT1 (Stoll et al, 2007)). Finally, because there are so few domains that have been functionally annotated, the actual number of domains within certain categories was low, making it difficult to determine if these results represent real correlations between function and spacing.…”

“…Originally crystallized by Pavletich and Pabo (1991), there are now almost 20 structures of Zif268, and its variants, bound to their target DNA sequences (Berman et al, 2000). Non-Canonical (Omichinski et al, 1997) GLI F2 6 Non-Canonical 2GLI (Pavletich and Pabo, 1993) F4 1, 2, 3, 6 Non-Canonical F5 -1, 2, 3, 5, 6 Non-Canonical TFIIIA F1 -1, 2 Non-Canonical 1TF3, 1TF6 (Nolte et al, 1998;Wuttke et al, 1997) F2 2, 3, 6 Non -Canonical F3 2, 3, 6, 10 Canonical F5 -1, 2, 3, 6 Non-Canonical TTK F1 2, 3, 6 Non-Canonical 2DRP (Fairall et al, 1993) F2 -1, 2, 3 Canonical YY1 F1 6 Non-Canonical 1UBD (Houbaviy et al, 1996) F2 -1, 2, 3, 6 Non -Canonical 1UBD F3 -1, 2, 3 Canonical F4 -1, 2, 3 Non-Canonical WT1 F2 -1, 6 Non-Canonical (Stoll et al, 2007) F3 -1, 2, 3 Non-Canonical F4 -1, 6 Non-Canonical Zif268 F1 -1, 2, 6 Canonical 1AAY (Pavletich and Pabo, 1991) F2 -1, 2, 3 Canonical F3 -1, 2, 6 Canonical As shown in figure 1.3, Zif268's C2H2 domains interact with DNA by inserting the N-terminal portion of the α-helix into the DNA's major groove (Pavletich and Pabo, 1991). Key residues include the arginine immediately preceding the start of the α-helix, henceforth referred to as position -1, as well as the glutamic acid and the arginine in positions 3 and 6, respectively, of the α-helix (Figure 1.3).…”