are intertwined in a compact body from which two -ribbons extend and interact with the DNA (Figure
10.6b). Unlike those of the met repressor, the -ribbons of IHF interact with the DNA in the minor groove.
Conserved proline residues from the tips of the -ribbon intercalate between base steps to support the
DNA trajectory. The protein binds a 35 base pair target, and although there is little apparent consensus
between the sequences of the binding sites, IHF prefers to intercalate at pyrimidine–purine steps. IHF binds
preferentially to these sites 10^3 –10^4 times more strongly than to random DNA sites. Pyrimidine–purine
steps have the greatest flexibility of the dinucleotide steps (Section 10.4.5), and it seems that IHF may rec-
ognize these sites indirectly through their ability to become conformationally distorted. The IHF protein inter-
acts extensively with the phosphate backbone and the minor groove of the DNA, where only three amino
acid side chains form hydrogen bonds with the DNA bases. However, these hydrogen-bonding interactions
contacts can be made with an arbitrary sequence. Thus, IHF recognises DNA sites through sequence-
dependent structural deformation but not through base-specific contacts.
The TBP binds to the TATA element that is 27–30 bases upstream of the start site of many eukaryotic
and archaea genes. This protein also uses a -sheet as the means of recognition of DNA, like the met
repressor and IHF, except that the sheet is fitted into a greatly widened minor groove, rather than the major
groove (Figure 10.9, Section 10.4.6). While IHF lies on the concave side of the bent DNA, TBP lies on the
convex side of its deformed target. In binding its target site, TBP unwinds and sharply bends the TATA elem-
ent. The conserved -sheet of TBP forms a saddle-shaped surface that matches the curvature of the exposed
bases in the deformed minor groove. A similar distortion is seen in the crystal structure of the HMG protein
bound to a TATA-like element, although an -helix is used in that complex.^16
The TBP/DNA interaction is highly conserved, as can be seen in the similar structures of the TBP/DNA
complexes from plants, animals and archaea. The TBP protein is used as a scaffold for the recruitment of
other general transcription factors, such as TFIIB (Figure 10.9b, Section 10.4.6). It is curious that such
extreme distortion of the DNA by TBP and its associated proteins occurs and is so highly maintained in
evolution. The distortion appears to be required to orient the DNA so that it enters the active site of the
RNA polymerase, and it might facilitate unwinding of the DNA at the transcription start site (Section 10.7.2).
10.3.7 Loops and Others Elements
A polypeptide does not necessarily need a defined secondary structure to interact with DNA. For example,
loops from certain DNA-binding proteins are used in the recognition of defined DNA sequences. Often
Protein–Nucleic Acid Interactions 395
Figure 10.6 Representative b-motifs and their docking into the grooves of DNA. (a) The E. coli met repressor (PDB:
1TGH). Only a single dimer complex is shown in complex with a ‘met box’, but the repressor binds as
a co-operative oligomer of dimers to adjacent met boxes (PDB: 1CMA). (b) The integration host factor
(IHF) from bacteriophage. Here, two b-ribbons engage the minor groove