Homeodomains and specific DNA recognition

There is a variety of types of proteins that can specifically bind to the DNA. Some of them bind to the major groove, and others to the minor. Many bind to the standard B-form of the DNA, while others require it to be prebent or unwound, and still others induce conformational changes on the DNA. Some proteins require ligands like metal ions in order to be able to recognize the DNA. Many proteins bind as oligomers, but monomer binding is also common. The proteins themselves can also change conformation upon binding to the DNA. The DNA-binding proteins also differ in their functions, which include transcription factors, repressors, etc. The function of many structurally known DNA-binding proteins is not known or understood.

One of the many DNA-binding motifs in proteins is the homeodomain motif. Homeodomains are self-folding 60 amino acid polypetides, participating in the gene expression in eucaryots. They are coded by 180 base pair homeoboxes in the DNA and are usually part of larger homeoproteins. The homeoboxes are responsible for the homeotic mutations, or mistakes in tissue differentiation. Historically such mutations were first discovered in Drosophila melanogaster (fruit fly), where the fly would have two pairs of fully developed wings (bithorax mutation) or its antennae would be replaced by legs (Antennapedia mutation). Subsequently, homeodomains were found in fungi (yeast), plants, amphibians and mammals.

Over all of this range, homeodomains are extremely conserved among the species, e.g. the human Hox-A7 and the Antennapedia homedomains differ only in 1 residue out of 60 (for reviews see [1,2]), which implies strong evolutionary pressure. Even between different types of homeodomains, there is a remarkable similarity in structure. All homeodomains consist of three $\alpha$-helices (termed I, II, and III or recognition helix), connected by a loop between helices I and II, and a turn between helices II and III (Figure 1.1). One also distinguishes a flexible N-terminal arm, and sometimes helix IV at the C-terminus.

The specificity of recognition is mainly assured by four residues in helix III at positions 47, 50, 51 and 54, which contact the bases in the major groove of the DNA. The positions of these residues are indicated on Figure 1.1. Some data show that also positions 5 and sometimes 3 at the N-terminus contact the DNA bases in the minor groove thus adding to the specificity. In certain cases, the N-terminal arm can also serve to dimerize the homeodomains [3].

  

Figure 1.1: The homeodomain-DNA complex scheme. The recognition helix is shown in the major groove of the DNA. Position 47 is coloured yellow, 50 blue, 51 red and 54 white. The bases of the DNA contacted by the four side chains are shown in plate representation with sugars absent.