7956 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. 2005, 117, 7956–7960 factors relies on short α helices that bind in the major groove. Isolated α-helical peptides could therefore be used to interfere selectively with protein–DNA interactions. However, isolated peptides do not normally adopt well-defined secondary structures, and the recognition helices in DNA binding proteins often rely on interactions with other proteins for stability with the potential for selectivity modulation.[1–7] For instance, the activity of the muscle-specific transcription factor MyoD is regulated through interaction with coactivators from the myocyte enhancer factor-2 (MEF-2) family.[8] MyoD belongs to a family of transcription factors that rely on a basic helix-loop-helix (bHLH) domain for DNA binding. MyoD dimerizes through the HLH domain and contacts the major groove of the DNA target sequences through its N-terminal recognition α helix. The production of MyoD in a wide variety of cell types, including fibroblasts and myoblasts, activates a cascade of genes that eventually results in cellular differentiation and the production of muscle cells.[9] The physiological activity of MyoD depends on the presence of DNA sequences that contain the symmetrical core motif CANNTG (E-box) that is found in the promoters and enhancers of many muscle specific genes, such as the creatine kinase enhancer.[10] In stark contrast to its high physiological specificity, MyoD displays only limited DNA-binding specificity in vitro.[11] The specificity of transcriptional activation that is needed to explain the physiological specificity of MyoD is therefore most likely achieved through cooperative interactions with other components of the transcriptional machinery such as MEF-2C.[8] The identification of mechanisms to control the DNA-binding properties of transcription factors such as MyoD would permit the regulation of their activity. Previously, we have shown that stabilization of the DNA-recognition helix of MyoD, through two disulfide bonds from an N-terminally fused apamin extension, led to a tenfold increase in the DNA-binding specificity.[12] However, due to the lability of disulfide bonds in cells, the application of such apamin-stabilized recognition helices is limited. Furthermore, it would be desirable to control the properties of DNA-binding proteins in response to external signals such as light. Previous work had shown significant stabilization of the αhelical conformation of peptides linked through cysteine residues in an i, i+ 7 spacing, to azobenzene-derived crosslinkers in their cis configuration (Figure 1a).[13] Met 116 and Ser 123, which are located on the water-exposed face of the recognition helix of MyoD, were therefore replaced with cysteine residues (Figure 1b). MyoD-bHLH-M116C-S123C was alkylated with 3, 3’-bis (sulfo)-4, 4’-bis (chloroacetamido)-azobenzene to generate photoMyoD (Figure 1 c).[14] The i, i+ 7 spacing avoided unfavourable steric interactions between residues in the helix and the cross-linker. Furthermore, this spacing maximized the conformational differences in the cis and trans configurations of the cross-linker of the residues that make specific contacts to the nucleobases, namely Arg 111, Thr 115, and Glu 118.[15] UV/Vis spectroscopy was used to follow the isomerization of photoMyoD. When the cross-linker was in its thermally stable trans configuration, the spectrum showed the strong maximum at 363 nm typical for amide-substituted azobenzene π–π* transitions (Figure 2).[16] However, irradiation with light at a wavelength of 360nm led to the loss of that absorption maximum. The percentage of isomerization was determined by HPLC separation of the …