Structural and functional studies of titin's fn3 modules reveal conserved surface patterns and binding to myosin S1--a possible role in the Frank-Starling mechanism of the heart.

The A-band part of titin, a striated-muscle specific protein spanning from the Z-line to the M-line, mainly consists of a well-ordered super-repeat array of immunoglobulin-like and fibronectin-type III (fn3)-like domains. Since it has been suspected that the fn3 domains might represent titin's binding sites to myosin, we have developed structural models for all of titin's 132 fn3-like domains. A subset of eight experimentally determined fn3 structures from a range of proteins, including titin itself, was used as homology templates. After grouping the models according to their position within the super-repeat segment of the central A-band titin region, we analyzed the models with respect to side-chain conservation. This showed that conserved residues form an extensive surface pattern predominantly at one side of the domains, whereas domains outside the central C-zone super-repeat region show generally less conserved surfaces. Since the conserved surface residues may function as protein-binding sites, we experimentally studied the binding properties of expressed multi-domain fn3 fragments. This revealed that fn3 fragments specifically bind to the sub-fragment 1 of myosin. We also measured the effect of fn3 fragments on the contractile properties of single cardiac myocytes. At sub-maximal Ca(2+) concentrations, fn3 fragments significantly enhance active tension. This effect is most pronounced at short sarcomere length, and as a result the length-dependence of Ca(2+) activation is reduced. A model of how titin's fn3-like domains may influence actomyosin interaction is proposed.

1H and 15N NMR resonance assignments and secondary structure of titin type I domains.

Titin/connectin is a giant muscle protein with a highly modular architecture consisting of multiple repeats of two sequence motifs, named type I and type II. Type I modules have been suggested to be intracellular members of the fibronectin type III (Fn3) domain family. Along the titin sequence they are exclusively present in the region of the molecule located in the sarcomere A-band. This region has been shown to interact with myosin and C-protein. One of the most noticeable features of type I modules is that they are particularly rich in semiconserved prolines, since these residues account for about 8% of their sequence. We have determined the secondary structure of a representative type I domain (A71) by 15N and 1H NMR. We show that the type I domains of titin have the Fn3 fold as proposed, consisting of a three- and a four-stranded beta-sheet. When the two sheets are placed on top of each other to form the beta-sandwich characteristic of the Fn3 fold, 8 out of 10 prolines are found on the same side of the molecule and form an exposed hydrophobic patch. This suggests that the semiconserved prolines might be relevant for the function of type I modules, providing a surface for binding to other A-band proteins. The secondary structure of A71 was structurally aligned to other extracellular Fn3 modules of known 3D structure. The alignment shows that titin type I modules have closest similarity to the first Fn3 domain of Drosophila neuroglian.

The leucine zippers of the HLH-LZ proteins Max and c-Myc preferentially form heterodimers.

c-Myc and Max are members of a subfamily of the helix-loop-helix transcription-regulating proteins. Their function is mediated by switches in the dimerization partners; c-Myc does not homodimerize in vivo but competes with Mad, another member of the subfamily, to form heterodimers with Max, leading to either activation or repression of transcription. Max is also able to form homodimers. In an attempt to identify which regions of the proteins carry the information to determine specific recognition of the dimerization partner, we have investigated the dimerization properties of synthetic peptides corresponding to the leucine zipper sequence of Max and c-Myc using circular dichroism and nuclear magnetic resonance techniques. We show that the heterodimer is obtained readily by simply mixing the peptides and that at neutral pH it is more stable than the homodimer of the Max leucine zipper. We have shown in a previous paper [Muhle-Goll, C. et al. (1994) Biochemistry 33, 11296-11306] that the leucine zipper of c-Myc does not form stable homodimers under these conditions. Thus, the leucine zipper regions of these two proteins by themselves display the same behavior as the entire proteins. However, even the heterodimer is less stable than dimers of leucine zippers of the basic leucine zipper family such as GCN4 and Fos-Jun. The specificity of the interaction between different monomers can be explained by polar interactions. We investigate the structural role of the polar and charged residues in the hydrophobic interface by molecular-modeling studies.

The dimerization stability of the HLH-LZ transcription protein family is modulated by the leucine zippers: a CD and NMR study of TFEB and c-Myc.

In the HLH-LZ protein family, the helix-loop-helix DNA-binding dimerization domain is followed in the sequence by a leucine zipper motif. The precise function of this second dimerization domain is still unclear, since the HLH motif of a subset of this family has been shown to be necessary and sufficient for dimerization. However, deletion and mutagenesis studies of the leucine zipper in various HLH-LZ proteins have shown a clear influence of this motif on homo- and heterodimerization. In this paper, we present a structural characterization of synthetic peptides encompassing the leucine zipper sequences of c-Myc and TFEB, using circular dichroism, analytical ultracentrifugation, and nuclear magnetic resonance. We show that the different ability of the synthetic leucine zippers of c-Myc and TFEB to homodimerize at neutral pH reflects the different dimerization properties reported for the entire proteins. The TFEB protein is known to form homodimers. c-Myc, on the other hand, does not homodimerize in vivo, but is mostly found in heterodimeric complexes with Max, another protein of the HLH-LZ family. Accordingly, our results show that the TFEB peptide homodimerizes at neutral pH whereas the Myc peptide dimerizes to a comparable amount only at acidic pH and high ionic strength. Both synthetic peptides are far less stable than leucine zippers of the b-ZIP family. The relative stability of the two leucine zippers and the factors which stabilize the dimer formation are discussed.