Solution Structure, Backbone Dynamics And Function Of The ScPDCD5 Protein

Human PDCD5 protein is a novel programmed cell death (PCD)-promoting molecule, although the mechanism by which PDCD5 participates in the regulation of PCD is not very clear. Recently, it has been reported that human PDCD5 interacts with Tip60 and functions as a cooperator in acetyltransferase activity and is involved in the Tip60-p53 apoptotic pathway induced by DNA damage. The NMR structures of the N-terminal helix and the helical core have been separately determined by two groups, however, the global structural features of such an important protein family has not been reported yet. Although PDCD5 is an evolutionary conserved gene, the function of Ymr074cp, a single S. cerevisiae homologue of hPDCD5, is still unknown. Up to now, in the Saccharomyces genome database, this protein has been described as an uncharacterized protein of unknown function.The paper consists of three chapters. In the chapter one, we reviewed the theory and applications of paramagnetic relaxation enhancement (PRE) for the characterization of transient low-population states of biological macromolecules and their complexes. We also summarized some paramagnetic probes attached to proteins and nucleic acids, and several paramagnetic cosolutes for probing molecular surfaces, usually employed in the studies of the structure and dynamics of macromolecules. Additionally, we introduced several pulse sequences for the measurement of PRE 1ΗΝ?Γ2 rates. Finally, two examples of the applications of PRE were given.In the chapter two, the heteronuclear NMR methods were used to determine the solution structure of the N-terminal 116-residue fragment (N116), a predicted dsDNA-binding domain, and that of the core residues (36-116,Δ81Δ) of Ymr074cp protein. The solution structure ofΔ81Δis stabilized by hydrophobic interactions and adopts a well-defined fold of extended triple-helix bundle flanked by two disordered tails. The helical core has a"variable loop"of six residues. The NMR relaxation H experiments were employed to investigate the backbone dynamics ofΔ81Δ. By a reduced spectral density mapping and a rotational diffusion analysis of the backbone 15N relaxation parameters, we found that theΔ81Δprotein has a slightly anisotropic rotational diffusion tensor with a well-defined correlation time of about 8.2 ns for its overall tumbling in solution. The model-free dynamic parameters indicate that the core helices constitute the"relatively rigid body"of theΔ81Δprotein but the two flanking helices are permitted to undergo slight internal motions. In contrast with the core helices, the"variable loop"and the two tails are much more flexible.The 15N relaxation parameters, a reduced spectral density mapping and a Lipari-Szabo mapping based on the relaxation rates were also used to interpret the backbone dynamics of the N116 protein. The spectral density functions delineate the spectrum of frequencies available to the secondary structural elements of the N116 protein. The power of the motion given by the intensity of the spectral density function is gathered toward low frequencies in the core helices of the protein, while it is more spread toward high frequencies in the unstructured parts, especially in the highly flexible linker. It has been indicated that the N-terminal helix might undergo a faster overall tumbling motion than the helical core. Taken together, there are three typical states in the N116 protein: (1) the"relatively rigid body"represented by the core helices; (2) the highly disordered regionΝ1 16( 18?40); (3) the third state represented byΝ1 16( 4?15;α1), a much more flexible helix than the core helices.We failed to characterize any long-range (|i ? j|≥5) intramolecular NOE connecting the N-terminal helix and the helical core. By the nitroxide spin label, attached to the mutant cysteine residue at position 7 or 11, however, significant transient interactions were probed between the N-terminal helical portion and the core moiety plus several residues in the C-terminal tail. It has been indicated that the N-terminal helical structure has a unique electrostatic potential character and such transient interactions may take place via electrostatic interactions between the N-terminal helical portion and the core moiety. Because of the r?6 averaging effects, caused by the flexibility of the N-terminal helix and that of the nitroxide side chain, the distance restraint dR is interpreted as an r?6-weighted, time and ensemble-averaged distance between the spin label and an amide proton. The spirit of a semiquantitative restraint is used in the structural interpretation of the PBE data, but with relatively loose (±4 ?) bounds. Clearly, calculated structures will combine structural features that may not necessarily be present simultaneously in any one conformation and will inevitably be more compact than the true ensemble. The fact that a flexibly folded protein is most likely to be a heterogeneous ensemble makes more sense to interpret the calculated structures as an ensemble rather than as individual or average structures. A comparison of the solution structures of PDCD5-related proteins indicates that the structure of the triple-helix bundle is significantly conserved during evolution, especially in the evolution from yeast to man, while the N-terminal part transits from a high disorder to a relative order.In the third chapter, we are the first to demonstrate that YMR074c overexpression promotes H2O2-induced apoptosis in yeast, not only in a metacaspase Yca1p-dependent manner but also in an Yca1p-independent manner and that deletion of the N-terminal helical portion greatly attenuates the apoptosis-promoting activity of this protein. Based on the data on the structure and function of S. cerevisiae Ymr074cp, we conclude that Ymr074cp indeed functions as a bona fide apoptosis-promoting molecule in yeast and propose the name ScPDCD5 (gene ScPDCD5).