Structural Characterization of Peptides Derived from the Sequence of Cytochrome b5

Open Access
Davis, Ronald B
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
August 13, 2007
Committee Members:
  • Dr Juliette Lecomte, Committee Chair
  • Alan James Benesi, Committee Member
  • John H Golbeck, Committee Member
  • Philip C. Bevilacqua, Committee Member
  • trifluoroethanol
  • N-cap
  • cytochrome
  • helical propensity
Rat microsomal cytochrome b5 is a membrane-bound hemoprotein consisting of two domains: an N-terminal heme binding region and C-terminal hydrophobic membrane anchor. Removal of the hydrophobic region results in an excised heme-binding domain of approximately 100 residues (referred to in this document as “cyt b5”), the solubility properties of which make it suitable for spectroscopic study in aqueous solutions. Cyt b5 can be further sub-divided into residues responsible for the formation of two hydrophobic cores. The first of these cores (core 1) forms the contacts between the protein and bound heme and loses most of its secondary structure in the apoprotein state. The second hydrophobic core (core 2) is responsible for stabilization of a scaffold that retains its secondary and tertiary structure in the apoprotein state. The secondary structure of the holoprotein is described sequentially by 1-H1-4-3-H2-H3-5-H4-H5-2-H6. This is the - topology that defines the cyt b5 fold. In this study, the role of local interactions in the behavior of apo- and holocyt b5 was investigated using peptide fragments. Each fragment was designed to preserve the ability to sample the secondary structure of the native holoprotein state, while eliminating conformational restraints arising from higher order structure in the protein. In this way, the effect of sequence on structural propensity in both the presence and absence of heme coud be examined. One of the elements of secondary structure of interest was the C-terminal 310 helix, H6. The sequence of this helix starts with HPDD and, with respect to the cis-trans isomerization of the Xxx-Pro bond, contains a potentially catalytic, positively-charged imidazolium side chain (His80). In the apoprotein, H6 fluctuates between two states. The conformational exchange is slow on the NMR chemical shift time scale, The peptidyl-prolyl bond of the model peptide for helix H6 (H6P-1) was found by NMR spectroscopy to isomerize at rates inconsistent with the fluctuations observed in this region of the intact apoprotein,. This led to the conclusion that a cis-trans isomerization event did not account for the minor conformer observed in apocyt b5. NMR data demonstrated that the peptide sampled a native-like conformation including an N-capping interaction between His80 and Asp82, implicating it as a possible nucleation site for protein folding. In the apoprotein, the region binding the heme (H2–H5) is partially disordered. Peptide models of the heme binding loop and its variants were examined by NMR and CD spectroscopy in the presence of TFE to determine the relative helical propensity of each region (H2–H5). CD and NMR in TFE solutions demonstrated that H2 and H3 contained small, but significant helical propensity and sequence-encoding for the intervening turn. The H4 region of the binding loop proved to have the least helical propensity of the four regions. CD experiments showed that replacement of Asp60 with an Arg residue acted to stabilize H4 through an electrostatic interaction (most likely with Glu56). Increased stability of the D60R holoprotein variant was consistent with the native-like stabilization provided by the putative side chain interaction. An Asn-to-Pro substitution at position 57 proved to have little effect on helical propensity of H4, indicating that the observed stability of the corresponding holoprotein variant was the result of complex interactions in the apo- and holoprotein states. To model H4 and H5, covalently conjugated mesoheme-peptide compounds were synthesized. UV-visible spectroscopy revealed a strained bis-histidine coordination state in the constructs, but no appreciable helix stabilization was detected in the attached peptides. These results implied that the proximity and register of the loop termini imposed by the loop-core 2 contacts were crucial to the cofactor induced folding of core 1 in the protein. Theses studies offered insights into the relationship among local structural propensity, loop-scaffold contacts and heme binding in core 1, as well as a potential protein folding nucleation site in core 2.