Amide hydrogens reveal a temperature-dependent structural transition that enhances site-II Ca2+-binding affinity in a C-domain mutant of cardiac troponin C

Figures

• 

  Figure 1. Mutation D145E affects the chemical environment. (a) Superposition of 1H-15N-HSQC NMR spectra for Ca2+-bound HcTnC WT (red) and D145E (blue) reveals the most perturbed residues in the D145E protein at 25 °C. (b) Chemical-shift perturbation (CSP) plots for the N- and C-domains as a function of residue. Dashed lines show the average and one and two s.d. above the average. Residues with CSP values ≥ 2 s.d. are shown by yellow backbones on HcTnC in c, the WT NMR structure (PDB: 1AJ4) for the C-domain. Ca2+ ions are shown as green spheres and D145 as a green stick.

• 

  Figure 2. Thermostability and hydrophobic exposure of WT and D145E. Mean residue ellipticity at 222 nm was measured for HcTnC WT and D145E. Data shown are sample derivatives from 3 experiments with 2 different protein batches in apo (a) and Mg2+/Ca2+-bound (b) states. All melting curves were run in reverse after reaching 90 °C, and the structural changes promoted by temperature were reversible. For average values, see text. In (c), 5 μM bis-ANS was excited at 360 nm in the presence of 1 μM WT or D145E to compare fluorescence intensity at low (pCa 9.0) and high (pCa 4.0) Ca2+ concentrations as a measure of binding to hydrophobic surface areas. Average ratios (holo/apo ± s.e.m.) were 1.68 ± 0.11 for WT and 1.13 ± 0.02 for D145E (p = 0.007).

• 

  Figure 3. Secondary structure at 25 °C (a,b) and distribution of amide hydrogen temperature coefficients over the range 15°–55 °C (c) among WT () and D145E () residues. Deviation of experimental (obs.) chemical shifts (δ) from random coils for backbone Cα (a) and Hα (b). Random-coil chemical shift values were obtained from ref. 27. Rectangles (I–IV) and black line segments (αN, αA, etc) identify Ca2+-binding sites and α-helices known to be present in WT cTnC. Alpha-helical segments show positive deviations for Cα and negative deviations for Hα, while β-strands and loops show the opposite. Dashed lines between α-D and α-E in a and b represent the D/E linker region. In (c), amide hydrogens with temperature coefficients that were linear but very different (>95% C.I.) for WT and mutant proteins are identified by residue. Temperature coefficients of the protons coincide (or differ only slightly) for the two isoforms except where one of the paired values is labeled with an arrow.

• 

  Figure 4. Residues with non-linear temperature coefficients for amide hydrogens are sampling a different chemical environment above the break. (a,b) Ca2+-binding sites III and IV reveal multiple residues with non-linear temperature dependence. Residues listed to the left of each panel are located before or after the cartoons in panels a and b (NL – nonlinear, L – linear and NA – not analysed). (c,d) Collection of 1H chemical shifts with nonlinear behavior for residues located within sites III and IV, respectively. Numbers beside line segments in (c–d) represent temperature coefficients in ppb/°C. Calcium ions (green) are depicted in both sites but available evidence indicates they are bound only weakly (or not at all) in sites III and IV of D145E.

• 

  Figure 5. C-domain residues of WT and D145E isoforms differ in stability, as measured by NMR cross-peak intensities between 15° and 55 °C. (a,b) Changes in resonance intensities of the C-domain as a function of temperature, normalized to values at 15 °C. Green residue intensities are unperturbed or increased by temperature changes and classified as non-decaying group (NDG). Red residues lose intensity and are classified as decaying group (DG). In Black is the biphasic decaying group (“mountain-like”, ML-DG) for D145E. (c) Bottom view of the mutant cTnC C-domain, highlighting in black and red the residues identified respectively as ML-DG and DG in panel (b). Calcium ions are shown as green spheres. Truncated curves in (a) and (b) represent peaks that were lost due to line broadening or overlap at higher temperatures.

• 

  Figure 6. The increase in site II Ca2+ binding affinity with temperature is greater for D145E than WT. Proteins double-labeled at their native cysteines with the fluorescent probe IAANS were titrated with Ca2+ in the presence of Mg2+ at 30 °C (a) or 45 °C (b). Panel (c) summarises the changes in Ca2+ affinity (pCa50) from 21° to 45 °C, with a vertical dashed line to indicate pCa50 values extrapolated to 37 °C. Protein concentration was 1 µM. Excitation was at 330 nm and emission was monitored at 450 nm. Symbols show mean ± s.e.m. (n = 4).

• 

  Figure 7. Temperature dependence for Ca2+ sensitivity and maximum force of cardiac skinned fibres is more pronounced following reconstitution with D145E cTnC. The Ca2+ sensitivity (pCa 8.0 to 4.0) of isometric force was measured at (a) 15° (b) 25° and (c) 30 °C in fibres depleted of native cTnC and reconstituted with HcTnC WT (○) or D145E (●). Ca2+ sensitivity was measured at all three temperatures on each fibre. Each curve is normalised to its own maximum (at pCa 4.0). (d) In a different set, each fibre was reconstituted with one isoform and maximal tension was measured at 21 °C and at one other temperature (see Methods). Data are normalised to the value at 21 °C. *p < 0.05 compared to WT at the same temperature. In (a–d), values shown are means ± s.e.m. (n = 4–5). In some cases error bars fall within the symbols.