Structural and functional effects of hereditary hemolytic anemia-associated point mutations in the alpha spectrin tetramer site

Abstract

The most common hereditary elliptocytosis (HE) and hereditary pyropoikilocytosis (HPP) mutations are alpha-spectrin missense mutations in the dimer-tetramer self-association site. In this study, we systematically compared structural and functional properties of the 14 known HE/HPP mutations located in the alpha-spectrin tetramer binding site. All mutant alpha-spectrin recombinant peptides were well folded, stable structures, with only the R34W mutant exhibiting a slight structural destabilization. In contrast, binding affinities measured by isothermal titration calorimetry were greatly variable, ranging from no detectable binding observed for I24S, R28C, R28H, R28S, and R45S to approximately wild-type binding for R34W and K48R. Binding affinities for the other 7 mutants were reduced by approximately 10- to 100-fold relative to wild-type binding. Some sites, such as R28, were hot spots that were very sensitive to even relatively conservative substitutions, whereas other sites were only moderately perturbed by nonconservative substitutions. The R34W and K48R mutations were particularly intriguing mutations that apparently either destabilize tetramers through mechanisms not probed by the univalent tetramer binding assay or represent polymorphisms rather than the pathogenic mutations responsible for observed clinical symptoms. All alpha0 HE/HPP mutations studied here appear to exert their destabilizing effects through molecular recognition rather than structural mechanisms.

Figures

Figure 1

Relationship between the human red cell spectrin dimer-tetramer equilibrium and tetramer site univalent recombinant peptides. (A) A model depicting the 2 equilibria in the overall dimer-tetramer equilibrium of human red cell spectrin. The first step in tetramer formation is opening of a closed dimer (top panel), followed by head-to-head association of 2 open dimers to form a tetramer. Dimer and tetramer models schematically illustrate the repeats that comprise the α and β monomers as follows: rectangles represent the many tandem homologous “spectrin type” repeats; the loop attached to the α9 repeat depicts the SH3 domain, which is inserted in the loop between the B and C helices of repeat 9 (this SH3 domain is designated α10 for historical reasons); the hexagons at the α chain C-terminus represent EF-hand regions (calmodulin-like domains); the elongated rectangle (ABD) at the N-terminus of β-spectrin represents the actin-binding domain (calponin homology domain); the squiggly “tail” represents the nonhomologous phosphorylated C-terminal end of β-spectrin. An enlarged view of the tetramerization site schematically illustrates the α0-1 and β16-17 recombinant peptides using cylinders to represent the 3 helix bundles. In this model, the tetramer binding site is composed of a C helix from the partial α0 repeat and a B and C helix from the β17 partial repeat. The amino acid residues and residue numbers in the α0 C helix that are mutated in HE/HPP patients are shown in the black bar immediately below the tetramer site model. (B) A 1-D 12% Bis-Tris SDS gel stained with Coomassie Brilliant Blue of the purified recombinant proteins (2 μg). Molecular weights of standard proteins are indicated on the left (in kilodaltons).

Figure 2

Thermal denaturation analysis of α0-1 recombinant peptides. Representative thermal denaturation experiments of wild-type and 2 HE mutant α0-1 peptides. DSC analysis was performed at a scan rate of 90°C/hr in 10 mM sodium phosphate, 130 mM NaCl, 1 mM β-ME, pH 7.4, at concentrations ranging from 0.6 to 1.0 mg/mL. Normalized spectra are shown for: the α0-1 wild-type (—), Tm = 54.6°C; the α0-1 R28C (···), which had the highest observed Tm of 55.6°C; and the α0-1 R34W (----), which had the lowest Tm at 49.1°C and the most atypical melting curve.

Figure 3

Representative isothermal titration calorimetry analyses of tetramerization site complexes. All experiments were conducted at 23°C with β16-17 in the reaction cell and the α0-1 peptide in the titration syringe. Top panels show baseline subtracted titration curve raw data; bottom panels show blank corrected integrated areas of the peaks from the top panel (●) and the best data fit using a nonlinear least-squares method (—). The protein concentrations of α0-1 mutants in the syringe were: R34W = 330 μM; K48R = 560 μM; G46V = 680 μM; R28H = 600 μM; V31A = 600 μM; R41W = 650 μM; R45T = 660 μM; R45S = 600 μM. The protein concentrations of the β16-17 wild-type recombinant in the cell were: 18 μM for R34W and 25 to 30 μM in all other cases.

Figure 4

Comparison of evolutionary sequence conservation within the spectrin α0 C-helix. Alignments of spectrin sequences encompassing approximately the C-helix of the α0 partial repeat involved in tetramer binding are shown. All known N-terminal region sequences of α-spectrins were extracted from the National Center for Biotechnology Information nonredundant protein sequence database using BLASTP 2.2.17 and human red cell spectrin (SPTA1) residues 1 to 158 as the query sequence. The severity of decreased tetramer binding affinity (Table 1) and residue numbers of the 14 known HE/HPP mutations are shown above the sequences. For these positions, residues that match the wild-type human red cell spectrin sequence are shown in bold type.