Anti-Correlation between the Dynamics of the Active Site Loop and C-Terminal Tail in Relation to the Homodimer Asymmetry of the Mouse Erythroid 5-Aminolevulinate Synthase

Abstract

Biosynthesis of heme represents a complex process that involves multiple stages controlled by different enzymes. The first of these proteins is a pyridoxal 5′-phosphate (PLP)-dependent homodimeric enzyme, 5-aminolevulinate synthase (ALAS), that catalyzes the rate-limiting step in heme biosynthesis, the condensation of glycine with succinyl-CoA. Genetic mutations in human erythroid-specific ALAS (ALAS2) are associated with two inherited blood disorders, X-linked sideroblastic anemia (XLSA) and X-linked protoporphyria (XLPP). XLSA is caused by diminished ALAS2 activity leading to decreased ALA and heme syntheses and ultimately ineffective erythropoiesis, whereas XLPP results from “gain-of-function” ALAS2 mutations and consequent overproduction of protoporphyrin IX and increase in Zn²⁺-protoporphyrin levels. All XLPP-linked mutations affect the intrinsically disordered C-terminal tail of ALAS2. Our earlier molecular dynamics (MD) simulation-based analysis showed that the activity of ALAS2 could be regulated by the conformational flexibility of the active site loop whose structural features and dynamics could be changed due to mutations. We also revealed that the dynamic behavior of the two protomers of the ALAS2 dimer differed. However, how the structural dynamics of ALAS2 active site loop and C-terminal tail dynamics are related to each other and contribute to the homodimer asymmetry remained unanswered questions. In this study, we used bioinformatics and computational biology tools to evaluate the role(s) of the C-terminal tail dynamics in the structure and conformational dynamics of the murine ALAS2 homodimer active site loop. To assess the structural correlation between these two regions, we analyzed their structural displacements and determined their degree of correlation. Here, we report that the dynamics of ALAS2 active site loop is anti-correlated with the dynamics of the C-terminal tail and that this anti-correlation can represent a molecular basis for the functional and dynamic asymmetry of the ALAS2 homodimer.

Keywords: ALAS; anti-correlated dynamics; homodimer asymmetry; intrinsically disordered region; molecular dynamics.

Figure 1

Analysis of a protein homologous to ALAS2 but with a structured C-terminal tail region. (A) Evaluation of the intrinsic disorder propensity of ALAS2 by a series of disorder predictors. (B) Modelled 3D structure of one of the ALAS2 protomers (ALASdChB) upon MD simulation at 310 K (ALASdChB-310 K 1st cluster). (C) Evaluation of the intrinsic disorder propensity of Methanothermobacter thermautotrophicus phosphoribosyl-AMP cyclohydrolase HisI by a series of disorder predictors. (D) 3D structure of HisI (PDB ID 1ZPS). Intrinsic disorder predispositions of both proteins were evaluated using PONDR^® VLXT, PONDR^® VSL2, and PONDR^® FIT, and the outputs of individual predictors were averaged (FX2). The ALAS2 active site loop and the C-tail of His I are indicated in red in plots (B,D), respectively.

Figure 2

Two-dimensional backbone RMSD representation of the MD simulation results for comparison of the dynamics of the conserved region and the C-terminal tail region of three different ALAS2 systems at 300 K. (A) ALASm, (B) ALASdChA, (C) ALASdChB. The 288 K and 310 K RMSD landscapes also showed similar patterns (Supplementary, Figure S1).

Figure 3

Backbone RMSF plots for ALAS2 and HemAm at three different temperatures. (A) ALASm, (B) ALASdChA, (C) ALASdChB, (D) ALASmWoT, (E) HemAm.

Figure 4

Analysis of the secondary structure propensity of the active site loop in MD simulations of different ALAS2 and HemAm systems. (A) The number of hydrogen bonds in the active site loop at different temperatures. (B) The number of β-strand forming amino acids in the active site loop at different temperatures. Legends for proteins and color code in plot B are the same as in A. Central dots indicate the mean values, and error bars represent the standard deviations. (C) Correlation between the mean hydrogen bond number and the mean number of β-strand forming amino acids for four ALAS2 systems (i.e., ALASm, ALASdChA, ALASdChB and ALASmWot). (y = −0.04 + 1.41 x, where y is the N. β-strand-forming residues mean, and x is the mean number of hydrogen bonds) (D) Correlation between the mean number of hydrogen bond and the mean number of the β-strand forming amino acid for four ALAS2 systems (i.e., ALASm, ALASdChA, ALASdChB and ALASmWoT). (y = 2.22 + 0.20 x, where y is the mean number of β-strand-forming residues, and x is the hydrogen bond number mean. R² is the adjusted r-squared value from R).

Figure 5

Correlation maps for the MD simulations of the wild-type C-terminal tail-containing ALAS2 systems at 300 K. (A) ALASm, (B) ALASdChA, (C) ALASdChB. 288 K, and 310 K correlation maps are provided in Supplementary, Figure S3. Black circles indicate the positioning of the anti-correlated data. (D–F). Zoomed in versions of (A–C), respectively.

Figure 5

Correlation maps for the MD simulations of the wild-type C-terminal tail-containing ALAS2 systems at 300 K. (A) ALASm, (B) ALASdChA, (C) ALASdChB. 288 K, and 310 K correlation maps are provided in Supplementary, Figure S3. Black circles indicate the positioning of the anti-correlated data. (D–F). Zoomed in versions of (A–C), respectively.

Figure 6

Representative structure of the ALAS2 homodimeric form observed in the molecular dynamics simulation. ∆C_ij, and ∆RMSF measurement description is provided in a diagram. Red part of protein structure represents the active site loop, and blue part represents C-terminal tail (C-tail) in both protomers (Green: Chain A, Cyan: Chain B).