A crucial active site network of titratable residues guides catalysis and NAD+ binding in human succinic semialdehyde dehydrogenase

Abstract

Human succinic semialdehyde dehydrogenase is a mitochondrial enzyme fundamental in the neurotransmitter γ-aminobutyric acid catabolism. It catalyzes the NAD⁺-dependent oxidative degradation of its derivative, succinic semialdehyde, to succinic acid. Mutations in its gene lead to an inherited neurometabolic rare disease, succinic semialdehyde dehydrogenase deficiency, characterized by mental and developmental delay. Due to the poor characterization of this enzyme, we carried out evolutionary and kinetic investigations to contribute to its functional behavior, a prerequisite to interpreting pathogenic variants. An in silico analysis shows that succinic semialdehyde dehydrogenases belong to two families, one human-like and the other of bacterial origin, differing in the oligomeric state and in a network of active site residues. This information is coupled to the biophysical-biochemical characterization of the human recombinant enzyme uncovering that (i) catalysis proceeds by an ordered bi-bi mechanism with NAD⁺ binding before the aldehyde that exerts a partial non-competitive inhibition; (ii) a stabilizing complex between the catalytic Cys340 and NAD⁺ is observed and interpreted as a protective mechanism; and (iii) a concerted non-covalent network assists the action of the catalytic residues Cys340 and Glu306. Through mutational analyses of Lys214, Glu306, Cys340, and Glu515 associated with pH studies, we showed that NAD⁺ binding is controlled by the dyad Lys214-Glu515. Moreover, catalysis is assured by proton transfer exerted by the same dyad networked with the catalytic Glu306, involved in catalytic Cys340 deprotonation/reprotonation. The identification of this weak bond network essential for cofactor binding and catalysis represents a first step to tackling the molecular basis for its deficiency.

Keywords: bioinformatic and evolutionary analysis; enzymatic mechanism; kinetics; succinic semialdehyde dehydrogenase; succinic semialdehyde dehydrogenase deficiency.

SCHEME 1

Metabolic pathways of GABA. The red cross denotes the misfunctioning or absence of hSSADH, leading to GABA and GHB accumulation (red arrows).

SCHEME 2

Proposed mechanism of reaction of hSSADH highlighting the role of Cys340 as a nucleophile and of Glu306 as acid–base catalyst. The catalytic Cys340 is deprotonated by the catalytic Glu306 (through a water molecule) (step 1) to generate a thiolate anion that (step 2) makes a nucleophilic attack to the formyl carbon of SSA. The produced thiohemiacetal intermediate transfers the hydride to NAD⁺, which is reduced to NADH (step 3) with the concomitant formation of the thioester intermediate. Glu306 acts as a general base (through a water molecule) to trigger the nucleophilic attack of a hydroxyl ion on the carbonyl carbon of the ester (step 4). This generates a tetrahedral intermediate with subsequent release of succinic acid (SA) and of Cys340 thiolate, which is finally reprotonated by Glu306 (step 5).

FIGURE 1

Diversity and evolution of SSADHs within ALDH superfamily and maximum likelihood unrooted phylogenetic tree of characterized SSADHs and ALDHs coded by human genes. (a) Sequence similarity networks (SSNs) of ALDH superfamily homologous to human ALDHs or characterized SSADHs from non‐human organisms. See the text for info on how the sequences were collected. The edge‐weighted spring embedded layout with respect to the edge weights was used for visualization. A dashed blue circle highlights the region of the sequence space encompassing nodes that have a minimum of 45% sequence identity to hSSADH. (b) See Table 1 for details on sequence identifiers. The structures used for aligning the sequences are from PDB, AlphaFold‐EBI protein structure database (https://alphafold.ebi.ac.uk/) or predicted with AF2; info reported in Table 1. The branch length is proportional to the expected substitution rate, according to the reference bar. Only nodes with bootstrap support <97 are labeled.

FIGURE 2

Evolutionary comparison of representative sequences from ALDH5 and Other SSADH families. (a) Taxonomy distribution of the source organisms for each SSADH family. (b) Box plot of evolutionary coupling (EC) scores (see section 4) for each residue pair predicted to interact at oligomeric interfaces in hSSADH simulations (Table S1). DI, residue pairs at dimeric interface; TI, residue pairs at tetrameric interface; Backbone, EC scores of residue pairs interacting by backbone atoms; Sidechain, EC scores of residue pairs interacting by sidechain atoms. (c) Frequency logo of catalytic and titratable residues of hSSADH within representative sequences of the two SSADHs families. (d) hSSADH active site with the catalytic Cys340 manually deprotonated (PDB ID: 2W8O). NAD⁺ and SSA were fully flexible docked with DynamicBind v.1.0 (plDDT_SSA ≈ 0.75, plDDT_NAD ⁺ ≈ 0.69; see section 4) and visualized in ball‐and‐sticks together with catalytic and titratable active site residues. Black dashed lines connect the sulfur atom of the catalytic Cys340 to the reactive carbon centers of NAD⁺ and SSA.

SCHEME 3

Proposed reaction mechanism for hSSADH. NAD⁺ binds first to the enzyme followed by SSA with the formation of the ternary complex. This then releases the SA and NADH products (straight lines). At high SSA concentrations, a partial non‐competitive inhibition exerted by SSA occurs, with the formation of a partial inhibitory ternary complex E‐SSA‐NADH. The reduced cofactor is then replaced by NAD⁺ to regenerate the catalytically competent ternary complex E‐SSA‐NAD⁺ (dashed lines). This outlined mechanism is based on extensive investigations previously reported for other aldehyde dehydrogenases that show a substrate (aldehyde) partial inhibition behavior similar to that of hSSADH (Munoz‐Clares and Casanova‐Figueroa 2019).

FIGURE 3

Burst kinetics and active site titration of hSSADH. Reactions were carried out at 20 μM SSA and 500 μM NAD⁺ by adding the following hSSADH concentrations: 4 μM (black), 8 μM (red), 10 μM (blue), 12 μM (green), and 14 μM (pink). The absorbance change at 340 nm was monitored using a stopped‐flow spectrophotometer at 4°C in 100 mM potassium phosphate buffer, 10 mM BME at pH 8. The inset shows the correlation of the concentration of NADH formed during the burst phase with hSSADH concentration.

FIGURE 4

Spectral characterization of WT hSSADH. All spectra were collected in 100 mM potassium phosphate buffer, 10 mM BME at pH 8 at 25°C. (a) Far UV CD spectra of 2 μM (monomer concentration) WT hSSADH in the absence (black line) or presence of 100 μM NAD⁺ (red line). (b) Near UV–visible spectra of 18 μM WT hSSADH in the absence (black line) or presence (blue line) of 200 μM NAD⁺. (c) Absorbance spectra of 18 μM WT hSSADH in the absence (black line) or presence (blue line) of 200 μM NAD⁺. (d) Plot of the quenching in intrinsic fluorescence spectra of 0.1 μM WT hSSADH in the presence of different concentrations of NAD⁺. Data were fitted to a hyperbola equation to obtain the K _D value. Inset: Intrinsic fluorescence spectra of 0.1 μM WT hSSADH (black line) in the presence of different concentrations of NAD⁺ (0.1 μM, red; 0.5 μM, dark green; 1 μM, orange; 4 μM, blue; 8 μM, gray; 20 μM, purple; 50 μM, light green; 75 μM, pink; and 100 μM, brown). The arrow indicates the change at increasing cofactor concentrations.

FIGURE 5

Dependence of K _DNAD+ on pH. All measurements were performed in 50 mM BTP and 10 mM BME at the indicated pH values. At each pH value, K _DNAD+ was obtained by measuring the quenching of intrinsic fluorescence determined by exciting 0.1 μM WT hSSADH (black), 0.1 μM Lys214Ala (blue), 0.1 μM Glu515Ala (green), and 0.4 μM Glu306Ala (red) at different NAD⁺ concentrations (from 0.1 to 100 μM) and fitting the data to a hyperbolic equation. The pK_a determination was obtained by fitting the resulting K _D values to Equation (6).

FIGURE 6

Dependence of kinetic parameters on pH. Kinetic parameters were determined at the following pH values: 5.4, 6.0, 6.5, 7.0, 7.5, 8.0, 8.5, 9.0, 9.5, and 10.0. All measurements were performed in 50 mM BTP and 10 mM BME at the indicated pH values. Log k _cat for NAD⁺ (a) and for SSA (b) versus pH, as well as Log k _cat/K _m for NAD⁺ (c) and for SSA (d) versus pH were determined and fitted to Equations (7) and (8), respectively.

FIGURE 7

Dependence of activity on pH for active SSADH variants. Activity was determined by measuring the formation of NADH (see section 4 for details) obtained by incubating 8 nM WT (black), or 0.1 μM Glu515Ala (green), or 0.35 μM Lys214Ala (blue) with 10 μM SSA and 500 μM NAD⁺. The amount of NADH produced was expressed as micromol/sec/micromol of tetrameric enzyme (v₀).