Regular article
Structural and biochemical investigations of the catalytic mechanism of an NADP-dependent aldehyde dehydrogenase from Streptococcus mutans1

https://doi.org/10.1006/jmbi.2000.3824Get rights and content

Abstract

The NADP-dependent non-phosphorylating glyceraldehyde-3-phosphate dehydrogenase from Streptococcus mutans (abbreviated Sm-ALDH) belongs to the aldehyde dehydrogenase (ALDH) family. Its catalytic mechanism proceeds via two steps, acylation and deacylation. Its high catalytic efficiency at neutral pH implies prerequisites relative to the chemical mechanism. First, the catalytic Cys284 should be accessible and in a thiolate form at physiological pH to attack efficiently the aldehydic group of the glyceraldehyde-3-phosphate (G3P). Second, the hydride transfer from the hemithioacetal intermediate toward the nicotinamide ring of NADP should be efficient. Third, the nucleophilic character of the water molecule involved in the deacylation should be strongly increased. Moreover, the different complexes formed during the catalytic process should be stabilised.

The crystal structures presented here (an apoenzyme named Apo2 with two sulphate ions bound to the catalytic site, the C284S mutant holoenzyme and the ternary complex composed of the C284S holoenzyme and G3P) together with biochemical results and previously published apo and holo crystal structures (named Apo1 and Holo1, respectively) contribute to the understanding of the ALDH catalytic mechanism.

Comparison of Apo1 and Holo1 crystal structures shows a Cys284 side-chain rotation of 110°, upon cofactor binding, which is probably responsible for its p Ka decrease. In the Apo2 structure, an oxygen atom of a sulphate anion interacts by hydrogen bonds with the NH2 group of a conserved asparagine residue (Asn154 in Sm-ALDH) and the Cys284 NH group. In the ternary complex, the oxygen atom of the aldehydic carbonyl group of the substrate interacts with the Ser284 NH group and the Asn154 NH2 group. A substrate isotope effect on acylation is observed for both the wild-type and the N154A and N154T mutants. The rate of the acylation step strongly decreases for the mutants and becomes limiting. All these results suggest the involvement of Asn154 in an oxyanion hole in order to stabilise the tetrahedral intermediate and likely the other intermediates of the reaction. In the ternary complex, the cofactor conformation is shifted in comparison with its conformation in the C284S holoenzyme structure, likely resulting from its peculiar binding mode to the Rossmann fold (i.e. non-perpendicular to the plane of the β-sheet). This change is likely favoured by a characteristic loop of the Rossmann fold, longer in ALDHs than in other dehydrogenases, whose orientation could be constrained by a conserved proline residue. In the ternary and C284S holenzyme structures, as well as in the Apo2 structure, the Glu250 side-chain is situated less than 4 Å from Cys284 or Ser284 instead of 7 Å in the crystal structure of the wild-type holoenzyme. It is now positioned in a hydrophobic environment. This supports the p Ka assignment of 7.6 to Glu250 as recently proposed from enzymatic studies.

Introduction

Aldehyde dehydrogenases (ALDH) belong to a family of enzymes that oxidise aldehydes into their carboxylic acids in the presence of NAD(P). For example, they participate in ethanol metabolism by conversion of acetaldehyde into acetic acid (Steinmetz et al., 1997), or in the cellular differentiation as they produce retinoic acid by retinal oxidation (Moore et al., 1998). In Streptococcus mutans, an ALDH (Sm-ALDH) catalyses NADPH production during the irreversible oxidation of glyceraldehyde-3-phosphate (G3P) into 3-phosphoglycerate (3-PGA) (Boyd et al., 1995).

Also found in photosynthetic organisms, this enzyme is defined as an NADP-dependent non-phosphorylating glyceraldehyde-3-phosphate dehydrogenase (GAPDH) on the basis of its reaction. In spite of its name, this 200 kDa enzyme has to be classified in the ALDH family, as it does not possess sequence identity with the other phophorylating GAPDHs for which the molecular mass is 145 kDa.

In ALDHs, the catalytic reaction occurs in two steps. Firstly, the acylation step occurs via the nucleophilic attack of the catalytic cysteine residue on the substrate (Cys284 in Sm-ALDH), leading to formation of the hemithioacetal intermediate next to the hydride transfer and the thioacylenzyme and NAD(P)H. Secondly, the deacylation step involves a water molecule. To be efficient, these acylation and deacylation steps require cysteine and water molecule activations at physiological pH as well as stabilisation of the different intermediates of the reaction.

Concerning the cysteine activation, it was proposed that a conserved glutamate residue (Glu250 in Sm-ALDH) might act as the general base. Its putative role might be to activate a water molecule observed in the ALDH2 holoenzyme structure (Steinmetz et al., 1997). This water molecule might then deprotonate the cysteine residue to form the thiolate ion (Steinmetz et al., 1997). However, the thiolate formation might be due to the proximity of the main-chain NH groups on both sides of the catalytic cysteine residue (i.e. the peptide nitrogen atoms of Cys284 and Thr285 in Sm-ALDH) (Steinmetz et al., 1997) and/or to the positive charge of the NAD(P) nicotinamide ring Steinmetz et al 1997, Johansson et al 1998 thus decreasing the cysteine p Ka. Recently, it was proposed that Cys284 activation is accompanied by a slight conformational change during cofactor binding (Marchal & Branlant, 1999). A careful investigation of the comparison of the previously solved Sm-ALDH apo and holo enzyme structure catalytic sites (Cobessi et al., 1999) (named Apo1 and Holo1) shows that a Cys284 side-chain rotation accompanies cofactor binding.

In the deacylation, the thioacylenzyme undergoes nucleophilic attack by a water molecule likely positioned and/or activated by its protein environment. Recently, it was proposed that Glu250 in Sm-ALDH could be the residue involved (Marchal et al., 2000). Movements of the reduced cofactor are also postulated to be necessary to position the water molecule between the cysteine and the glutamate residues (Moore et al., 1998). In the Sm-ALDH Apo1 and Holo1 structures (Cobessi et al., 1999), the Glu250 carboxylate group is situated approximately 7 Å from the Cys284 sulphur atom, as observed in the other ALDH structures. The present analyses of the catalytic sites of the Apo1, Holo1, Apo2 and ternary complex structures of Sm-ALDH in relation with the biochemical results Marchal and Branlant 1999, Marchal et al 2000 suggest a catalytic mechanism in which Glu250 has no role in the acylation. The data presented here highlight the role of an oxyanion hole in the stabilisation of the tetrahedral intermediate and likely of the other intermediates of the reaction. We compare the Rossmann fold of the ALDHs and other dehydrogenases in order to understand the structural determinants of the new NAD(P) binding mode in ALDHs, which seem important for an efficient catalytic reaction.

Section snippets

The Apo2 enzyme structure

Apo2 crystals were obtained under the same crystallisation conditions as those used for Apo1 and Holo1. The structure was solved at 2.5 Å resolution (see Materials and Methods) and revealed a fold similar to the two other reported Sm-ALDH structures (Cobessi et al., 1999) (rms between Apo1 and Apo2 tetramers, 0.51 Å). The major difference concerns the position of a few side-chains in the catalytic site, most notably Glu250.

In the catalytic site of Apo2, the side-chain orientation of Cys284

Conclusion

Surprisingly, two Sm-ALDH apoenzyme structures show catalytic sites differing by the side-chain positions of a few residues, notably the catalytic Glu250 and Arg437 observed in interaction with Glu250 in Apo1. The side-chain rotation, due to the binding of a second sulphate anion in Apo2, reveals Glu250 flexibility. This flexibility is confirmed by comparing its observed position in the Apo1, Holo1, binary and ternary complex of C284S mutant structures. The present structural studies combined

Materials

C284S, N154A and N154T mutants were produced and purified following a procedure similar to that described for wild-type and other Sm-ALDH mutants (Marchal & Branlant, 1999).

Crystallisation and data collection

Apo2 was crystallised under two sets of conditions, using the hanging drop method at 20°C: 2 μl of protein solution at a concentration of 5 mg/ml was mixed with 2 μl of reservoir containing 2.0 M ammonium sulphate, 0.1 M cacodylate buffer (pH 6.5) and over a range of 1 to 5 % acetone. These Apo2 crystals diffract poorly at 3

Acknowledgements

We gratefully acknowledge Professor Vitoux and Dr Rahuel-Clermont for helpful discussions. This work was technically supported by the Service Commun de Diffraction X sur Monocristaux at the Henri Poincaré University. We thank E. Habermacher, S. Boutserin and S. Azza for their efficient technical help. This work was financially supported by the Centre National de la Recherche Scientifique, the University Henri Poincaré Nancy I and the Institut Fédératif de Recherche Protéines.

References (25)

  • A.T. Brünger

    X-PLOR Manual

    (1990)
  • A.T. Brünger

    The Rfree valuea novel statistical quantity for assessing the accuracy of crystal structures

    Nature

    (1992)
  • Cited by (0)

    1

    Edited by I. A. Wilson

    View full text