Structure and mutational analysis of a plant mitochondrial nucleoside diphosphate kinase. Identification of residues involved in serine phosphorylation and oligomerization

Abstract

We report the first crystal structure of a plant (Pisum sativum L. cv Oregon sugarpod) mitochondrial nucleoside diphosphate kinase. Similar to other eukaryotic nucleoside diphosphate kinases, the plant enzyme is a hexamer; the six monomers in the asymmetric unit are arranged as trimers of dimers. Different functions of the kinase have been correlated with the oligomeric structure and the phosphorylation of Ser residues. We show that the occurrence of Ser autophosphorylation depends on enzymatic activity. The mutation of the strictly conserved Ser-119 to Ala reduced the Ser phosphorylation to about one-half of that observed in wild type with only a modest change of enzyme activity. We also show that mutating another strictly conserved Ser, Ser-69, to Ala reduces the enzyme activity to 6% and 14% of wild-type using dCDP and dTDP as acceptors, respectively. Changes in the oligomerization pattern of the S69A mutant were observed by cross-linking experiments. A reduction in trimer formation and a change in the dimer interaction could be detected with a concomitant increase of tetramers. We conclude that the S69 mutant is involved in the stabilization of the oligomeric state of this plant nucleoside diphosphate kinase.

Figure 1.

Arrangement of the six pea mtNDPK monomers in the asymmetric unit. A, Ribbon representation of the hexamer viewed down the 3-fold axis, showing monomers A, C, and E in the foreground and monomers B, D, and F in the background. B, The hexamer in A rotated 90° with respect to the horizontal axis; monomers C and D are in the foreground. The position of Ser-69 and Trp-148 of adjacent monomers at the trimer interface are circled in red. C, Close-up around Ser-69 (blue) showing its interaction with Phe-66 of the same monomer and Trp-148 of an adjacent monomer of the trimer (orange). Drawn and rendered using Pymol (http://www.pymol.org).

Figure 2.

Overall structure and sequence similarity of pea mtNDPK and other representative NDPKs. A, Superposition of C-α traces of A monomers from pea mtNDPK (yellow), human NM23-H2 cytosolic NDPK (cyan), human NM23-H4 mtNDPK (blue), and Dictyostelium discoideum cytosolic NDPK (red). PDB codes for the structures are 1W7W, 1NSK, 1EHW, and 1F6T, respectively. Residues S69, H117, and S119 of the pea mtNDPK structure are represented as ball-and-sticks (drawn with Molscript; Kraulis, 1991) and rendered in Molray (Harris and Jones, 2001). B, Sequence alignment of pea mtNDPK and representative NDPKs. Secondary structure elements are indicated below in blue for α-helix and in yellow for β-strand and numbered consecutively along the sequence. The conserved amino acids Ser-69, Ser-119, and the active-site His, His-117, are highlighted in red, and the conserved Killer of prune Pro is in bold. The sequences are numbered according to the mature pea mtNDPK sequence. Residues for which density was not observed are indicated in italics. Accession codes for the sequences are: AAF08537, mitochondrial pea mtNDPK-3; O49203, mitochondrial Arabidopsis NDPK-3; O00746, mitochondrial human NH23-H4; P22392, cytosolic human NM23-H2; CAA50511.1, cytosolic pea NDPK-1; P39207, cytosolic Arabidopsis NDPK-1; and P22887, cytosolic D. discoideum. The structure and sequence alignments were made using the Indonesia package (http://xray.bmc.uu.se/dennis/manual/).

Figure 2.

Overall structure and sequence similarity of pea mtNDPK and other representative NDPKs. A, Superposition of C-α traces of A monomers from pea mtNDPK (yellow), human NM23-H2 cytosolic NDPK (cyan), human NM23-H4 mtNDPK (blue), and Dictyostelium discoideum cytosolic NDPK (red). PDB codes for the structures are 1W7W, 1NSK, 1EHW, and 1F6T, respectively. Residues S69, H117, and S119 of the pea mtNDPK structure are represented as ball-and-sticks (drawn with Molscript; Kraulis, 1991) and rendered in Molray (Harris and Jones, 2001). B, Sequence alignment of pea mtNDPK and representative NDPKs. Secondary structure elements are indicated below in blue for α-helix and in yellow for β-strand and numbered consecutively along the sequence. The conserved amino acids Ser-69, Ser-119, and the active-site His, His-117, are highlighted in red, and the conserved Killer of prune Pro is in bold. The sequences are numbered according to the mature pea mtNDPK sequence. Residues for which density was not observed are indicated in italics. Accession codes for the sequences are: AAF08537, mitochondrial pea mtNDPK-3; O49203, mitochondrial Arabidopsis NDPK-3; O00746, mitochondrial human NH23-H4; P22392, cytosolic human NM23-H2; CAA50511.1, cytosolic pea NDPK-1; P39207, cytosolic Arabidopsis NDPK-1; and P22887, cytosolic D. discoideum. The structure and sequence alignments were made using the Indonesia package (http://xray.bmc.uu.se/dennis/manual/).

Figure 3.

Autophosphorylation of recombinant mtNDPK. The recombinant purified mtNDPK proteins were incubated with [γ-³²P]ATP for 8 min at room temperature. The sample was divided into two aliquots and analyzed as indicated. A, Alkali-stable His autophosphorylation. B, Acid-stable Ser autophosphorylation. Phosphorylation was detected with a phosphor imager. Lanes 1 to 3, wild-type protein; lanes 4 to 6, S69A protein; lanes 7 to 9, S119A protein; lane 10, H117D protein; lane 11, S69A/S119A protein. Protein amounts in lanes 1, 4, and 7, 0.1 μg; 2, 5, and 8, 0.2 μg; 3, 6, 9, 10, and 11, 1 μg.

Figure 4.

Determination of the oligomerization state of recombinant mtNDPK. Cross-linking of mtNDPK with glutaraldehyde. Lanes 2 to 4 and 6 to 8 show cross-linking of wild-type and S69A proteins, respectively. Glutaraldehyde amounts in lanes 2 and 6, 0.007%; 3 and 7, 0.006%; 4 and 8, 0.005%. An equal protein amount was loaded for each lane. The relevant molecular masses are indicated in lanes 1 and 5.