Structural Insights into the Nucleotide-Binding Domains of the P1B-type ATPases HMA6 and HMA8 from Arabidopsis thaliana

Abstract

Copper is a crucial ion in cells, but needs to be closely controlled due to its toxic potential and ability to catalyse the formation of radicals. In chloroplasts, an important step for the proper functioning of the photosynthetic electron transfer chain is the delivery of copper to plastocyanin in the thylakoid lumen. The main route for copper transport to the thylakoid lumen is driven by two PIB-type ATPases, Heavy Metal ATPase 6 (HMA6) and HMA8, located in the inner membrane of the chloroplast envelope and in the thylakoid membrane, respectively. Here, the crystal structures of the nucleotide binding domain of HMA6 and HMA8 from Arabidopsis thaliana are reported at 1.5Å and 1.75Å resolution, respectively, providing the first structural information on plants Cu+-ATPases. The structures reveal a compact domain, with two short helices on both sides of a twisted beta-sheet. A double mutant, aiding in the crystallization, provides a new crystal contact, but also avoids an internal clash highlighting the benefits of construct modifications. Finally, the histidine in the HP motif of the isolated domains, unable to bind ATP, shows a side chain conformation distinct from nucleotide bound structures.

Fig 1. 3D structures of HMA6 and HMA8 N-domains.

A The structure of HMA6^N-AA with β-strands shown in blue and α-helices in red. Missing loops are outlined by dashed lines. B The structure of HMA8^N; colouring is as in A. C Superimposition of the two structures with HMA6^N-AA in red and HMA8^N in blue.

Fig 2. Sequence alignment of P_IB-type ATPase N-domains.

Alignment of the region covering the N-domains of HMA1, HMA6 and HMA8 from A. thaliana, CopA and CopB from A. fulgidus, CopA from L. pneumophila, CopB from S. solfataricus, the human Menkes (ATP7A) and Wilson proteins (ATP7B) and ZntA from S. sonnei. Disordered regions in HMA6 and HMA8 are marked by a red rectangle. The mutated residues in HMA6 are marked in orange. The secondary structures of HMA6 and HMA8 are indicated above and below the alignment, respectively. Multiple sequence alignment was done using MUSCLE (http://www.ebi.ac.uk/Tools/msa/muscle) and visualized using ESPript 3.0 [35].

Fig 3. Comparison of P_IB-type ATPase N-domains of structural homologs.

Superimposition of HMA6^N-AA (gray), HMA8^N (red), LpCopA (3RFU, green), AfCopA (2B8E, cyan) and AfCopB (3SKX, orange), SsCopB (2IYE, yellow) and the Wilson disease protein (2ARF, blue).

Fig 4. Effect of the double alanine mutant in HMA6^N.

Interface of helix 4 from chain A and the symmetry related chain A* with glutamate modelled at position 709 and 710, matching the WT sequence instead of the two alanine, as found in the structure. Helix 4 is shown in orange in the symmetry related side chain marked with an asterisk.

Fig 5

A Superimposition of the ATP binding site of HMA6^N-AA (red), HMA8^N (orange), SsCopB (blue, 2YJ4) and AfCopA (light-blue, 3A1D). Conserved residues are shown as sticks, coloured as the corresponding chain and numbered according to their position in HMA6. ATP and ADP of SsCopB and AfCopA are shown as sticks with carbon atoms show in grey, nitrogen atoms in blue, oxygen atoms in red and phosphate atoms in orange. B Structure of HMA6^N-AA and C HMA8^N, each with the ATP of the SsCopB structure modelled. For the positioning of the nucleotide the N-domains of HMA6^N-AA, HMA8^N and SsCopB (2YJ4) were superimposed and as the conserved residues are spatially in close proximity a model of HMA6^N-AA and HMA8^N together with the ATP of SsCopB was generated. Conserved residues important for the binding are shown in sticks and the ATP as spheres. Colouring as in A.