Structure of Active IspH Enzyme from Escherichia coli Provides Mechanistic Insights into Substrate Reduction
Structure of Active IspH Enzyme from Escherichia coli Provides Mechanistic Insights into Substrate Reduction
复制标题
DOI:
10.1002/anie.200900548
复制
发表时间:
2009-01-01
影响因子:
16.6
通讯作者:
Groll, Michael
中科院分区:
文献类型:
--
作者:
Graewert, Tobias;Rohdich, Felix;Groll, Michael
Eukaryotes and most prokaryotes require isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) as biosynthetic precursors of terpenes. Whereas animals generate these essential metabolites via the mevalonate pathway,[1] many human pathogens including Plasmodium falciparum and Mycobacterium tuberculosis are known to use the more recently identified non-mevalonate pathway, which is a potential target for drug development.[2–4] The final step of this pathway is catalyzed by IspH protein, which generates a mixture of IPP and DMAPP by reductive dehydration of 1-hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate(HMBPP, Figure 1 a).[5–11] Recently, Rekittke et al. described the first X-ray structure of IspH protein from the hyperthermophilic eubacterium Aquifex aeolicus in its open state.[12] Herein, we report the crystal structure of the IspH protein from Escherichia coli [11] in its closed conformation, which serves as basis for a detailed discussion of the catalytic pathway. Recombinant E. coli IspH protein (comprising an N-terminal His6 fusion tag) was purified and crystallized under anaerobic conditions. Its structure was determined to a resolution of 1.8 by single-wavelength anomalous diffraction methods. Three iron sites per protein unit were localized in the anomalous difference Patterson map and were used for phasing. Successive rounds of model building and refinement afforded a well-defined electron density for the entire IspH molecule except for the N-terminal His6 tag and five C-terminal amino acid residues (Rfree= 23.8%, SupportingInformation, Table S2). The root mean square (rms) deviation between the Cα positions of the two protein molecules in the asymmetric unit is less than 0.3. The folding pattern of the monomeric protein involves three structurally similar domains, D1 to D3, which are related by pseudo-C3 symmetry but are devoid of detectable sequence similarity (Figure 1b and Supporting Information, FigureS1). Relative to domain D1, domains D2 and D3 appear rotated by angles of approximately 1008 and 1408, respectively. Each domain starts with a conserved cysteine residue that protrudes into a cavity at the center of the protein where it coordinates one respective iron atom of a [3Fe-4S] cluster. The cluster appears to be tilted relative to the pseudotrigonal axis of the apoprotein by about 208. The trigonal symmetric [3Fe-4S] cluster is located in a hydrophobic pocket of the central cavity, which is formed by residues located on D1 (G14 and V15), D2 (P97 and V99), D3 (A199) as well as the C-terminus (F302 and P305), which stabilizes the arrangement of the individual domains. Furthermore, the methylene moiety of C96 in D2 is turned inward generating additional hydrophobic shielding of atom Fe2 (see Figure 2). Residual electron density located inside the central cavity was identified as inorganic diphosphate (PPi; see Supporting