Article Type: Assignment Notes.

Mark Jeeves, K. John Smith*, Philip G. Quirk, Nick P.J. Cotton and J. Baz Jackson.

School of Biochemistry, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK.

*To whom correspondence should be addressed.

Telephone: 44-(0)-121-414-5394

Fax: 44-(0)-121-414-3982

Email: K.J.Smith@bham.ac.uk

Key words: NADP(H), NMR assignment, nucleotide-binding, proton translocation, Rhodospirillum rubrum, transhydrogenase.

Transhydrogenase is a proton pump, found in the inner membrane of animal mitochondria, and the cytoplasmic membrane of bacteria. It has a tripartite structure. Domains I and III protrude from the membrane (on the cytoplasmic side in bacteria, and on the matrix side in mitochondria). The domain II component spans the membrane, and serves as a channel for proton conduction. Transhydrogenase couples the transfer of reducing equivalents (hydride ion equivalents) between NAD(H) and NADP(H) to the translocation of protons across the membrane (reviewed in Jackson et al., 1998),

nH⁺_out + NADH + NADP⁺ Û nH⁺_in + NAD⁺ + NADPH (equ. 1)

where n is probably 1.0. Hydride transfer between NAD(H) bound to domain I, and NADP(H) bound to domain III, is direct and proceeds without involvement of intermediate redox reactions, implying that the C4 atoms of the nicotinamide rings of the two nucleotides must be brought into close apposition during catalysis. Under physiological conditions the equilibrium is driven from left to right (equ. 1) by the D p generated by the primary proton pumps of respiration or photosynthesis. The function of transhydrogenase in bacteria is to provide NADPH for amino acid biosynthesis and for the reduction of glutathione. In mitochondria, it is again required in the production of reduced glutathione, and it may have a role in the regulation of flux through the tricarboxylic acid cycle.

In different species, the three domains of transhydrogenase are distributed across one, two or three polypeptide chains. For example, in the photosynthetic bacterium Rhodospirillum rubrum, there are three polypeptides, PntAA (comprising domain I), PntAB (comprising domain IIa) and PntB (comprising domains IIa and III). In Escherichia coli there are two independent polypeptides, the first consisting of domains I-IIa, and the second of domains IIb-III. In bovine mitochondria and in the parasitic protozoan Eimeria tenella there is a single polypeptide, with the domains arranged in the orders I-II-III and IIb-III-I-IIa, respectively. The isolated domains I and III of transhydrogenase from a number of organisms have been cloned and expressed, for example, domain I protein (Diggle et al., 1995) and domain III protein (Diggle et al., 1996) from R.rubrum. In solution, mixtures of expressed domain I and domain III (even from enzymes of different species) are catalytically active (Diggle et al., 1996; Fjellstrom et al., 1997), and hence provide a convenient system in which to investigate the relationship between the structure of the domains I and III and the mechanism of hydride transfer. We here report the sequence-specific assignments for isolated recombinant domain III from R.rubrum. Calculation of the structure of this isolated domain III-NADP⁺ complex is underway.

Transhydrogenase domain III from R. rubrum (molecular weight 21.5 kDa, 203 amino acids) was cloned, expressed and purified as described in Diggle et al. (1996), except that the protein was cloned into pET11c vector and E.coli BL21(DE3) host cells were used for the expression. Typically yields of purified labeled protein were 20 mg per litre minimal medium (M9, 2g/L ¹³C₆-labeled glucose, 1g/L ¹⁵NH₄Cl). During the preparation procedure all the nucleotide bound to domain III was converted to NADP⁺ by mixing domain III with domain I (in a molar ration of 1:20 with domain III) and 50m M acetlypyridine adenine dinucleotide (AcPAD⁺) followed by subsequent chromatographic isolation of the domain III protein. Protein solutions of the U-¹⁵N- or U-¹³C,¹⁵N-labelled domain III-NADP⁺ complex were prepared at 800m M in 20mM Hepes, pH 7.2, 0.01% (w/v) NaN₃, 20m M AEBSF protease inhibitor (ICN Biomedicals Inc.), 2m M excess NADP⁺, 90% ¹H₂O / 10% ²H₂O. All experiments were performed on a 3-channel Varian Unityplus 600 spectrometer at 30 ° C, using a 5mm triple resonance ¹H/¹³C/¹⁵N Z-gradient probe.

Main-chain ¹H^N, ¹⁵N, ¹³Ca and side-chain ¹³Cb resonances were assigned using HNCACB and CBCA(CO)NNH (Muhandiram & Kay, 1994) and Cbd-HNCA (Matsuo et al., 1996) experiments to establish segments of sequential connectivity. Main-chain ¹Ha and ¹³C¢ assignments were made using CBCACO(CA)HA (Kay, 1993), HNCO (Muhandiram & Kay, 1994), HBHA(CBCA)(CO)NNH (Grzesiek & Bax, 1993) and HNHA (Kuboniwa et al., 1994). Where appropriate, selective carbon decoupling was achieved using WURST-2 adiabatic decoupling schemes (Matsuo et al., 1996).

Sequence-specific assignments (¹H^N, ¹⁵N, ¹³Ca , ¹Ha , ¹³Cb , ¹³C¢ ) for recombinant domain III from R. rubrum have been deposited in the BioMagResBank (http://www.bmrb.wisc.edu) database (accession number 4236). Figure 1 shows an annotated ¹H-¹⁵N HSQC spectrum of domain III. Virtually complete resonance assignments have been made for the protein between residues F13 and N203 (98% of assignments made, with only residue Y147, and the proline residues having less than four resonances assigned). At the N-terminus no data are given for residues M1-G12 (only tentative assignments have been made). Based upon the alignment of transhydrogenase sequences from different species (Kramer et al, 1993) this N-terminal sequence appears to form part of a linker region between domains IIb and III. This is confirmed by ¹⁵N relaxation and NOE data which show that, independent of the remainder of the protein, residues M1-G12 are mobile and largely unstructured (data not shown). The frequencies of ¹H resonances were referenced to internal DSS (2,2-dimethyl-2-silapentane-5-sulphonate) and the chemical shifts of ¹³C and ¹⁵N resonances were then indirectly referenced to DSS.

Financial support from the BBSRC and the Wellcome Foundation is gratefully acknowledged. We thank Mr. A. J. Pemberton for skillful maintenance of the NMR facilities, and Dr. Eriks Kupce (Varian UK) for assistance with the Cbd-HNCA spectrum.

References:

Diggle, C., Hutton, M., Jones, G. R., Thomas, C. M. and Jackson, J. B. (1995) Eur.J.Biochem. 228, 719-726.

Diggle, C., Bizouarn, T., Cotton, N. P. J. and Jackson, J. B. (1996) Eur.J.Biochem. 241, 162-170.

Fjellstrom, O., Johansson, C. and Rydstrom, J. (1997) Biochemistry 36, 11331-11341.

Grzesiek,S. and Bax,A. (1993) J.Biomol.NMR, 3, 185-204.

Jackson, J. B., Quirk, P. G., Cotton, N. P. J., Venning, J. D., Gupta, S., Bizouarn, T., Peake, S. J. and Thomas, C. M. (1998) Biochim.Biophys.Acta 1365, 79-86.

Kay,L.E. (1993) J.Am.Chem.Soc., 115, 2055-2057.

Kramer, R.A., Tomchak, L.A., McAndrew, S.J., Becker, K., Hug, D., Pasamontes, L., and Humbelin, M. (1993) Mol. Biochem. Parasitol., 60, 327-332.

Kuboniwa, H., Grzesiek, S., Delaglio, F. and Bax, A. (1994) J.Biomol.NMR, 4, 871-878.

Matsuo, H., Kupce, E., Li, H. and Wagner, G. (1996) J.Magn.Reson. 113, 91-96.

Muhandiram, D.R. and Kay, L.E. (1994) J.Magn.Reson.SerB. 103, 203-216.

Sensitivity enhanced 2D ¹H-¹⁵N HSQC spectrum of 800m M ¹⁵N-labeled transhydrogenase domain III from R. rubrum at 30 ° C and pH 7.2. Backbone resonances are labeled with the residue number, and side chain NH₂ resonances of asparagine and glutamine are connected by bars. Peaks marked ? are not assigned. The numbering system places residue M1 at amino acid position M262 in the Pnt B sequence of Diggle et al. (1996).