Title: Involvement of the Pyrophosphate and the 2′-Phosphate Binding Regions of Ferredoxin-NADP+ Reductase in Coenzyme Specificity
Abstract: Previous studies indicated that the determinants of coenzyme specificity in ferredoxin-NADP+ reductase (FNR) from Anabaena are situated in the 2′-phosphate (2′-P) NADP+ binding region, and also suggested that other regions must undergo structural rearrangements of the protein backbone during coenzyme binding. Among the residues involved in such specificity could be those located in regions where interaction with the pyrophosphate group of the coenzyme takes place, namely loops 155–160 and 261–268 in Anabaena FNR. In order to learn more about the coenzyme specificity determinants, and to better define the structural basis of coenzyme binding, mutations in the pyrophosphate and 2′-P binding regions of FNR have been introduced. Modification of the pyrophosphate binding region, involving residues Thr-155, Ala-160, and Leu-263, indicates that this region is involved in determining coenzyme specificity and that selected alterations of these positions produce FNR enzymes that are able to bind NAD+. Thus, our results suggest that slightly different structural rearrangements of the backbone chain in the pyrophosphate binding region might determine FNR specificity for the coenzyme. Combined mutations at the 2′-P binding region, involving residues Ser-223, Arg-224, Arg-233, and Tyr-235, in combination with the residues mentioned above in the pyrophosphate binding region have also been carried out in an attempt to increase the FNR affinity for NAD+/H. However, in most cases the analyzed mutants lost the ability for NADP+/H binding and electron transfer, and no major improvements were observed with regard to the efficiency of the reactions with NAD+/H. Therefore, our results confirm that determinants for coenzyme specificity in FNR are also situated in the pyrophosphate binding region and not only in the 2′-P binding region. Such observations also suggest that other regions of the protein, yet to be identified, might also be involved in this process. Previous studies indicated that the determinants of coenzyme specificity in ferredoxin-NADP+ reductase (FNR) from Anabaena are situated in the 2′-phosphate (2′-P) NADP+ binding region, and also suggested that other regions must undergo structural rearrangements of the protein backbone during coenzyme binding. Among the residues involved in such specificity could be those located in regions where interaction with the pyrophosphate group of the coenzyme takes place, namely loops 155–160 and 261–268 in Anabaena FNR. In order to learn more about the coenzyme specificity determinants, and to better define the structural basis of coenzyme binding, mutations in the pyrophosphate and 2′-P binding regions of FNR have been introduced. Modification of the pyrophosphate binding region, involving residues Thr-155, Ala-160, and Leu-263, indicates that this region is involved in determining coenzyme specificity and that selected alterations of these positions produce FNR enzymes that are able to bind NAD+. Thus, our results suggest that slightly different structural rearrangements of the backbone chain in the pyrophosphate binding region might determine FNR specificity for the coenzyme. Combined mutations at the 2′-P binding region, involving residues Ser-223, Arg-224, Arg-233, and Tyr-235, in combination with the residues mentioned above in the pyrophosphate binding region have also been carried out in an attempt to increase the FNR affinity for NAD+/H. However, in most cases the analyzed mutants lost the ability for NADP+/H binding and electron transfer, and no major improvements were observed with regard to the efficiency of the reactions with NAD+/H. Therefore, our results confirm that determinants for coenzyme specificity in FNR are also situated in the pyrophosphate binding region and not only in the 2′-P binding region. Such observations also suggest that other regions of the protein, yet to be identified, might also be involved in this process. Although the general catalytic mechanism of many NAD(P)+/H-dependent flavoenzymes is known, and despite the fact that the only difference between the two coenzymes is the 2′-P of the NADP+/H, the mechanism by which each enzyme is able to recognize either NAD+/H or NADP+/H is not yet completely understood (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 2.Elmore C.L. Porter T.D. J. Biol. Chem. 2002; 277: 48960-48964Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 3.Carrillo N. Ceccarelli E.A. Eur. J. Biochem. 2003; 270: 1900-1915Crossref PubMed Scopus (221) Google Scholar). Nevertheless, during the last decade considerable progress has been made in the study of the determinants of coenzyme specificity in different pyridine nucleotidedependent enzymes with the ultimate goal of shifting such coenzyme specificity via site-specific mutagenesis (2.Elmore C.L. Porter T.D. J. Biol. Chem. 2002; 277: 48960-48964Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 4.Scrutton N.S. Berry A. Perham R.N. Nature. 1990; 343: 38-43Crossref PubMed Scopus (646) Google Scholar, 5.Chen R. Greer A. Dean A.M. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 11666-11670Crossref PubMed Scopus (99) Google Scholar, 6.Perozich J. Kuo I. Wang B.C. Boesch J.S. Lindahl R. Hempel J. Eur. J. Biochem. 2000; 267: 6197-6203Crossref PubMed Scopus (46) Google Scholar, 7.Döhr O. Paine M.J.I. Friedberg T. Roberts G.C.K. Wolf C.R. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 81-86Crossref PubMed Scopus (73) Google Scholar). The flavoenzyme ferredoxin-NADP+ reductase (FNR) 1The abbreviations used are: FNRferredoxin-NADP+ reductaseFNRoxFNR in the oxidized stateFNRrdFNR in the reduced stateFNRsqFNR in the semiquinone stateFdferredoxinFdrdFd in the reduced statedRf5-deazariboflavin2′-P2′-phosphateETelectron transfer2′-P-AMP2′-phospho-AMP portion of NADP+/HWTwild typeMes4-morpholineethanesulfonic acid. catalyzes the reduction of NADP+ to NADPH in photosynthesis (3.Carrillo N. Ceccarelli E.A. Eur. J. Biochem. 2003; 270: 1900-1915Crossref PubMed Scopus (221) Google Scholar, 8.Hurley J.K. Morales R. Martínez-Júlvez M. Brodie T.B. Medina M. Gómez-Moreno C. Tollin G. Biochim. Biophys. Acta. 2002; 1554: 5-21Crossref PubMed Scopus (75) Google Scholar). This reaction is highly specific for NADP+/H relative to NAD+/H (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 9.Piubelli L. Aliverti A. Arakaki A.K. Carrillo N. Ceccarelli E.A. Karplus P.A. Zanetti G. J. Biol. Chem. 2000; 275: 10472-10476Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). Extensive characterization of FNR from different sources has been reported, with its structural arrangement being the prototype for a family of flavin oxidoreductases that interact specifically with either NADP+/H or NAD+/H (8.Hurley J.K. Morales R. Martínez-Júlvez M. Brodie T.B. Medina M. Gómez-Moreno C. Tollin G. Biochim. Biophys. Acta. 2002; 1554: 5-21Crossref PubMed Scopus (75) Google Scholar, 10.Bruns C.M. Karplus P.A. J. Mol. Biol. 1995; 247: 125-145Crossref PubMed Scopus (172) Google Scholar, 11.Serre L. Vellieux F.M.D. Medina M. Gómez-Moreno C. Fontecilla-Camps J.C. Frey M. J. Mol. Biol. 1996; 263: 20-39Crossref PubMed Scopus (134) Google Scholar, 12.Aliverti A. Deng Z. Ravasi D. Piubelli L. Karplus P.A. Zanetti G. J. Biol. Chem. 1998; 273: 34008-34015Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar, 13.Deng Z. Aliverti A. Zanetti G. Arakaki A.K. Ottado J. Orellano E.G. Calcaterra N.B. Ceccarelli E.A. Carrillo N. Karplus A. Nat. Struct. Biol. 1999; 6: 847-853Crossref PubMed Scopus (184) Google Scholar). Previous studies on Anabaena FNR indicated that all of the residues interacting with the 2′-P are not involved to the same extent in determining coenzyme specificity (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Thus, whereas in Anabaena FNR Ser-223 (previous studies on the single S223D mutant) and Tyr-235 (previous studies on the Y235F and Y235A mutants) have been shown to be critical residues in determining NADP+/H orientation and specificity, Arg-224 and Arg-233 only provide secondary interactions (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Moreover, it has also been shown that the determinants for coenzyme specificity are not provided solely by those residues directly interacting with the 2′-P. Thus, in pea FNR, as well as in related flavoenzymes containing an FNR module, it has been shown that a conserved aromatic side chain shields the flavin ring and occupies the putative position of the nicotinamide moiety during ET involving protein substrates. It has been shown that this amino acid in FNR, and in many members of the FNR family, occupies the terminal sequence position and modulates the enzyme NADP+/H binding affinity and selectivity (2.Elmore C.L. Porter T.D. J. Biol. Chem. 2002; 277: 48960-48964Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 7.Döhr O. Paine M.J.I. Friedberg T. Roberts G.C.K. Wolf C.R. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 81-86Crossref PubMed Scopus (73) Google Scholar, 9.Piubelli L. Aliverti A. Arakaki A.K. Carrillo N. Ceccarelli E.A. Karplus P.A. Zanetti G. J. Biol. Chem. 2000; 275: 10472-10476Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar, 14.Adak S. Sharma M. Meade A.L. Stuehr D.J. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 13516-13521Crossref PubMed Scopus (48) Google Scholar, 15.Gutierrez A. Döhr O. Paine M. Wolf C.R. Scrutton N.S. Roberts G.C. Biochemistry. 2000; 39: 15990-15999Crossref PubMed Scopus (50) Google Scholar). ferredoxin-NADP+ reductase FNR in the oxidized state FNR in the reduced state FNR in the semiquinone state ferredoxin Fd in the reduced state 5-deazariboflavin 2′-phosphate electron transfer 2′-phospho-AMP portion of NADP+/H wild type 4-morpholineethanesulfonic acid. Recent studies (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 16.Hermoso J.A. Mayoral T. Faro M. Gómez-Moreno C. Sanz-Aparicio J. Medina M. J. Mol. Biol. 2002; 319: 1133-1142Crossref PubMed Scopus (69) Google Scholar) also point to other regions of the protein that must undergo specific structural rearrangements of the backbone for proper coenzyme binding, which might contribute to the observed specificity. Sequence and structural analysis of different members of the FNR family with affinity for either NADP+/H or NAD+/H suggested that the regions formed by residues 155–161 and 261–265 of Anabaena FNR might be involved in such conformational changes and, therefore, in coenzyme discrimination (see Fig. 1 and Table I in Ref. 1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Previous studies (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar) have shown that replacement by site-directed mutagenesis of the Anabaena FNR Thr-155 (a residue present in all NADP+/H-dependent members) by Gly (a residue present in all NAD+/H-dependent members) produced an increase in the affinity of FNR for NAD+/H while producing an important structural modification in the 261–265 loop. Moreover, the structure recently solved for an Anabaena FNR-NADP+ complex shows a conformation for the 261–265 loop more similar to that shown by the corresponding residues of several NAD+/H-dependent members of the FNR family than the conformation observed for this loop in free FNR (see Fig. 2 in Ref. 16.Hermoso J.A. Mayoral T. Faro M. Gómez-Moreno C. Sanz-Aparicio J. Medina M. J. Mol. Biol. 2002; 319: 1133-1142Crossref PubMed Scopus (69) Google Scholar). In particular, a Pro residue from a "Pro-rich loop" in the NAD+/H-dependent enzymes is systematically placed in the uncomplexed enzyme at the same position that the critical Leu-263 residue occupies in the co-crystallized FNR-NADP+ complex upon NADP+ binding (Fig. 1A, CII) (16.Hermoso J.A. Mayoral T. Faro M. Gómez-Moreno C. Sanz-Aparicio J. Medina M. J. Mol. Biol. 2002; 319: 1133-1142Crossref PubMed Scopus (69) Google Scholar). In addition, structural comparison of the Anabaena FNR-NADP+ complex, obtained after soaking native FNR crystals in an NADP+ solution (Fig. 1A, CI) (11.Serre L. Vellieux F.M.D. Medina M. Gómez-Moreno C. Fontecilla-Camps J.C. Frey M. J. Mol. Biol. 1996; 263: 20-39Crossref PubMed Scopus (134) Google Scholar), with the complex between NADP+ and a pea FNR mutant in which the C-terminal Tyr residue was replaced by Ser (Fig. 1A, CIII) (13.Deng Z. Aliverti A. Zanetti G. Arakaki A.K. Ottado J. Orellano E.G. Calcaterra N.B. Ceccarelli E.A. Carrillo N. Karplus A. Nat. Struct. Biol. 1999; 6: 847-853Crossref PubMed Scopus (184) Google Scholar) allows us to postulate a mechanism of coenzyme recognition and binding involving structural reorganization of the enzyme (16.Hermoso J.A. Mayoral T. Faro M. Gómez-Moreno C. Sanz-Aparicio J. Medina M. J. Mol. Biol. 2002; 319: 1133-1142Crossref PubMed Scopus (69) Google Scholar) (Fig. 1A). However, these structural rearrangements observed in FNR upon coenzyme binding are already observed in the NAD+/H-dependent enzymes in the absence of coenzyme (16.Hermoso J.A. Mayoral T. Faro M. Gómez-Moreno C. Sanz-Aparicio J. Medina M. J. Mol. Biol. 2002; 319: 1133-1142Crossref PubMed Scopus (69) Google Scholar). These observations clearly suggest that the mechanisms for coenzyme recognition and complex reorganization in FNR, and therefore in NADP+/H-dependent members of the FNR family, are different from those of the NAD+/H-dependent enzymes. Thus, in the case of the FNR, it seems that the free enzyme possesses a large cavity to accommodate the 2′-P-AMP moiety of the coenzyme, which upon binding is reorganized in order to perfectly match the charge and shape of the adenine portion of the substrate. However, in the case of the NAD+/H-dependent family members, such a narrow cavity is already preformed in the free enzyme and probably does not need to undergo significant structural rearrangements in order to accommodate the adenine moiety of NAD+/H, as has been described for the case of the phthalate dioxygenase reductase-NADH complex (16.Hermoso J.A. Mayoral T. Faro M. Gómez-Moreno C. Sanz-Aparicio J. Medina M. J. Mol. Biol. 2002; 319: 1133-1142Crossref PubMed Scopus (69) Google Scholar, 17.Correll C.C. Batie C.J. Ballou D.P. Ludwig M.L. Science. 1992; 258: 1604-1610Crossref PubMed Scopus (264) Google Scholar). Therefore, this cavity formation must contribute to the coenzyme specificity in the FNR family, which might be explained as a consequence of the nature of both the residues interacting directly with the 2′-P and the residues shaping the pocket that accommodates the pyrophosphate moiety. In the present study further site-directed mutagenesis studies have been carried out in these Anabaena FNR regions (Fig. 1) in order to clarify the subtle structural features that confer coenzyme specificity. The following groups of mutations on Anabaena FNR have been investigated (Fig. 1B): (a) single and combinational mutations in the pyrophosphate binding region involving residues Thr-155, Ala-160, and Leu-263; (b) combinational mutations at the 2′-P binding region involving residues Ser-223, Arg-224, Arg-233, and Tyr-235; and (c) simultaneous mutations at the pyrophosphate and the 2′-P-binding sites. The obtained binding, kinetic, and structural data have been compared with those of WT FNR, and other FNR mutants previously reported, in order to clarify the subtle structural features that confer coenzyme specificity to this enzyme. Oligonucleotide-directed Mutagenesis—Anabaena FNR site-directed mutants were prepared by two different methods. Some of the mutants were prepared by using the Transformer site-directed mutagenesis kit (Clontech) in combination with suitable synthetic oligonucleotides (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar) and by using as template a construct of the petH gene previously cloned into the expression vector pTrc99a (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). A new construct, pET28-FNR, was also prepared by cloning the petH gene into the NcoI and HindIII sites of the pET-28a(+) vector (Novagen). The mutants prepared by using this latter construct as template were produced using the QuikChange mutagenesis kit (Stratagene) with suitable oligonucleotides. The pTrc99a vectors with the desired mutation were used to transform the Escherichia coli PC strain 0225 (18.Medina M. Martínez-Júlvez M. Hurley J.K. Tollin G. Gómez-Moreno C. Biochemistry. 1998; 37: 2715-2728Crossref PubMed Scopus (90) Google Scholar), whereas the pET28-FNR vectors were transformed into E. coli BL21(DE3) Gold cells (Stratagene). Purification of Anabaena Ferredoxin and FNR Mutants—FNR mutants from pTrc99a vectors were purified from isopropyl-1-thio-β-d-galactopyranoside-induced LB cultures as described previously (18.Medina M. Martínez-Júlvez M. Hurley J.K. Tollin G. Gómez-Moreno C. Biochemistry. 1998; 37: 2715-2728Crossref PubMed Scopus (90) Google Scholar). An analogous protocol was used for mutants of pET28-FNR, but cultures were grown at 30 °C for 24 h, and no isopropyl-1-thio-β-d-galactopyranoside was added. Some of the mutants were not retained by the Cibacron blue gel and were purified by using a fast protein liquid chromatography system from Amersham Biosciences with a Mono-Q column (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Recombinant WT Fd from Anabaena was prepared as described (19.Hurley J.K. Salamon Z. Meyer T.E. Fitch J.C. Cusanovich M.A. Markley J.L. Cheng H. Xia B. Chae Y.K. Medina M. Gómez-Moreno C. Tollin G. Biochemistry. 1993; 32: 9346-9354Crossref PubMed Scopus (101) Google Scholar). UV-visible spectra and SDS-PAGE were used as purity criteria. Spectral Analysis—Ultraviolet/visible spectral analyses were carried out using either a Hewlett-Packard diode array 8452 spectrophotometer, a KONTRON Uvikon 860 spectrophotometer, or a KONTRON Uvikon 942 spectrophotometer. Circular dichroism was carried out on a Jasco 710 spectropolarimeter at room temperature. Far-UV spectra were carried out with either 0.7 μm protein in a 1-cm path length cuvette or 5 μm protein in a 1-mm path length cuvette. Protein concentrations for the near-UV/visible spectra were between 3 and 5 μm in a 1-cm path length cuvette. Photoreduction of different FNR forms was performed at room temperature in an anaerobic cuvette containing 32–65 μm FNR samples and 3 μm dRf in 50 mm Tris/HCl buffer, pH 8, under anaerobic conditions as described previously (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Dissociation constants of the complexes between oxidized FNR mutants and either NADP+ or NAD+ were measured by difference spectroscopy as described previously (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Errors in the estimated Kd values were ±15%. Enzymatic Assays—Diaphorase activity, assayed with 2,6-dichlorophenolindophenol as artificial electron acceptor, was determined for all the FNR mutants in 50 mm Tris/HCl, pH 8.0, as described previously (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). In the case of the diaphorase reactions studied with NADH, high enzyme concentrations (0.5–9 μm) were required in order to detect and follow their activity. Therefore, in some of these cases the coenzyme concentration used was only 100 times higher than that of the corresponding enzyme. This was also the case for all FNR forms containing the S223D mutation when using NADPH, where the enzyme concentration in the cuvettes was 4 μm. When assaying the reaction of the other FNR enzymes with NADPH, enzyme concentrations in the range 3–500 nm were used. Errors in the estimated values of Km and kcat were ±25 and ±10%, respectively. Stopped-flow Kinetic Measurements—Fast electron transfer (ET) processes between NADPH or NADH and the different FNRox mutants were studied by stopped-flow methodology in 50 mm Tris/HCl, pH 8.0, under anaerobic conditions using an Applied Photophysics SX17.MV spectrophotometer interfaced with an Acorn 5000 computer using the SX18.MV software of Applied Photophysics as described previously (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 18.Medina M. Martínez-Júlvez M. Hurley J.K. Tollin G. Gómez-Moreno C. Biochemistry. 1998; 37: 2715-2728Crossref PubMed Scopus (90) Google Scholar). The apparent observed rate constants (kapp) were calculated by fitting the data to a mono- or bi-exponential equation. Errors in their estimated values were ±15%. Final FNR concentrations were kept between 6 and 11 μm. Unless otherwise stated final NADPH concentrations were in the range of 160–200 μm, whereas NADH was used at ∼2.5 mm. The time course of the reactions was followed at 460 nm, although other wavelengths, 340 and 600 nm, were also analyzed. Laser Flash Photolysis Measurements—The laser flash photolysis and the photochemical systems that generate reduced protein in situ were as described previously (18.Medina M. Martínez-Júlvez M. Hurley J.K. Tollin G. Gómez-Moreno C. Biochemistry. 1998; 37: 2715-2728Crossref PubMed Scopus (90) Google Scholar, 20.Hurley J.K. Fillat M.F. Gómez-Moreno C. Tollin G. J. Am. Chem. Soc. 1996; 118: 5526-5531Crossref Scopus (46) Google Scholar, 21.Tollin G. J. Bioenerg. Biomembr. 1995; 27: 303-309Crossref PubMed Scopus (57) Google Scholar). Samples containing 0.1 mm dRf and 1 mm EDTA in 4 mm potassium phosphate buffer, pH 7.0, were deaerated in a long stem 1-cm path length cuvette by bubbling with H2O-saturated argon gas for 1 h. Microliter volumes of concentrated protein were introduced through a rubber septum using a Hamilton syringe under anaerobic conditions. Generally, 4–10 flashes were averaged. Kinetic traces were analyzed using a computer fitting routine (Kinfit, OLIS, Bogart, GA). Experiments were performed at room temperature. Crystal Growth, Data Collection, and Structure Refinement—Crystals of the L263P, T155G/A160T, and T155G/A160T/L263P FNR mutants were grown by the hanging drop method. The 5-μl droplets consisted of 2 μl of 0.75 mm protein solution buffered with 10 mm Tris/HCl, pH 8.0, 1 μl of unbuffered β-octyl glucoside at 5% (w/v), and 2 μl of reservoir solution containing 18–20% (w/v) polyethylene glycol 6000, 20 mm ammonium sulfate, 0.1 m Mes/NaOH, pH 5.0. The droplets were equilibrated against 1 ml of reservoir solution at 20 °C. Under these conditions crystals grew to a maximum size of 0.8 × 0.4 × 0.4 mm within 1–7 days in the presence of phase separation caused by the detergent. Cryoprotectant additives were tested in order to find suitable conditions to use cryotechniques. Finally, crystals were soaked in a solution containing 70–75% of mother liquor and 25–30% glycerol for 1 min. One crystal for each FNR mutant was mounted in a fiber loop and frozen at 100 K with a cryogenic system in a nitrogen stream. X-ray data were collected on a Mar Research (Germany) IP area detector using graphite monochromated CuKα radiation generated by an Enraf-Nonius rotating anode generator to a maximum resolution of 1.6 Å. Crystals belong to the P65 hexagonal space group. The Vm is 3.0 Å3/Da with one FNR molecule in the asymmetric unit and 60% solvent content. All data sets were processed with MOSFLM (22.Leslie A.G.W. Helliwell J.R. Machin P.A. Papiz M.Z. Proceedings of the CCP4 Study Weekend. SERC Daresbury Laboratory, Warrington, UK1987: 39-50Google Scholar) and scaled and reduced with SCALA from the CCP4 package (23.Collaborative Computational Project Number 4 Acta Crystallogr. Sect. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19797) Google Scholar). The L263P, T155G/A160T, and T155G/A160T/L263P structures were solved by molecular replacement using the program AmoRe (24.Navaza J. Acta Cryst. Sect. A. 1994; 50: 157-163Crossref Scopus (5030) Google Scholar) on the basis of the 1.8-Å resolution native FNR model (11.Serre L. Vellieux F.M.D. Medina M. Gómez-Moreno C. Fontecilla-Camps J.C. Frey M. J. Mol. Biol. 1996; 263: 20-39Crossref PubMed Scopus (134) Google Scholar) without the FAD cofactor. An unambiguous single solution for the rotation and translation functions was obtained for all proteins. These solutions were refined by the fast rigid body refinement program FITING (25.Castellano E. Oliva G. Navaza J. J. Appl. Crystallogr. 1992; 25: 281-284Crossref Scopus (57) Google Scholar). The models were subjected to alternate cycles of conjugate gradient refinement with the program X-PLOR (26.Brünger A.T. X-PLOR: A System for X-ray Crystallography and NMR. Version 3.851. Yale University Press, New Haven, CT1993Google Scholar) and manual model building with the software package O (27.Jones T.A. Zou J.Y. Cowan S.W. Kjeldgaard M. Acta Crystallogr. Sect. A. 1991; 47: 110-119Crossref PubMed Scopus (13014) Google Scholar). Finally, water molecules were added. The resulting model was again subjected to more cycles of positional and B factor refinement. The final models comprise residues 9–303 (the first 8 residues were not observed in the electron density map), one FAD moiety, oneSO42- molecule, and solvent molecules. Relevant refinement parameters are presented in Table I. The coordinates and structure factors for all these mutants have been deposited in the Protein Data Bank with accession numbers 1OGJ, 1OGI, and 1H42 for the L263P, T155G/A160T, and T155G/A160T/L263P mutants, respectively.Table IStructure determination statisticsData collectionFNR formT155G/A160T/L263PT155G/A160TL263PTemperature (K)100100100X-ray sourceRotating anodeRotating anodeRotating anodeSpace groupP65P65P65Cell a, b, c (Å)85.81; 85.81; 96.1387.32; 87.32; 96.5587.26; 87.26; 96.69Resolution range (Å)31.94-2.1520.49-1.6320.49-1.63No. unique reflections21,60550,90251,351Completeness of data (%) All data98.998.999.8 Outer shell98.9 (2.00-2.15Å)98.9 (1.73-1.63Å)99.8 (1.73-1.63Å) RsymaRsym = Σhkl Σi |Il - 〈I〉|/Σhkl Σi 〈I〉.0.0950.0670.053Refinements statistics Resolution range (Å)14.86-2.1520.5-1.619.9-1.6 No. protein atoms233623372337 No. hetero atoms585858 No. solvent atoms280358355 RfactorbRfactor = ||Fo| - |Fc||/|Fo| × 100. (%)20.020.620.7 Free Rfactor (%)23.122.122.8 r.m.s.cr.m.s., root mean square. deviationBond lengths (Å)0.0180.0100.010Bond angles (degrees)1.71.41.4a Rsym = Σhkl Σi |Il - 〈I〉|/Σhkl Σi 〈I〉.b Rfactor = ||Fo| - |Fc||/|Fo| × 100.c r.m.s., root mean square. Open table in a new tab Expression and Purification of the Different FNR Mutants— The level of expression in E. coli of all the mutated FNR forms was judged to be similar to that of the recombinant WT protein. All of the mutants were obtained in homogeneous form and in amounts suitable to perform the demanding characterization studies described herein. FNR forms containing the S223D mutation and also those with more than three mutations in the 2′-P interaction region interacted weakly with the Cibacron blue column (1.Medina M. Luquita A. Tejero J. Hermoso J.A. Mayoral T. Sanz-Aparicio J. Grever K. Gómez-Moreno C. J. Biol. Chem. 2001; 276: 11902-11912Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar), thereby requiring the use of a fast protein liquid chromatography Mono-Q column to further purify them. Spectral Properties—No major differences were detected in the UV-visible absorption or in the near-UV or visible CD spectra of any of the FNR forms (not shown). Therefore, no major structural perturbations appear to have been introduced by the mutations in the FAD environment, and the extinction coefficient of Anabaena WT FNR (9.4 mm–1 cm–1 at 458 nm) (28.Pueyo J.J. Gómez-Moreno C. Prep. Biochem. 1991; 21: 191-204PubMed Google Scholar) has been assumed herein for all the FNR mutants. The far-UV CD spectra showed only a slight reduction of the 208 nm peak in two mutants, T155G/R224Q/R233L/Y235F and T155G/S223D/R224Q/R233L/Y235F, indicating that only very subtle modifications, if any, have occurred in the protein folding because of the introduced mutations. Illumination of the FNR forms in the presence