Bills

A 31P-nuclear-magnetic-resonance study of NADPH-cytochrome-P-450 reductase and of the Azotobacter flavodoxin/ferredoxin-NADP+ reductase complex

Description
A 31P-nuclear-magnetic-resonance study of NADPH-cytochrome-P-450 reductase and of the Azotobacter flavodoxin/ferredoxin-NADP+ reductase complex
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Eur. J. Biochem. IYO 531 -537 (1990) C FEBS 1990 A and of the zyxwvut zotobacter flavodoxin/ferredoxin NADP zy   reductase complex P-nuclear-magnetic-resonance study of NADPH - ytochrome-P-450 reductase Peter J. M. BONANTS', Franz MULLER', Jacques VERVOORT' and Dale E. EDMONDSON' Department of Biochemistry, Agricultural University, Wageningcn, The Netherlands Emory University School of Medicine, Atlanta, Georgia, USA (Rcceived November 24, 1989/March 13, 1990) EJB 89 1412 'P-nuclear-magnetic-resonance spectroscopy has been employed to probe the structure of the detergent- solubilized form of liver microsomal NADPH ytochrome-P-450 reductase. In addition to the resonances due to the FMN and FAD coenzymes, additional phosphorus resonances are observed and are assigned to the tightly bound adenosine 2'-phosphate (2'-AMP) and to phospholipids. The phospholipid content was found to vary with the preparation; however, the 2'-AMP resonance was observed in all preparations tested. In agreement with published results [Otvos et al. (1986) Biochemistry 25, 7220 2281 for the protease-solubilized enzyme, the addition of Mn(I1) to the oxidized enzyme did not result in any observable line-broadening of the FMN and FAD phosphorus resonances. The phospholipid resonances, however, were extensively broadened and the line width of the phosphorus resonance assigned to the bound 2'-AMP was broadened by 70 Hz. The data show that only the phosphorus moieties of the phospholipids and the 2'-AMP, but not the flavin coenzymes are exposed to the bulk solvent. Removal of the FMN moiety from the enzyme substantially alters the 31P-NMR spectrum as compared with the native enzyme. The 2'-AMP is removed from the enzyme during the FMN-depletion procedure and the pyrophosphate resonances of the bound FAD are significantly altered. Reconstitution of the FMN- depleted protein with FMN results in the restoration of the coenzyme spectral properties. Reduction of FMN to its air-stable paramagnetic semiquinone form results in broadening of the FMN and 2'-AMP resonances in the detergent-solubilized enzyme. In agreement with previous results, FMN semiquinone formation had little or no effect on the line width of the FMN phosphorus resonance for the proteolytically solubilized enzyme. 31P-NMR experiments with Asotobacter flavodoxin semiquinone, both in its free form and in a complex with spinach ferredoxin-NADP' reductase, mimic the differential paramagnetic effects of the flavin semiquinone on the line width of the FMN phosphorus resonance, observed by comparison of the detergent-solubilized and protease-solubilized forms of the reductase. The data demonstrate that assignment of the site of flavin semiquinone formation to a particular flavin coenzyme may not always be possible by 31P-NMR experiments in multi-flavin containing enzymes. NADPH ytochrome-P-450 reductase is a flavoprotein present in the endoplasmic reticulum of liver and certain other organs [l 1 which catalyzes the transfer of electrons to cytochrome P-450 [6 1 in the microsomal hydroxylation of a variety of compounds [lo- 121. The P-450 reductase contains 1 in01 each of FMN and FAD/SO-kDa polypeptide chain [6 , 131. Proteolytic treatment of the detergent-solubilized enzyme results in cleavage of a hydrophobic N-terminal frag- ment of 12 kDa [14, 151 from the enzyme. This protease- treated enzyme still contains both flavins and exhibits NADPH ytochrome-c reductase activity but not cyto- chrome-P-450 reductase activity [16, 171. The two flavin co- enzymes play an important role in the electron-transfer pro- cess. Reducing equivalents are donated from NADPH to the FAD site [9] and are sequentially transferred to FMN. The FMN moiety then serves as the electron donor for cytochrome Correspondence to F. Muller, Sandoz Agro Ltd, Department of Abbreviations. FNR, ferredoxin-NADP+ reductase; 2'-AMP, Enzymes. NADPH -cytochromc-P-450 reductase (EC 1.6.2.4); Toxicology, CH-4002 Basel, Switzerland adenosine 2'-phosphate. ferredoxin-NADP' reductase (EC 1.18.1.2). P-450. During this electron transfer, the FAD and FMN coenzymes function at both semiquinone and hydroquinone redox levels. 31P-NMR spectroscopy is a powerful technique to investi- gate the environments of phosphorus groups in proteins and has been profitably used to study the flavin coenzyme and protein-bound phosphorus residues in Megasphaera elsdenii flavodoxin [18], Aspergillus niger glucose oxidase [19], bovine milk xanthine oxidase [20], and Azotobucter vinelandii flavodoxin [21, 221. Otvos et al. [23] have recently investigated the properties of the bound flavin phosphates in the proteo- lytically solubilized form of the P-450 reductase, by 31P NMR. The work reported here, on the detergent-solubilized form of the reductase, represents an extension of their published studies. In addition, it is of importance to compare the properties of the two forms in as much as the detergent-solubilized en- zyme retains activity in the cytochrome P-450 assay while the proteolytically solubilized form does not [16, 171. As documented in this paper, the most significant differ- ence in the 31P-NMR spectral properties of the two forms of the enzyme is that the FMN phosphorus resonance z s ex- tensively broadened on formation of the air-stable FMN  532 semiquinone in the detergent-solubilized enzyme, but remains relatively unaffected (compared to the diamagnetic oxidized form) on FMN semiquinone formation in the proteolytically treated enzyme. This latter observation is in full agreement with the results reported by Otvos et al. [23]. To investigate this unusual property in more detail, we have also investigated the effect of FMN semiquinone formation on the FMN phos- phorus resonance of zyxwvutsrq zotobactrr flavodoxin, using both the free form and the complex with spinach ferredoxin-NADP' reductase (FNR). Ongoing work in one of our laboratories on the flavodoxin from Azotobacter (strain OP Berkeley) has shown the FMN phosphate resonance to be unaffected on semiquinone formation when the flavodoxin form lacking the 'labile' phosphorus group is examined by NMR. This observation is in direct contrast to previous work published on the form containing the labile phosphate, where semiquinone formation results in extensive line-broadening of the FMN side-chain phosphate resonance [19]. Since the P-450 reductase shows extensive similarity to FNR and to flavodoxin [24], it seemed worthwhile to investigate this system as an aid towards understanding the spectral properties of the two forms of the P-450 reductase. This paper demonstrates subtle changes in the properties of the FMN coenzyme, on comparison of the two forms of the reductase, which may be of significance in further work on the mechanism of electron transfer from the FMN moiety to the heme of cytochrome P-450. Furthermore, conditions are examined to explain why flavin radical formation may not lead to extensive line-broadening of the side-chain phosphate rcsonance, although existing views in the literature would suggest some paramagnetic effect even in an extended side- chain conformation. Preliminary reports of this work have been published elsewhere [25, 261. MATERIALS AND METHODS NADPH zyxwvutsrq   ytochroine-P-450 reductase was purified after detergent-solubilization of liver microsomes of phenobarbital- treated rabbits and from pig liver as described previously [8]. The rabbit liver preparations were homogeneous, as judged by gel electrophoresis run using the Laemmli procedure [27]. Pig liver preparations showed both 60-kDa and 20-kDa pro- tein bands (in addition to the native 80-kDa enzyme band) which arise from proteolysis of the C-terminal region as de- scribed by Haniu et al. [28, 291. All P-450 reductase prep- arations contained stoichiometric quantities of FMN and FAD. Azotobucter flavodoxin was purified according to the method of Hinkson and Bulen [30] as modified from other publications [21]. Spinach FNR was purified from fresh spin- ach (obtained at a local market) as described by Curti and Zanetti [31]. All oxidized and semiquinone flavoenzyme con- centrations were determined from their visible absorption spectra using published absorption coefficients [13, 30, 311. The seiniquinone form of the flavoproteins was obtained by addition of small amounts of a dithionite solution under anaerobic conditions. The reduction was followed spectro- photometrically to ensure complete formation of the semiquinone. The solutions were deoxygenated by flushing with argon for about 10 min. FMN-depleted P-450 reductase was prepared by repetitive ultrafiltration on an Amicon appa- ratus (YM-30 membrane) using 100 mM Tris/HCl, pH 8.4, 20% (by vol.) glycerol, 1 mM dithiothreitol, 0.1 mM EDTA, and 2 M KBr. Reconstitution of the FMN-depleted enzyme with FMN was performed by addition of an excess of FMN in 20 mM potassium phosphate, pH 7.7, containing 10% (by vol.) glycerol, 1 mM dithiothreitol, 0.1 mM EDTA for several minutes at 25°C. Excess flavin and inorganic phosphate were removed by gel filtration. The 80-kDa form of the P-450 reductase was converted to the 68-kDa form by incubation with trypsin for 1 h at 4°C. No cytochrome-c reductase ac- tivity was lost during this treatment. All phosphorus analyses were performed as described by Bartlett [32]. All 31P-NMR spectra on the P-450 reductase were record- ed at 120.8 MHz on a Bruker CXP-300 spectrometer, while the flavodoxin and FNR spectra were recorded at 81 MHz on an IBM/Bruker WP2OOSY spectrometer using 10-mm NMR tubes (Wilmad). The chemical shifts were determined relative to an external standard, 85 phosphoric acid. P-450 re- ductase spectra were obtained at 17°C using 0.5 W broad- band proton decoupling and a 4800 Hz spectral width. Flavodoxin and FNR spectra were obtained at 22°C using 1 W broad-band proton decoupling and a 4800 Hz spectral width. zyxw   flip angle of 30 was used for the acquisition of all spectra. Free induction decays were subjected to a 10 Hz exponential line-broadening for the P-450 reductase spectra and a 1 Hz line-broadening for the flavodoxin and FNR spectra, prior to Fourier transformation. RESULTS 31P-NMR pectral properties of pig and rabbit liver cytochrome-P-450 reductase The 31P-NMR spectrum of the oxidized form of the deter- gent-solubilized reductase from pig liver is shown in Fig. 1 A. More phosphorus resonances are observed than expected based on the flavin coenzymes present and, as will be shown below, are due to the presence of bound phospholipid and bound 2'-AMP. By analogy with previous studies on the FAD-containing enzymes, glucose oxidase [ 191 and xanthine oxidase [20], the two phosphorus resonances at .4 ppm and 1.3 ppm are assigned to the pyrophosphate moiety of FAD. The line widths of the peaks (AV~,~ 100Hz) are consistent with the FAD coenzyme being tightly bound to the protein. The two resonances appearing around Oppm (-0.3 ppm and 0.4ppm) are probably due to the phospholipids bound to the enzyme. This suggestion is sup- ported by the observation that the intensity of the -0.3 ppm resonance increases strongly upon the addition of 0.5 mol phosphatidylcholine/mol enzyme. Absolute assignment of these resonances to phospholipids would require the compari- son of spectral properties of the isolated enzyme before and after delipidation. Unfortunately, despite a number of at- tempts, we were unable to completely delipidate the enzyme under conditions that would not denature it, at least in concen- trations needed for NMR experiments. The relatively sharp. resonance at 2.9 ppin is presently unassigned. This resonance is variable among preparations tested and may be due to some inorganic phosphate that survives after gel chromatography of the preparation in Tris to remove phosphate buffer srcinally used in the purification procedure. The peaks appearing at 1.8 ppm and 4.1 ppm are assigned to bound 2'-AMP and to FMN, respectively. The 2'-AMP resonance is based on the assignment by Otvos et al. [23] for the proteolytically solubilized P-450 reductase and the FMN resonance is assigned by analogy with previous assignments of flavodoxins [18, 21, 22, 331 and on the P-450 reductase [23]. Confirmation of this assignment is provided by 31P-NMR studies of the FMN-depleted P-450 reductase  533 I zyxwvutsrqponmlkj CHEMICAL SHIFT [ppm] Fig. 1. zyxwvutsrqpon 20.8-MHz 31P-NMR pectra qfNADPH-cytochrome-P-450 reductase,frompig liver A, D) orfrom rahhit zyxwvuts iv r (B, C). The enzyme was dissolved in 50 mM Tris/HCI, pH 7.7, containing 20% (by vol.) glycerol, 0.1 mM dithiothreitol. The temperature was 17°C. zyxwvut A) 365 pM oxidized enzyme, 206278 acquisitions; (B) 390 pM oxidized enzyme, 82632 acquisitions; (C) as (B) with 36.5 pM Mn2+, 82632 acquisitions; (D) semiquinone form (93%) of A), 143961 acqui- sitions. The sharp signal appearing at about 5 ppm in the spectra B and D is due to a small amount of free FMN isolated from rabbit liver (see below). The 31P-NMR spectrum of the rabbit liver P-450 reductase (Fig. 1 B) is quite similar to the pig liver enzyme. The rabbit liver enzyme contains a larger amount of phospholipid and a lowered 2'-AMP content as compared to the pig liver enzyme. The line widths and chemical shift values of the phosphorus resonances are quite similar in both enzymes. Previous 31P-NMR studies of flavoenzymes have shown the utility of probing for solvent accessibility of the phos- phorus groups, by examining the effect of addition of para- magnetic Mn(I1) on the line widths of the observed resonances [19, 20, 221. Otvos et al. [23] concluded that the FMN and FAD phosphate groups in the 68-kDa P-450 reductase were not exposed to the solvent bulk since no observable line- broadening of the flavin coenzyme resonances was observed in the presence of Mn(I1). This situation has been found in all flavoenzyme tested to date. The addition of 10 mol Mn(II)/ 100 mol rabbit liver P-450 reductase results in extensive broadening of the phospholipid resonances and extensive n l zyxwv 10 5 0 zyx   10 -15 CHEMICAL SHIFT [ppm] Fig. 2. 120.8-MHz 31 P-NMR spectru of trypsin-treated NADPH- cytorhrome-P-450 reductase from pig liver. The enzyme was dissolved in 50 niM Tris/HCl, pH 7.7, containing 20% (by vol.) glycerol, 0.1 mM dithiothreitol. The temperature was 17°C. The enzyme con- centration was 345 1M. A) Oxidized enzyme, 61 147 acquisitions; (B) semiquinone form (78 ) of A), 99990 acquisitions broadening (line width 80 50 Hz) of the 2'-AMP resonance, but little or no observable effect on the line widths of the FMN and FAD resonances (Fig. 1 C). These data show that the phosphate groups of both the bound FMN and FAD coenzymes are buried in the interior of the protein (isolated as detergent-solubilized P-450 reductase), as observed by Otvos et al. [23] for the proteolytically solubilized preparation. NADPH ytochrome-P-450 reductase forms an air- stable neutral FMN semiquinone [34] which can be obtained on reduction of the protein by one electron under aerobic conditions. Previous 31P-NMR studies on the proteolytically solubilized enzyme [23] showed that the FMN phosphate resonance decreased in intensity by ~40 n FMN semi- quinone formation, relative to the oxidized form. Resonances due to the FAD pyrophosphate moiety are unaffected by FMN radical formation [23] which demonstrates that the FAD side chain is not in close proximity to the FMN ring. The observed decreased in intensity of the FMN resonance on FMN radical formation was suggested to result from the presence of two different conformational environments of FMN in the preparation of the enzyme used [23]. In one state, the FMN phosphate would be in close proximity to the flavin ring and be able to undergo dipolar line-broadening which would follow an zyx  6 distance relationship. In the other sub- state, presumably the conformation about the FMN was such that the side chain would assume an extended conformation away from the ring with little or no line-broadening being observed. To confirm these observations, we treated a prep- aration of the detergent-solubilized pig liver P-450 reductase with trypsin (see Materials and Methods) to convert it to a 68-kDa form similar to that used by Otvos et al. [23]. 31P- NMR spectral data on the oxidized and air-stable semi- quinone forms of this preparation demonstrate the following properties (Fig. 2). The phospholipid resonances are no longer present which shows they are not bound to this form of the enzyme. Formation of the FMN semiquinone does not influ- ence the spectral properties of the FMN resonance in any observable way. There does not appear to be any reduction in  534 Covalent Phosphate FMN zyxwvutsr B Covalent Phosphate I zyxwvutsrqpo   I 6 4 i zyx   CHEMICAL SHIFT ppm) Fig. 3. zyxwvutsrqpon 1-MHz 31P-NMR pectra zyxwvutsrq f Azotobacter Jlavodoxin in the form after rernoval ofthe luhile phosphate zyxwvutsrq esidue. The protein (2 mM) was dissolved in 50 mM Tris/acetate, pH 8.0. A) Oxidized form after 5406 acquisitions; (B) semiquinone form after 3223 acquisitions intensity of the peak as judged from its relative intensity to the FAD and 2'-AMP resonances in the two redox forms of the enzyme (Fig. 2). Thus, qualitatively, these results confirm those reported by Otvos et al. [23], however, the reduction in the FMN resonance intensity on semiquinone formation observed in their preparations is not observed in the prep- aration used in this experiment. It was of interest to determine if this behavior would also occur in the detergent-solubilized form of the enzyme. The 31P-NMR spectrum for the FMN-semiquinone form of the pig-liver P-450 reductase is shown in Fig. 1 D. The spectrum in Fig. 1 D should be compared directly with that in Fig. 1 A. Formation of the FMN semiquinone has little or no effect on the FAD resonances in the detergent-solubilized enzyme as also observed for the protease-solubilized form ([23], Fig. 2 . The resonances due to the FMN and the bound 2'-AMP are broadened considerably on semiquinone formation. The intensities of both lines are substantially decreased to a level much less than the observed alterations in intensity observed for the FMN semiquinone form of the protease-solubilized enzyme. The residual signal occurring for both assigned reso- nances is probably due to the level of proteolysis (approxi- mately 20%) which cannot be avoided with the pig liver en- zyme [28,29]. These data show a marked difference in respect to the NMR properties of the detergent-solubilized form of the enzyme as compared to those of the proteolytically solubilized form. This difference in properties between the two forms suggests that the FMN phosphate might be closer to the flavin ring in the detergent-solubilized form than in the protease- solubilized form. zyx llh I. .,- I I I' I 10 0 -5 -10 CHEMICAL SHIFT ppm) Fig. 4. 81-MHz 31P-NMR spectra of the complex of Azotobacter ,flavodo.xin )+>ith pinach FNR in 50 mM Trislacetate, pH 8.0. The con- centration of each protein was mM. A) Oxidized form after 30850 acquisitions; (B) reduction of the FMN moiety of flavodoxin to its neutral semiquinone by the addition of a stoichiometric amount of sodium dithionite under an argon atmosphere, 30003 acquisitions 31 P-NMR properties of Azotobacter luvodoxin and in a complex with spinuch ferredoxin-NADP' reductase The discovery of a flavin phosphate in a protein whose 31P resonance is unaffected by semiquinone formation is without precedent in the studies of flavoproteins such as flavodoxin [15,19] or glucose oxidase [16]. In the past, it has been assumed that even in an extended conformation, the FMN phosphate would be close enough to the flavin ring to undergo some line-broadening on semiquinone formation as predicted by the Solomon-Bloembergen equation [35,36]. This assumption has been the basis for assignment of the flavin coenzyme in which the radical species resides in multiflavin-containing enzymes. 31P-NMR studies on Azotobacter flavodoxin which dif- fered from the preparation used in earlier studies [21, 221 in that the labile phosphate was no longer present (due to the fact that it dissociates from the protein on prolonged storage [22]) resulted in properties which paralleled those observed for the two forms of the P-450 reductase. The FMN resonance is virtually unaffected on reduction of the FMN moiety to its neutral semiquinone (an increase in line width of approximate- ly 1 Hz) (Fig. 3). This behavior is in marked contrast to pre- vious results [22] on the flavodoxin form in which the labile phosphate is present, and in which FMN semiquinone forma- tion results in extensive line-broadening of the FMN reso- nance such that it can no longer be observed. These data would suggest that the distance from the side-chain 5'-phosphate to the flavin ring is quite different for the two forms of the flavodoxin. In an attempt to mimic the P-450 reductase, Azotobacter flavodoxin was bound to an equivalent molar quantity of the FAD-containing enzyme, spinach FNR. Pre- vious sequence analysis of cytochrome-P-450 reductase dem- onstrated extensive similarity within the FNR and flavodoxin  535 5 10 -15 zyxwvu HEMICAL SHIFT [ppm] zyxwvutsrq ig. 5. 120.8 MHz zyxwvutsrqp 1 P-NMR spectra zyxwvutsrq f zyxwvutsrq he FAD pyrophosphute reso- nances of zyxwvutsrqpon ADPH- cytochrome-P-450 reductuse,from rubbit liver. The enzyme was dissolved in 50 mM Tris/HCl, pH 7.7, containing 20% (by vol.) glycerol, 0.1 mM dithiothreitol. The temperature was zyxwvut 7 C. A) 390 pM oxidized enzyme, 109777 acquisitions; (B) 380 pM FMN- depleted enzyme from A), 109777 acquisitions; (C) 400 pM FMN- reconstituted enzyme from (B), 39291 acquisitions; (D) sample (C), two weeks after reconstitution (enzyme concentration, 400 pM) 89 364 acquisitions sequences [24]. No perturbation of the 31P-NMR resonances assigned to the FAD moiety of FNR or the FMN and phosphodiester moieties of Azotobacter flavodoxin are ob- served on complexation of the two proteins (Fig. 4). It is assumed that the binding of Azotobacter flavodoxin to spinach FNR is of the same magnitude as its binding to A. vnrahilis FNR (Kd = 7 pM) [37]. Reduction of FMN to its semiquinone form now results in extensive line-broadening of the FMN resonance (Fig. 4B) in contrast to the situation observed with the unbound flavodoxin (Fig. 3). As in the case of the P-450 reductase, the FAD resonances of the FNR-flavodoxin com- plex are unaffected by FMN semiquinone formation. These data suggest the distance from the FMN phosphate to the flavin ring is altered on binding the flavodoxin to FNR. The relevance of these findings to the P-450 reductase will be discussed in more detail in the Discussion. Influence of FMN resolution and reconstitution on the P-NMR properties of the P-450 reductase-bound FAD The reversible resolution and reconstitution of the deter- gent-solubilized P-450 reductase reported by Vermillion and Coon [9] provided important insights into the functional as- signments of the FMN and FAD coenzymes, with FAD serv- ing as the site for NADPH oxidation and FMN serving as an electron-donor to the heme of cytochrome P-450. FMN- depleted P-450 reductase was prepared and investigated by 31P-NMR to verify the FMN resonance assignment due to the presence of other resonances from bound phospholipids and 2'-AMP. These experiments confirmed the assignment of the resonance at 4 ppm as srcinating from the FMN phos- phate. An interesting aspect of this study was the influence of FMN removal on the spectral properties of the FAD pyrophosphatc resonance (Fig. 5). As shown in Fig. 5A and B, removal of FMN from the enzyme causes line narrowing which may result from an increased mobility of the pyro- phosphate moiety. It would appear that the conformation of the FAD side chain has been altered such that the P-0-P torsional angles are affected by FMN removal. Reconstitution of the holoprotein by the addition of exogenous FMN (incu- bation at 4 C, 24 h) results in the reappearance of catalytic activity similar to the level found in the native protein and in the re-establishment of the chemical shifts similar to but not identical with the native enzyme (Fig. 5C and D). The line widths of the FAD resonances for the reconstituted enzyme (see above) are somewhat narrower than those of the native enzyme. These observations show that pronounced structural changes occur about the FAD binding site on FMN removal and that these conformational alterations are reversible on FMN reconstitution. These observations should be considered in any future work on the P-450 reductase requiring FMN depletion and reconstitution. DISCUSSION The results of this study both document and extend the 31P-NMR studies published by Otvos and coworkers [23] and bring forth some unexpected aspects in the application of 31P- NMR techniques to flavoenzyme systems. The major differ- ence observed between the protease-solubilized P-450 re- ductase and the detergent-solubilized form is the influence of formation of the air-stable neutral FMN semiquinone on the line width of the phosphorus resonance of the FMN side-chain phosphate monoester. The observation that it is extensively broadened in the detergent-solubilized form but relatively un- affected in the protease-solubilized form suggests the side- chain phosphate to be in closer proximity to the isoalloxazine ring in the former than in the lattcr enzyme preparation. If this premise is accepted, then a reasonable conclusion is that the presence of the hydrophobic tail derived from the N- terminal portion of the molecule and/or its associated phospholipids, alters the conformation about the FMN-bind- ing site. In this respect, it is worth noting that the N-terminal sequence of the P-450 reductase shows a high degree of simi- larity with flavodoxin [24] and therefore would mimic the interaction between the Azotobacter flavodoxin and FNR. The influence of bound phospholipids on protein confor- mation should also be considered. Magdalou et al. [38] have reported that the a-helical content of the P-450 reductase increased from 28 to 41% upon the addition of phospho- lipids to the P-450 reductase, as measured by circular dichroism spectroscopy. Resolution of the relative contri- butions to protein conformation from either phospholipid binding or from the presence of the N-terminal hydrophobic domain must await further detailed structural studies on the P-450 reductase. However the results of Magdalou et al. [38] are corroborated by our analytical ultracentrifugation studies. This data showed that the sedimentation behavior of deter- gent-solubilized P-450 reductase differed in the presence or
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks