Studies on native and reconstituted FMN-depleted NADPH-cytochrome P-450 reductase

The research on flavins and flavoproteins started in 1879 with the discovery of the yellow pigment "lactochrome" in whey by Blyth. Later it turned out that "lactochrome" and other isolated yellow pigments were derivatives of riboflavin (vitamin B2). The first as such recognized flavoprotein was the "old yellow enzyme" NADPH dehydrogenase, which was isolated from yeast in 1932. At the moment about hundred flavoproteins are known, which contain FAD and/or FMN as prosthetic group. The flavin coenzyme can be covalently or noncovalently bound to the enzyme. Their function is different . Some flavoproteins act as dehydrogenases, while others are oxygenases. Willer wrote in 1981 a review on the different flavoproteins present in nature.The enzyme discussed In this thesis, NADPH-cytochrome P-450 reductase, contains two flavins : FAD and FMN, which are both noncovalently bound to the protein.Chapter 1 contains a broad review on this particular flavoprotein and the reader is referred to this for further information. A short survey of the enzyme will be on its place here. NADPH-cytochrome P-450 reductase, containing both FAD and FMN as prosthetic groups, was first discovered by Horecker In 1949, although he was not aware of it. Later other investigators were involved in Isolating the enzyme to homogeneity and proving its function. The NADPH-cytochrome P-450 reductase is a membrane bound flavoprotein, which Is particularly present in the endoplasmatic reticulum of the liver. The enzyme is part of the mixed function oxidase system and transports electrons from NADPH to cytochrome P-450 by means of its flavins . The multi-enzyme system cytochrome P-450 plays an important role in the detoxification of xenobiotica, activation of procarcinogens, the steroid- and fatty acid metabolism etc. The essential reaction is the hydroxylation by cytochrome P-450. The amino acid sequence of the 78 kDa protein has been determined. It turned out that there was some degree of homology between the reductase and some other flavoproteins . The redox potentials of the Individual flavins have been calculated. FMN and FAD could be removed from the enzyme and It was also possible to restore activity after reconstitution of the flavin-depleted enzyme with FMN or FAD. A 68 kDa proteolytic fragment of the native reductase which was obtained after cleavage of a 10 kDa hydrophobic part of the enzyme, was unable to reduce cytochrome P-450. Several studies were conducted to modify the SH groups present in the protein. The interaction of cytochrome P-450 and of cytochrome c with the enzyme was studied. The latest investigations were on cloning the gene of the enzyme and to bring the gene to expression.Chapter 2 deals with a 31P NMR study of the reductase. The 31 P spectrum of the enzyme showed besides the three resonances due to the flavins some other phosphate resonances. VAMP phosphate was present in every preparation in an almost 1:1 ratio. The amount of phospholipids present in the samples varied. It was demonstrated that these phospholipids were of importance for the configuration of the FMN binding site. 31P NMR spectra of the reductase in the various redox states gave some information on how the environment of the phosphates is influenced by the state of reduction. The semiquinone state of the reductase showed some broadening of the FMN-phosphate resonance due to the paramagnetic free electron only when phospholipids are present in the sample. In the absence of phospholipids no broadening has been observed. The function of these phospholipids has to be studied in more detail. Adding the paramagnetic Mn++ ion to the protein showed that the phospholipids are on the surface of the enzyme and that the 2' AMP phosphate is to some extent accessible by the solvent.In Chapter 3 time-resolved fluorescence studies on the NADPH- cytochrome P-450 reductase are reported. Although the flavins in the reductase are slightly fluorescent it was possible to study the time-resolved fluorescence- and anisotropy decays, which are both multi-exponential. It is shown that FMN in the FMN- reconstituted preparation was not bound in the same way as in the native reductase. It was bound much more loosely although the activity was almost completely regained. Because there is a small spectral overlap between the absorption and emission spectra of FMN and FAD, respectively, a certain degree of energy transfer between the two flavins is possible. We used dimers of methyllumiflavin as models. By measuring the steady state fluorescence anisotropy at different excitation wavelength we discovered that at certain excitation wavelength there was no energy transfer. The same was true for the reductase. By measuring the time-resolved fluorescence anisotropy decay at the wavelength where energy transfer occurs the decay could be fitted by three exponentials, while in the case of no energy transfer the decay could be fitted by two exponentials. The third component, a rather fast one, was absent in the former case and could be described to energy transfer. With this fast component it was possible to calculate the distance between the FAD and the FMN in the reductase (about 8-13 angstrom).In Chapter 4 we report on a recombination study of the FMN- depleted reductase with different modified flavins. NADPH- cytochrome P-450 reductase was easily made FMN-depleted and could be reconstituted with FMN to an active enzyme again. We studied if some modified flavins were able to do this as well and how fast the rate of recombination was compared to that of the native FMN. It was shown that an extra phenyl group on the 6,7 position of the isoalloxazine moiety of the FMN was of no concern to the recombination rate and the recovery of the activity. An extra phosphate group on the 31 position of the ribityl chain was disastrous to the recovery of the activity; the 3'5'biphosphate riboflavin did not bind at all on FMN-depleted reductase. Other modified flavins did bind and gave recovery of the activity to different degrees.Chapter 5 describes several procedures to make the FMN-depleted reductase. Further some physical techniques were applied to the native , the FMN-depleted- and the FMN-recombined reductase to see if the procedure changes some physical properties of the enzyme. Because the published procedure to make FMN-depleted reductase was of no use to us (dialysis of the reductase against a buffer containing 2 M KBr), we searched for a better procedure. Binding of the enzyme on a bioaffinity Sepharose column and removal of the FMN by washing with a certain buffer and elution of the FMN- depleted protein, led to loss of FMN and FAD. Therefore another procedure was tested. Dilution In a buffer containing 2 M KBr , concentration of the solution by ultrafiltration and. repeating this several times, led to a sample containing small amounts of FMN. A method Is described to calculate the percentage of FMN-depleted reductase if one starts with a certain amount of enzyme and a certain number of iterations of the procedure. Large amounts of reductase could not be inade FMN-depleted by this procedure. A third procedure (applying the enzym on a Blogel gelfiltration column and eluting the FMN-depleted protein with a buffer containing 2 M KBr) was also not succesfull In obtaining a FMN-free preparation of the enzyme. A combination of the procedures followed by binding of the FMN-depleted enzyme on the 2'5'ADP-Sepharose column, led to a completely FMN-depleted reductase preparation. Certain physical techniques (Circular dichroism, Analytical ultracentrifugation, Timeresolved fluorescence spectroscopy, Steady state fluorescence spectroscopy and 31P Nuclear magnetic resonance) showed that FMN in the FMN-recombined reductase was bound different from the FMN in the native enzyme. Aggregation of the enzyme occured only if the preparation was frozen at -70°C. The binding of the FAD is influenced by the apo-procedure. Also the conformation of the apo-enzyme is influenced slightly by this procedure as shown by CD