NADPH cytochrome P450 reductase

(aka P450 reductase, CPR, POR)

(yes, yes it also has a wikipedia page Cytochrome_P450_reductase)

P450 reductase (EC 1.6.2.4) belongs to a family of flavoproteins utilizing both FAD and FMN as cofactors. These diflavin reductases emerged from the ancestral fusion of a gene coding for a ferredoxin reductase with its NADP(H) and FAD binding domains with a gene coding for a flavodoxin with its FMN domain. This origin of the enzyme was first proposed by Porter and Kasper (1986) based on their analysis of the rat P450 reductase sequence. The fusion is illustrated by the three-dimensional structure of P450 reductase (Wang et al., 1997) where the domains are clearly distinguished.

The architecture of this domain fusion has been found in a handful of other enzymes (Murataliev et al., 2004) and in some cases, further fusion with a P450 gene has led to self-sufficient P450 proteins (Munro et al., 2007).

The insect P450 reductases sequenced to date are clearly orthologous to the mammalian P450 reductases, with an overall amino acid sequence identity of 54% for the house fly P450 reductase, first cloned and sequenced in 1993 (Koener et al., 1993). There is a single P450 reductase gene in insect genomes, although there are other diflavin reductases. The house fly P450 reductase gene codes for a protein of 671 amino acids, and was mapped to chromosome III. The P450 reductases of other insects are very similar to the house fly enzyme- 82 % identity for the Drosophila melanogaster enzyme (Hovemann et al., 1997), 57 % identity for the Bombyx mori enzyme (Horike et al., 2000), and 75% identity for the Anopheles gambiae P450 reductase (Nikou et al., 2003).

Early attempts to purify and characterize the enzyme from microsomes of house fly abdomens were hampered by the facile proteolytic cleavage of the N-terminal portion of the protein. This hydrophobic peptide anchors the reductase in the membrane, and its removal abolishes the ability of the remainder of the protein (“soluble” or “tryptic” reductase) to reduce P450s. The proteolytically-cleaved reductase nonetheless retains the ability to reduce artificial electron acceptors such as cytochrome c, DCPIP or ferricyanide (Hodgson, 1985). Heterologous expression of the cloned P450 reductase has been achieved in E. coli as an N-terminal fusion with bacterial pelB signal sequence (Andersen et al., 1994; McLaughlin et al., 2008) or as a C-terminal 6xHis-tagged protein (Kewpa et al., 2007), and as the native enzyme in the baculovirus expression system (Wen et al., 2003). The Anopheles minimus enzyme is relatively unstable and prone to lose FMN but mutagenesis (L86F and L219F) to the residues found in other insect reductases increases FMN retention (Sarapusit et al., 2008).

A purification scheme (Murataliev et al., 1999) has produced quantities of enzyme sufficient for a detailed catalytic characterization of the enzyme's functioning and of its reconstitution with redox partners.

P450 reductase is an obligatory partner of microsomal P450 enzymes. Antisera to Spodoptera eridania or house fly P450 reductase inhibit all P450-dependent activities tested (Crankshaw et al., 1981, Feyereisen and Vincent, 1984).

(from: Feyereisen and Vincent, 1984)

It follows therefore: (1) that RNAi of P450 reductase will have a similar effect in vivo to anti-reductase antibodies in vitro (2) that such antibodies can distinguish microsomal from mitochondrial activities in vitro (3) and will also distinguish P450 activities from FMO (flavin monooxygenase) activities.

(1) RNAi knockdown of P450 reductase in adult Anopheles gambiae incrases permethrin susceptibility (Lycett et al., 2006), suggesting P450 activity as a major determinant of its toxicity. Not surprisingly, many publications have now confirmed this in a number of arthropods for a number of pesticides.

(2) Immunoinhibition with P450 reductase antibodies serves as a strong indication of microsomal P450 involvement in NADPH-dependent activities, such as ecdysone 20-hydroxylation in the cockroach and in Bombyx mori eggs (Halliday et al., 1986, Horike and Sonobe, 1999, Horike et al., 2000) and (Z)-9-tricosene biosynthesis in the house fly (Reed et al., 1994).

(3) P450 reductase immunoinhibition could serve as a tool to distinguish P450 dependent activities from flavin monooxygenase (FMO) activities in microsomes from arthropod sources. Relatively little is known about arthropod FMOs.

P450 reductase can also transfer electrons to cytochrome b5 and to other microsomal enzymes such as heme oxygenase (Spencer et al.,2008).

In insect microsomes, the ratio of P450 enzymes to P450 reductase is about 6-18 to 1 (Feyereisen et al., 1990). In this ratio, all P450 enzymes are summed, so that the actual ratio of one specific P450 enzyme to P450 reductase is probably smaller. The rate of the overall microsomal P450 reaction (two transfers of one electron) is relatively slow so that dissociation of the P450-P450 reductase complex is possible between the first and the second electron transfer. Indeed, cytochrome b5 can replace P450 reductase for the supply of the second electron in some cases.

The effect of varying P450/P450 reductase ratio on catalytic rates was measured in a reconstituted system for heptachlor epoxidation by CYP6A1. The rate of epoxidation was determined by the concentration of the binary complex of P450 and P450 reductase, with the same high rate being observed in the presence of an excess of either protein. The half-saturating concentration of either protein was about 0.1 μM in the presence of cytochrome b5 (Murataliev et al., 2008). This is in good agreement with the Km of 0.14 and 0.5 μM for P450 reductase in the presence and absence of cytochrome b5 measured previously (Guzov et al., 1996).

Co-infection of Sf9 cells with baculoviruses carrying lepidopteran CYP6B1 and house fly P450 reductase has revealed that highest catalytic activity was achieved at an equivalent, moderate, multiplicities of infection for the two viruses (Wen et al., 2003; Mao et al., 2006b). Higher enzymatic activities of cell lysates towards furanocoumarins was not achieved when either protein was produced in excess. This result can be explained in part by documented limitations of the cell's ability to host, fold and provide cofactors for both P450 and reductase, but it mostly supports the idea than highest activity is achieved for the highest concentration of the binary complex of the two partners.

The insect, mammalian and yeast enzymes are functionally interchangeable in reconstituted systems of the purified proteins or in heterologous expression systems. However, there is little work to document how well a mammalian, yeast or other P450 reductase can support the activity of an insect P450 when compared to the cognate insect P450 reductase.

It would seem that reductase and P450 from the same species would be the optimal match.

In support of this idea is the observation that CYP392A16 of Tetranychus urticae confers resistance to abamectin when expressed transgenically in Drosophila, only when the mite reductase is co-expressed (Riga et al. 2020).

However, baculovirus-expressed CYP6AA3 from Anopheles minimus has a higher activity in benzoyloxyresorufin metabolism when reconstituted with rat reductase than with the mosquito reductase (Surapusit et al., 2013). In that study, the rat reductase had a higher specific activity, so whether the origin of the reductase or its activity was the determinant factor is unclear.

Purified yeast, human, and Arabidopsis (ATR1 and 2) reductases were compared as redox partners of rabbit CYP2B4 in reconstituted systems of 7-ethoxycoumarin deethylase activity. The human enzyme exhibited the highest affinity but supported the lowest kcat whereas the yeast reductase gave the best kcat but with the lowest affinity. ATR1 exhibited both high affinity and efficiency (Louërat-Oriou et al. 1998). There was no simple relation between reductase activities with artificial acceptors ( cyt c, ferricyanide, DCPIP ) and a natural (CYP2B4) acceptor.

P450 reductase accepts two electrons from NADPH, more precisely, it accepts a hydride ion (one proton and two electrons), and donates the two electrons, one at a time, to P450 enzymes. P450 reductase is therefore an enzyme with two substrates: NADPH and the electron acceptor (P450 or artificial acceptor such as cyt c), and two products: NADP+ and the reduced electron acceptor. With two bound flavins and a pathway of electron transfer NADPH > FAD > FMN > P450 (or cyt c), its reduction state during catalysis can theoretically vary between the fully oxidized state (0 electrons) and the fully reduced state (4 electrons).

Studies with the purified recombinant house fly P450 reductase (Murataliev et al., 1999, Murataliev and Feyereisen, 1999, Murataliev and Feyereisen, 2000)(reviewed in Murataliev et al., 2004) have shed light on two questions posed by this electron transfer function: what is the kinetic mechanism of this two-substrate enzyme (Ping-Pong or sequential Bi-Bi)? and what are the respective reduction states of the two flavins during catalysis ?

In the Ping-Pong mechanism, the first product of the reaction must be released before the second substrate binds to the enzyme, and no ternary complex is formed. In sequential Bi-Bi mechanisms both substrates bind to the enzyme to form a ternary complex. Although several kinetic mechanisms have been proposed (Hodgson, 1985), a careful study of the recombinant house fly P450 reductase clearly established a sequential random Bi-Bi mechanism (Murataliev et al., 1999). The great sensitivity of the enzyme to ionic strength hampers the comparison of different studies (Murataliev et al., 2004). Moreover, inhibition of the enzyme by phosphate above physiological (mM) concentrations (Murataliev and Feyereisen, 2000) make studies at high phosphate concentrations (e.g. 0.3 M, Sarapusit et al., 2008, 2010) physiologically irrelevant and result in both excessive Km values for NADPH and incorrect kinetic mechanisms.

The formation of a ternary complex of NADPH, P450 reductase and the electron acceptor suggested a role for reduced nucleotide binding in the catalysis of fast electron transfer. With the house fly CPR, the rate of cytochrome c reduction was shown to equal the rate of hydride ion transfer from the nucleotide donor to FAD (Murataliev et al., 1999). A faster electron transfer rate was observed with NADPH as compared to NADH (Murataliev et al., 1999) and the 2'-phosphate shown to contribute for more than half of the free energy of binding (Murataliev and Feyereisen, 2000). The affinity of the oxidized P450 reductase was 10x higher for NADPH than for NADP+ (Murataliev et al., 1999), and a conformational change induced by NADPH binding and important for fast catalysis was suggested by these studies.

Later work on active and inactive P450 reductases from various sources has indeed provided evidence for large conformational changes in P450 reductase. One would be the originally described “closed” conformation optimized for interflavin exchange, and the other would be a more “open” conformation, better suited for electron transfer to P450 (Aigrain et al., 2009; Ellis et al., 2009; Hamdane et al., 2009).In addition to the open/closed conformational changes of the FAD and FMN domains relative to each other, there is also a postulated rotation of the FMN cofactor around conserved Asp187 (of yeast, Asp205 of housefly) between the buried position and a more exposed position that has been described as a second FMN binding site (Lamb et al., 2006; Ivanov et al., 2010). Whether the reductase works by just flapping domains but also flipping FMN remains conjectural.

(from Ellis et al. 2009)

The state of reduction of the flavins of house fly P450 reductase during catalysis was deduced from kinetic experiments, rates of NADPH oxidation and EPR measurements of flavin semiquinone (free radical) levels. These revealed the existence of a catalytically competent FMN semiquinone, different from the “blue” neutral FMN semiquinone, known as the air-stable semiquinone that is not a catalytically relevant form of the house fly enzyme (Murataliev and Feyereisen, 1999). Furthermore, the detailed studies of house fly P450 reductase led to a proposed catalytic cycle where the reduction state of the enzyme does not exceed 2 electrons, and where an FMN semiquinone, and not an FMN hydroquinone, serves as electron donor to the acceptor P450 or cytochrome c. This “0-2-1-0 cycle” likely represents the general mechanism of P450 reductases, with strong evidence that it operates in P450BM3 and in other P450 reductases as well (Murataliev et al., 2004).