Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Dec 4.
Published in final edited form as: Drug Metab Dispos. 2007 Nov 29;36(3):582–588. doi: 10.1124/dmd.107.018473

Physical Incorporation of NADPH-cytochrome P450 Reductase and Cytochrome P450 into Phospholipid Vesicles using Glycocholate and Biobeads*

James R Reed 1, Lauren M Brignac-Huber 1, Wayne L Backes 1
PMCID: PMC2789009  NIHMSID: NIHMS160830  PMID: 18048487

Abstract

In a previous study from our lab (Drug Metab. Disp. (2006) 34, 660–666), we found several limitations with published methods (cholate gel filtration and cholate dialysis) for the incorporation of cytochromes P450 and P450 reductase into phospholipid vesicles. We found that a significant proportion of reductase was not incorporated in the vesicles when the amount of reductase was equal to or greater than that of CYP2B4 in the systems reconstituted with phosphatidylcholine. Furthermore, implementation of these methods compromised the ability of the CYP2B4 to form a ferrous carbon monoxy complex. In the current study, a comparison of results using the detergent-dialysis method with five similar detergents having the “bile salt” ring structure finds that glycocholate results in the greatest incorporation of reductase and the least loss in carbon monoxy ferrous CYP2B4 complex. The method is further improved by using Biobeads SM-2 to remove detergent instead of the lengthy dialysis procedure or the size exclusion chromatography which significantly dilutes the protein and lipid concentrations of the preparation. The method is shown to be applicable over a range of lipid to CYP2B4 ratios; and by employing assay methods for total lipid, reductase, and CYP2B4, this improved reconstitution method resulted in increased incorporation efficiencies while minimizing the protein degradation inherent with these procedures.

INTRODUCTION

The cytochromes P450 (P450) constitute a superfamily of heme-containing enzymes that display enormous diversity with regards to substrate specificity and catalytic activity (Guengerich, 2001). The enzymes utilize a complicated reaction mechanism that requires the physical coupling of separate proteins to deliver a single electron at two different steps of the catalytic cycle. The diflavin protein, NADPH cytochrome P450 reductase (reductase) can transfer both electrons needed for the catalytic cycle (White and Coon, 1980), whereas the second electron also can be transferred by cytochrome b5 (Schenkman and Jansson, 2003). The mammalian, drug-metabolizing P450s and their associated redox partners are anchored in the lipid membrane of the endoplasmic reticulum.

Lipid plays an important role in catalysis by these enzymes as the in vitro reconstitution of the active monooxygenase system from the purified proteins requires the addition of a lipid milieu, usually dilauroylphosphatidylcholine (DLPC) (Lu et al., 1969;Strobel et al., 1970). The lipid has been postulated to both function as an effector for catalysis (Ingelman-Sundberg, 1977) and provide a framework to allow for the correct orientation and binding of the P450s with the redox partners and substrates (Ingelman-Sundberg et al., 1983;Causey et al., 1990;Taniguchi and Pyerin, 1988). Furthermore, variations in the composition of the lipid milieu that is added to these in vitro reconstituted systems have dramatic effects on the stability and the catalytic activity of the P450 enzymes (Ingelman-Sundberg et al., 1996;Ingelman-Sundberg et al., 1981;Ahn et al., 2005).

The composition of lipids used also affects the manner in which proteins of the P450 system are organized in the membrane. Large proportions of the reductase and P450 enzymes in conventional or standard reconstituted systems using sonicated DLPC have been reported to be peripherally associated with micellar and/or vesicular liposomes (Reed et al., 2006;Autor et al., 1973). However, reconstitution methods that use longer chain phospholipids result in higher amounts of reductase and P450 enzymes being physically incorporated into monolamellar, bilayer phospholipid vesicles (Ingelman-Sundberg and Johansson, 1980;Taniguchi et al., 1979;French et al., 1980). The physical incorporation of reductase and P450 into phospholipid bilayers has been shown to result in preparations with different (higher) levels of catalytic activity (Ingelman-Sundberg et al., 1996;Reed et al., 2006). These commonly implemented methods use sodium cholate to disperse lipids and enzymes followed by a detergent removal step (either dialysis or gel filtration).

The existing methods for these types of systems are limited by several practical considerations. The gel filtration method (Ingelman-Sundberg and Johansson, 1980) results in significant dilution of the protein preparation and samples can only be made one at a time which limits direct comparisons of different conditions. The dialysis procedure (Taniguchi et al., 1979) is limited by the fact that samples must be dialyzed for several days with 3 to 4 changes of large volumes of dialysate. Besides being cumbersome and time consuming, this extended incubation has the potential to compromise the stabilities of the enzymes and lipids in the reconstituted system. Previously (Reed et al., 2006), we observed several limitations with implementation of both the cholate dialysis and cholate gel filtration methods for the vesicular incorporation of reductase and P450. For example, the incorporation of reductase was inefficient at higher reductase:P450 ratios. Although at subsaturating reductase ratios, the incorporation was complete. At a reductase:P450 ratio of 1:1, as much as 40% of the reductase resisted incorporation into the vesicles. In addition, the use of sodium cholate to solubilize the protein and lipid constituents prior to either the dialysis or gel filtration steps resulted in a significant loss of the ferrous CYP2B4 carbon-monoxy complex (even though total heme measured by 417 nm absorbance was unchanged). Thus, the goal of the present study was to improve these aspects of the published methods by comparing a variety of bile salt-derivative detergents using the dialysis procedure for the reconstitution of reductase and CYP2B4 into vesicles of bovine liver phosphatidylcholine (PC (Taniguchi et al., 1979). A method is described in which the detergent is removed from the mixture of detergent, lipid, and proteins by batch incubation with Biobeads SM2 (Holloway, 1973;Levy et al., 1990). This method improves several aspects of the commonly used methods for the physical incorporation of reductase and CYP2B4 into phospholipid vesicles. Finally, the method was tested over a range of phospholipid:CYP2B4 ratios in order to determine its practicality when the protein and lipid components are varied.

MATERIALS AND METHODS

Materials

Bovine liver phosphatidylcholine was purchased from Avanti Polar Lipids, Inc. (Alabaster, AL). Sodium cholate was purchased from Sigma-Aldrich (St. Louis, MO). Benzphetamine was a gift from Upjohn (Kalamazoo, MI). Other detergents used in this study were purchased from Calbiochem (La Jolla, CA). 7-ethoxy-4-trifluoromethylcoumarin and 7-hydroxy-4-trifluoromethylcoumarin were purchased from Molecular Probes (Eugene, OR). All other reagents were obtained from Sigma-Aldrich at the highest available purity.

Enzyme sources

Rabbit NADPH cytochrome P450 reductase was expressed from a recombinant plasmid, containing the wild type cDNA insert in a vector using the T7 promoter, which was provided by Dr. Lucy Waskell (University of Michigan) and has been described previously (Kelley et al., 2005). Recombinant CYP2B4 was expressed and purified from E. coli as described previously (Kelley et al., 2005). P450 levels were determined by measuring the carbon monoxy-ferrous complex (Omura and Sato, 1964).

Preparation of lipid vesicles by the detergent-dialysis method

Lipid vesicles were prepared by the detergent-dialysis method according to published methods (Taniguchi et al., 1979;Reed et al., 2006). Briefly, bovine liver phosphatidylcholine (approximately 1250 nmol), was dissolved in chloroform, lyophilized overnight, and bath-sonicated until clarified in a 0.1 mL solution of 250 mM of either Hepes or potassium phosphate buffer (pH 7.2) containing 75 mM MgCl2 and 5% (w/v) of one of the detergents listed in Figure 1. The solubilized lipid solution was then mixed with a solution containing the enzymes. It was found that less of the “CO-reactive” CYP2B4 was lost if the detergent solution was added gradually. Thus, the detergent solution was added to 0.4 mL of a solution of the CYP2B4 (2.5 nmol) and the reductase (ranging from 0.5 – 5 nmol) in four equal aliquots (≈ 0.025 mL each). Nitrogen was layered over the solution to prevent oxidation of the phospholipids and the samples were gently inverted to mix the solutions after each addition of the lipid-detergent mixture. The samples were then immediately transferred to 0.5 to 3 mL slide-a-lyzer cassettes (Pierce, Rockville, MD) and were dialyzed at 4°C for 36 hr with three changes (12 hr each) of 3 L of dialysis buffer containing 50 mM potassium phosphate (pH 7.25), 0.1 mM EDTA, and 0.1 mM dithiothreitol.

Figure 1. “Bile salt” structural motif common to all of the detergents compared in this study.

Figure 1

Open in a new tab

Chemical structure of the basic structural unit common to the detergents compared for the effects on the vesicular reconstitution of reductase and P450.

Preparation of lipid vesicles using glycocholate and Biobeads SM-2 for detergent removal

The amount of phosphatidylcholine indicated, usually 1 mg, was dissolved in chloroform and lyophilized overnight. The dried lipid was suspended and added to the CYP2B4 and the reductase as described above in the preparation of reconstituted systems using the cholate dialysis method. After the detergent-lipid solution was added to the solution of the CYP2B4 and the reductase, the mixture incubated at 4°C for 1 hr before adding 0.25 g of Biobeads SM-2 (prepared as described previously (Holloway, 1973)). The sample was rocked for 2 hr 4°C, and the solution and a 0.4 mL volume wash of the beads with 0.05 M Hepes buffer (pH 7.25) were drawn up with a 32 gauge needle and then filtered through a 5 micron syringe filter (Osmonics, Minnetonka, MN). In most cases, the final volume of the reconstituted system recovered from the beads was about 0.65 mL.

When different concentrations of lipid were used in the reconstituted systems, the concentration of detergent needed to solubilize the suspension was adjusted according to an approach described previously (Jackson et al., 1982;Bayerl et al., 1986). Briefly, using the CMC of the detergent and the fact that the solubilization of a membrane occurs when the detergent-binding sites in the liposomes are saturated and the monomeric detergent concentration approaches the CMC (Jackson et al., 1982), we estimated the amount of detergent bound to the lipid (as opposed to monomeric detergent in solution) from the amount of detergent needed to optimize the incorporation of 2.5 nmol of enzyme into the vesicles obtained from 1250 nmol of lipid (the lipid-bound detergent concentration = the optimal detergent concentration – the CMC of the detergent). The lipid-bound detergent to lipid ratio determined under the optimized condition could then be used to calculate the amount of detergent that would be bound to any given amount of lipid, and the amount of detergent needed to solubilize the lipid would be the lipid bound detergent concentration added to the CMC of the detergent. We found that vesicles were not efficiently formed when the concentration of the lipid used in the procedure was less than 1250 μM. The enzymatic activities of the reconstituted systems prepared under these conditions were extremely low (data not shown).

Characterization of vesicular preparations

The percentages of reductase and P450 actually incorporated in the vesicles were determined by injecting a fraction of the vesicular preparations on a Tricorn Superose 6 column run on an Akta FPLC system (Amersham Biosciences). Using this column, sample components with a molecular weight greater than 5,000,000 Da were eluted in the void volume. Proteins that were detected in the void volume were considered to be physically incorporated in the lipid vesicle bilayer preparation (French et al., 1980;Reed et al., 2006)

The amounts of P450 reconstituted into the lipid vesicles were detected from the column eluates by manually monitoring the 417 nm absorbance of the fractions using an extinction coefficient of 0.125 μM−1cm−1 (Estabrook et al., 1971). The percentages of CO-reducible P450 in vesicles were quantified by measuring the P450 concentrations using the CO difference spectra (Omura and Sato, 1964). The recoveries of reductase in lipid vesicles and after detergent treatment were determined by measuring cytochrome c reductase activity of fractions recovered from the Superose 6 size exclusion column (Phillips and Langdon, 1962).

A colorimetric method using ammonium ferrothiocyanate (Stewart, 1980) was adapted to determine the concentration of lipids in the fractions of Superose 6 column eluant. Lipids were extracted from 0.1 mL aliquots of the aqueous fractions using an established method (Bligh and Dyer, 1959). After taking the 0.1 mL aliquot from the column fraction, 1.68 mL of H2O and 2 mL each of methanol and chloroform were added. The mixture was vortexed and centrifuged, and 1.5 mL of the lipid/chloroform layer was transferred to a clean test tube. To this layer, 1.5 mL of an aqueous ammonium ferrothiocyanate solution (prepared by dissolving 0.1 M FeCl36H2O and 0.4 M NH4SCN in water) was added, and the mixture was vortexed for 7 min and centrifuged for another 7 min. The timing of these steps is critical as it was observed that the formation of the colored complex between ammonium ferrothiocyanate and phospholipids was light sensitive in the presence of the residual methanol remaining in the chloroform layer. The lower chloroform layer was removed again, and the absorbance was measured at 488 nm. The concentration of lipid was determined from a standard curve using known amounts of DLPC.

Enzymatic assays

N-demethylation of benzphetamine (BZP) was determined by measuring the fluorescence change from formaldehyde after the reaction with a modified Nash reagent (de Andrade et al., 1999;Hanna et al., 2000). O-Dealkylation of 7-ethoxy-4-trifluoromethylcoumarin (7EFC) was determined by monitoring the formation of the product, 7-hydroxy-4-trifluoromethylcoumarin by fluorescence (excitation - 410 nm, emission – 510 nm), and the amount was quantified by reference to a standard curve with 7-hydroxy-4-trifluoromethylcoumarin (Hanna et al., 2000). O-Dealkylation of 7-pentoxy-resorufin (7PRF) was measured by following the formation of the fluorescent product, 7-hydroxyresorufin (excitation – 559 nm, emission – 585 nm) using a standard curve for quantification (Hanna et al., 2000;Lubet et al., 1985;Hanna et al., 2000).

RESULTS

Comparison of the bile salt-derived detergents on the physical incorporation of reductase and P450 into PC vesicles

Our study compared a wide range of commercially available detergents with structures related to bile salts by preparing vesicles, using each of the bile salt-derived detergents shown in Fig. 1, that contained 1250 nmol of bovine liver PC and 2.5 nmol each of the reductase and CYP2B4. Each system was then characterized, focusing on the fraction of each protein physically incorporated into the vesicles, the loss of native protein during the reconstitution procedure, and the catalytic activity of the vesicular preparations (Fig. 2).

Figure 2. Comparison of vesicular preparations of reductase, CYP2B4, and PC using different “bile salt”-derived detergents.

Figure 2

Open in a new tab

(A) The dialysis method (see Materials and Methods) was used to prepare reconstituted systems with reductase and P450 physically incorporated into PC vesicles by using the sodium salt of each of the detergents identified in Fig. 1. The preparations were compared for the proportions of reductase and P450 physically incorporated in the vesicles as indicated by the proportions of enzyme eluting in the void volume of the Superose 6 column. These systems also were characterized by the proportions of active reductase (cytochrome c activity) and P450 (ferrous CO complex) recovered after detergent treatment. The errors associated with measurements for cholate (n=5) and glycocholate (n=4) represent averages ± standard deviation of the indicated parameter. All other conditions were measured in triplicate. The significance of the difference of an average relative to that observed with cholate is indicated by the * using an unpaired student t-test (P<0.01 ). (B) The catalytic activities as determined by the rates of benzphetamine N-demethylation of the vesicular reconstituted systems prepared using the detergents from Fig. 1. Benzphetamine N-demethylation was measured as described in Materials and Methods using 0.05 μM of P450 (as determined from the ferrous-CO complex) from each of the corresponding vesicular preparations. The values represent the averages ± the standard deviation of triplicate determinations.

One characteristic of the vesicular reconstituted systems that was previously reported was that the physical incorporation of reductase into lipid vesicles was incomplete at equimolar reductase to P450 ratios (Reed et al., 2006). As shown in Fig. 2A, only 50% of the reductase was eluted in the void volume after Sepharose 6B chromatography when cholate was used for vesicle preparation. Similar levels of reductase incorporation were observed with all of the detergents except CHAPS and sodium glycocholate. These two detergents resulted in statistically significant increases in the proportion of reductase eluting in the void volume of the Superose 6 chromatography.

The detergent dialysis reconstitution procedure also led to the loss of a portion of active reductase (Reed et al., 2006). About 80% of the reductase activity was recovered when sodium cholate was used in the incorporation of the proteins into the vesicles (Fig. 2A). Each of the tested detergents produced similar fractional recoveries of active protein with the exception of Big CHAPS and deoxycholate, which decreased active reductase recovery to less than 40%.

The incorporation of CYP2B4 into vesicles (as measured by elution of the protein in the void volume of the Superose 6 column) ranged from 88–98% for each of the detergents used with the exception of Big CHAPS, which resulted in only 80% incorporation (Fig. 2A). Despite the uniform response observed with each detergent on the incorporation of CYP2B4 into the PC vesicles, more variability was observed when P450 stability was examined. As reported previously (Reed et al., 2006), the use of cholate for physical incorporation of P450 into PC membranes resulted in a significant loss in the ability of the CYP2B4 to form the ferrous carbon monoxy-complex. This loss was attributable to the denaturation of a fraction of the CYP2B4 by the detergent. As Fig. 2A shows, only 44% of the CYP2B4 was recovered in a CO-reducible form after treatment with sodium cholate. Similar levels of P450 denaturation (or greater with sodium deoxycholate) were seen with all of the other detergents except sodium glycocholate. This was the only detergent that was effective in both facilitating the incorporation of the reductase into the vesicular lipid fraction and minimizing the damage to the CYP2B4. More specifically, the glycocholate procedure generated vesicles in which 85% of the reductase and 98% of the CYP2B4 were physically incorporated into the lipid (at 1:1 reductase:CYP2B4). Furthermore, the recovery of both proteins showed the best profile of the detergents tested – 90% and 75% for reductase and CYP2B4, respectively.

Benzphetamine metabolism was used to assess the catalytic function of the different reconstituted systems (Fig. 2B). Metabolism of this substrate was expressed as a function of the amount of CO-reducible CYP2B4 present in the vesicles. The highest rates of metabolism were attained using CHAPS and deoxycholate. This is due, in part, to the disparate effects of these detergents on the incorporation of reductase into the lipid vesicles and the preservation of CO-reactive ferrous CYP2B4 (discussed above). Given that the binding of reductase and CYP2B4 is characterized as a high affinity complex (Reed and Hollenberg, 2003; French et al., 1980; Reed et al., 2006 ), the result of these effects is that the reconstituted systems prepared with these two detergents probably have a saturating ratio of reductase to P450 (the reductase:CYP2B4 was >1.5:1). In contrast, the reconstituted systems with cholate and glycocholate have ratios less than 1. The preparations with both Big CHAPS and taurocholate were not catalytically active in the metabolism of benzphetamine.

FPLC profiles comparing reconstituted systems using glycocholate and cholate

The vesicular preparations using dialysis after treatment with cholate and glycocholate were further compared by the Superose 6 elution profiles (Fig. 3). The absorbance at 257 nm shows that the vast majority of each of the components eluted in the void volume of the column (7–8 mL of eluant) with both detergents. The most salient distinction between the two preparations was seen in the large amount of reductase, indicated by a broad peak from 9 to 13 mL that eluted after the void volume in the profile derived from the cholate-derived preparations (Fig. 3A). In contrast, the reconstituted system derived from the preparation with sodium glycocholate did not show the lower molecular weight complex of reductase (Fig. 3B). This is consistent with the results shown in Fig. 2A where the cholate preparation exhibited a much smaller fractional incorporation of reductase into the void volume of the column.

Figure 3. FPLC elution profiles from the Superose 6 column of reconstituted systems prepared using cholate (A) and glychocholate (B).

Figure 3

Open in a new tab

Preparations using 5 nmol each of CYP2B4 and reductase with 1% sodium cholate and 2500 nmol of PC were prepared by the dialysis method as described in Materials and Methods. The elution profiles of reductase, P450, and lipid were monitored as described in Materials and Methods. The volumes of elution of the total components, as indicated by the 257 nm absorbances, are also shown in the two profiles.

Preparation of vesicular reconstituted systems using Biobeads SM-2 beads

After finding that glycocholate was able to significantly circumvent several of the shortcomings associated with cholate, we tested whether the technical impracticalities of the dialysis and gel filtration methods could be improved by the use of Biobeads SM-2 to remove the detergent. Vesicular preparations in which glycocholate was removed by adsorption to Biobeads SM-2 in an overnight incubation were statistically identical to those prepared by removing the detergent using a two day dialysis incubation involving 3 changes of 3 L of dialysate (Fig. 4). More specifically, the reconstituted systems were indistinguishable with respect to the proportions of reductase and CYP2B4 physically incorporated into the PC vesicles; the proportions of active reductase and CYP2B4 recovered after treatment with glycocholate (Fig. 4A); and catalytic activity of the preparations (Fig. 4B).

Figure 4. Comparison of vesicular reconstituted systems prepared using Biobeads SM-2 with dialysis.

Figure 4

Open in a new tab

Vesicular preparations using 2.5 nmol each of reductase and P450, 1250 μM PC, and 1% sodium glycocholate were prepared using either Biobeads SM-2 or dialysis as described in Materials and Methods. (A) The preparations were compared for the proportions of reductase and P450 physically incorporated in the vesicles as indicated by the proportions of enzyme eluting in the void volume of the Superose 6 column. The systems also were characterized by the proportions of active reductase (cytochrome c activity) and P450 (ferrous CO complex) recovered after detergent treatment. The errors associated with the measurements represent averages ± standard deviation of the indicated parameter (n=3). (B) Benzphetamine N-demethylation was measured as described in Materials and Methods using 0.05 μM of P450 from each of the corresponding vesicular preparations. The values represent the averages ± the standard deviation of triplicate determinations.

Characterization of vesicular reconstituted systems of reductase and CYP2B4 at different PC to CYP2B4 ratios

For many studies, it is desirable to understand how the reconstitution conditions affect CYP2B4 activities. With this in mind, we have examined the effect of the lipid to CYP2B4 ratio on the ability of the proteins to physically incorporate into vesicles and on the enzymatic activities of the vesicular reconstituted systems. Using a method adapted from a previous study (Jackson et al., 1982) as described in the Materials and Methods, the detergent concentration needed to solubilize different amounts of lipid was optimized.

In the process of optimizing our method, we identified a lower limit of lipid concentration that could be used to obtain catalytically efficient preparations of physically incorporated reductase and CYP2B4. When the lipid concentrations in the reconstituted systems were less than 1250 μM, the incorporation of lipids and enzymes into large-sized aggregates (those eluting in the void volume of the Superose 6 column) was limited. More specifically, the percentages of lipid, reductase, and P450 incorporated in the void volume of the column were 33%, 46%, and 54% when this concentration of lipid was used to prepare the vesicles. Furthermore, reconstituted systems of CYP2B4 and reductase prepared using less than 1250 μM lipid resulted in large-sized aggregates with low catalytic activity (rates of benzphetamine N-demethylation less than 20 nmol/min/nmol CYP2B4 which were less than those catalyzed by conventional reconstituted systems using sonicated DLPC at identical lipid to P450 ratios (data not shown)). Conversely, in all cases where the lipid to P450 ratios were greater than or equal to 500 to 1 (2500 μM), the enzymes were physically associated with large vesicles having catalytic activities that were significantly greater than those prepared with sonicated DLPC (data not shown).

The rates of metabolism of three CYP2B4 substrates were examined in reconstituted systems prepared using different ratios of lipid:CYP2B4 (Figure 5). The optimal lipid:CYP2B4 ratio for each enzymatic reaction was about 500:1. A decrease in the rate of metabolism of each substrate was observed at higher lipid:P450 ratios. Interestingly, the rates of 7EFC and 7PRF metabolism were inhibited to a greater degree than that of benzphetamine at the higher lipid:P450 ratios.

Figure 5. Enzymatic activities of vesicular reconstituted systems prepared with a 1:3 ratio of reductase:CYP2B4 and variable amounts of phosphatidylcholine.

Figure 5

Open in a new tab

The vesicular reconstituted systems were prepared with reductase, 2.5 nmol of CYP2B4, and the indicated amounts of phosphatidylcholine (PC) using sodium glycocholate and Biobeads SM2 as described in Materials and Methods. The reductase:CYP2B4 ratio in the vesicles of PC for all of the preparations was 1:3 as was determined from the cytochrome c reductase activity and the reduced CO-CYP2B4 complex contained in the void volume after injecting 0.1 mL of the reconstituted system on the Superose 6 size exclusion column. The rates of benzphetamine N-demethylation (A); 7EFC O-dealkylation (B); and 7PRF O-dealkylation (C) were determined as described in Materials and Methods. The error bars represent the standard deviation associated with triplicate determinations.

Characterization of vesicular reconstituted systems of reductase and CYP2B4 at different reductase to CYP2B4 ratios

The catalytic activities of vesicular reconstituted systems prepared using the method at a variety of reductase:CYP2B4 ratios also were determined. As discussed in our previous study (Reed et al., 2006), the incorporation of reductase into the vesicular fraction was more efficient at the lower reductase:CYP2B4 ratios (ranged from 100% incorporation at the 1:4 ratio to approximately 50% at a 2:1 ratio), whereas the incorporation of CYP2B4 was greater than 90% at all of the reductase:CYP2B4 ratios.

Figure 6A–6C shows the rates of metabolism of BZP, 7EFC, and 7PRF, respectively, as a function of the vesicular-incorporated reductase concentration. The results indicate that the reductase and CYP2B4 form a high affinity catalytic complex as indicated by the low apparent Km values calculated from the nonlinear regressions (ranging from 14 to 60 nM). The roughly four-fold range in the apparent Km values measured in the presence of the different substrates is consistent with the findings from studies using conventional methods to reconstitute the enzymes with sonicated dilauroylphosphatidylcholine that have found that the affinity of the enzyme complex is dependent on the substrate used to measure the binding (French et al., 1980;Backes and Eyer, 1989;Reed and Hollenberg, 2003). Metabolism of 7PRF was characterized by the lowest apparent Km, whereas the highest value was calculated for 7EFC metabolism. We previously reported that the apparent Km valuescalculated for vesicular reconstituted systems using PC and standard systems prepared with sonicated dilauroylphosphatidylcholine were similar. However, the catalytic activities for the vesicular systems were significantly higher (Reed et al., 2006).

Figure 6. Enzymatic activities of vesicular reconstituted systems prepared with a 500:1 ratio of PC:CYP2B4 and variable amounts of the reductase.

Figure 6

Open in a new tab

The vesicular reconstituted systems were prepared with 1 mg (~1250 nmol) of phosphatidylcholine (PC), 2.5 nmol of CYP2B4, and variable amounts of the reductase by using sodium glycocholate and Biobeads SM2 as described in Materials and Methods. The concentrations of reductase and CYP2B4 in the bilayer vesicles of PC for all of the preparations were determined by measuring the cytochrome c reductase activity and the reduced CO-CYP2B4 complex, respectively in the void volume of the Superose 6 size exclusion column after injecting 0.1 mL of the vesicular reconstituted systems. The rates of benzphetamine N-demethylation (A); 7EFC O-dealkylation (B); and 7PRF O-dealkylation (C) were determined as described in Materials and Methods. All enzyme assays used 0.1 μM CYP2B4, and the ordinate of the plots represent the reductase concentration used in the respective assays. The error bars represent the standard deviation associated with triplicate determinations.

DISCUSSION

This work describes a significant improvement to the method for preparing reconstituted systems of reductase, CYP2B4, and PC in which the enzymes are physically incorporated into a lipid bilayer. We have demonstrated increases in the proportions of both the reductase that is incorporated into vesicles and the active CYP2B4 recovered when sodium glycocholate is substituted for sodium cholate. In addition, the described method using Biobeads SM-2 also eliminates many of the technical difficulties associated with previously published methods (cholate dialysis and cholate gel filtration) in that several different preparations can be prepared simultaneously without prolonged or laborious handling. We have shown that the characteristics of the vesicular preparations using Biobeads and dialysis are statistically identical with regard to enzyme incorporation and activity.

We have also defined a lower limit of the lipid concentration (> 1250 μM) needed for efficient vesicle incorporation of the lipid and proteins using the glycocholate-biobeads method. In our hands, the vesicles prepared under conditions in which the lipid was ≤ 1250 μM had significantly lower percentages (only 30–50%) of enzymes and lipid eluting in the void volume of the superpose 6 column (the criterion used to indicate vesicle-incorporation) and had low catalytic activity. Because the reconstituted systems using low PC concentrations were limited in the formation of large-sized liposomes and had low catalytic activities, it seems the enzymes and lipid were organized in a different structure from that studied in vesicular reconstituted systems using higher lipid concentrations. Previously, it has been shown that unilamellar vesicles merge into multilamellar structures at low lipid concentrations (Akhrem et al., 1982). Thus, it seems likely that the vesicles prepared in our hands at low lipid:protein ratios using either the detergent dialysis or Biobeads methods represent multilamellar vesicles that have low catalytic activity.

Our study has compared a wide range of commercially available detergents with structures related to bile salts. It is possible that deoxycholate and CHAPS may have facilitated the interaction of the reductase and CYP2B4 and in turn stimulated catalysis. The evidence for this assumption resides in the fact that the vesicular preparations derived using these detergents catalyzed extremely high rates of benzphetamine N-demethylation (> 175 nmol/min/nmol). This is significantly higher than the maximal rate of catalysis calculated from reconstituted systems prepared with sodium cholate (Reed et al., 2006). A previous study also has reported that CHAPS is able to activate the CYP2B-related activity in rats (Dutton et al., 1998). However, as mentioned in the results, the higher rates of activity associated with these detergents probably could be attributed to disparate effects on the incorporation of reductase into the phospholipid vesicles and the CYP2B4 stability which resulted in higher reductase:P450 ratios in the reconstituted systems.

Our results showed that the reductase and CYP2B4 formed a high affinity complex during the metabolism of three different substrates, and the catalytic activity of the vesicular reconstituted systems with CYP2B4 decreased at higher lipid:CYP2B4 ratios. The decrease in activity associated with the metabolism of each substrate as a function of the lipid:CYP2B4 is consistent with the reductase and CYP2B4 being incorporated into separate lipid vesicles at the higher ratios, which would hinder the enzymes from physically interacting and transferring the electrons needed for metabolism. Interestingly, the rates of O-dealkylation of 7EFC and 7PRF by the CYP2B4 reconstituted systems dropped to very low levels at a 5000:1 PC:CYP2B4 ratio, whereas that of benzphetamine only decreased gradually after the ratio exceeded 700:1. One possible explanation for the difference observed with BZP may be the evidence that a high proportion of the total CYP2B4-mediated metabolism of this substrate occurs by the peroxidative pathway which circumvents the need for the reductase and P450 to physically interact (Estabrook et al., 1984).

In conclusion, the efficiency of incorporation of reductase and CYP2B4 was examined in reconstituted systems prepared by the detergent-dialysis method as a function of the detergent used. Glycocholate was shown to be superior to the other detergents tested not only in the stability of the proteins, but also in their ability to be physically incorporated into the vesicular membrane. Building on this data, a modified procedure was developed using glycocholate for the initial solubilization of components, followed by replacement of the dialysis step with Biobeads treatment. The characteristics of reconstituted systems prepared by this method are comparable to those of the detergent-dialysis method with regard to efficiency of protein incorporation and catalytic activity. Additionally, the detergent-Biobeads method has the advantage of being more rapid to prepare and allows for the simultaneous preparation of several reconstituted systems under widely ranging conditions, such as variable lipid to protein and reductase to P450 ratios.

Abbreviations

PC

phosphatidylcholine

CD

cholate dialysis method for preparation of reconstituted system

DLPC

dilauroylphosphatidylcholine

reductase

NADPH-cytochrome P450 reductase

P450

cytochrome P450

CYP2B4

cytochrome P450 2B4

GC

sodium glycocholate

BZP

benzphetamine

7EFC

7-ethoxy-4-trifluoromethylcoumarin

7PRF

7-pentoxyresorufin

Footnotes

*

These studies were supported by a US Public Health Services research grant from the National Institute of Environmental Health Sciences, R01-ES004344.

Reference List

  1. Ahn T, Yun CH, Oh DB. Involvement of Nonlamellar-Prone Lipids in the Stability Increase of Human Cytochrome P450 1A2 in Reconstituted Membranes. Biochem. 2005;44:9188–9196. doi: 10.1021/bi050051e. [DOI] [PubMed] [Google Scholar]
  2. Akhrem AA, Andrianov VT, Bokut SB, Luka ZA, Kissel MA, Skornyakova TG. Thermotropic Behavior of Phospholipid Vesicles Reconstituted With Rat Liver Microsomal Cytochrome P450. Biochim Biophys Acta. 1982;692:287–295. doi: 10.1016/0005-2736(82)90533-8. [DOI] [PubMed] [Google Scholar]
  3. Autor AP, Kaschnitz RM, Heidema JK, Coon MJ. Sedimentation and Other Properties of the Reconstituted Liver Microsomal Mixed-Function Oxidase System Containing Cytochrome P-450, Reduced Triphosphopyridine Nucleotide-Cytochrome P-450 Reductase, and Phosphatidylcholine. Mol Pharmacol. 1973;9:93–104. [PubMed] [Google Scholar]
  4. Backes WL, Eyer CS. Cytochrome P-450 LM2 Reduction. Substrate Effects on the Rate of Reductase-LM2 Association. J Biol Chem. 1989;264:6252–6259. [PubMed] [Google Scholar]
  5. Bayerl T, Klose G, Blanck J, Ruckpaul K. Interaction of Nonionic Detergents With Phospholipids in Hepatic Microsomes at Solubilizing Concentrations As Studied by 31P-NMR. Biochim Biophys Acta. 1986;858:285–293. doi: 10.1016/0005-2736(86)90333-0. [DOI] [PubMed] [Google Scholar]
  6. Bligh EG, Dyer WJ. A Rapid Method for Total Lipid Extraction and Purification. Can J Biochem Physiol. 1959;37:911–917. doi: 10.1139/o59-099. [DOI] [PubMed] [Google Scholar]
  7. Causey KM, Eyer CS, Backes WL. Dual Role of Phospholipid in the Reconstitution of Cytochrome P-450 LM2-Dependent Activities. Mol Pharmacol. 1990;38:134–142. [PubMed] [Google Scholar]
  8. de Andrade JB, Andrade MV, Pinheire HLC, Martins RA, Borges EL. Determination of Formaldehyde and Acetaldehyde in Urine by High Performance Liquid Chromatography (HPLC) Am Lab. 1999;31:22–27. [Google Scholar]
  9. Dutton DR, McMillen SK, Parkinson A. Purification of Rat Liver Microsomal Cytochrome P-450b Without the Use of Nonionic Detergent. J Biochem Toxicol. 1998;3:131–145. doi: 10.1002/jbt.2570030207. [DOI] [PubMed] [Google Scholar]
  10. Estabrook RW, Hildebrandt AG, Baron J, Netter KJ, Leibman K. A New Spectral Intermediate Associated With Cytochrome P-450 Function in Liver Microsomes. Biochem Biophys Res Commun. 1971;42:132–139. doi: 10.1016/0006-291x(71)90372-x. [DOI] [PubMed] [Google Scholar]
  11. Estabrook RW, Martin-Wixtrom C, Saeki Y, Renneberg R, Hildebrandt A, Werringloer J. The Peroxidatic Function of Liver Microsomal Cytochrome P-450: Comparison of Hydrogen Peroxide and NADPH-Catalysed N-Demethylation Reactions. Xenobiotica. 1984;14:87–104. doi: 10.3109/00498258409151400. [DOI] [PubMed] [Google Scholar]
  12. French JS, Guengerich FP, Coon MJ. Interactions of Cytochrome P-450, NADPH-Cytochrome P-450 Reductase, Phospholipid, and Substrate in the Reconstituted Liver Microsomal Enzyme System. J Biol Chem. 1980;255:4112–4119. [PubMed] [Google Scholar]
  13. Guengerich FP. Common and Uncommon Cytochrome P450 Reactions Related to Metabolism and Chemical Toxicity. Chem Res Toxicol. 2001;14:611–650. doi: 10.1021/tx0002583. [DOI] [PubMed] [Google Scholar]
  14. Hanna IH, Reed JR, Guengerich FP, Hollenberg PF. Expression of Human Cytochrome P450 2B6 in Escherichia Coli: Characterization of Catalytic Activity and Expression Levels in Human Liver. Arch Biochem Biophys. 2000;376:206–216. doi: 10.1006/abbi.2000.1708. [DOI] [PubMed] [Google Scholar]
  15. Holloway PW. A Simple Procedure for Removal of Triton X-100 From Protein Samples. Anal Biochem. 1973;53:304–308. doi: 10.1016/0003-2697(73)90436-3. [DOI] [PubMed] [Google Scholar]
  16. Ingelman-Sundberg M. Phospholipids and Detergents As Effectors in the Liver Microsomal Hydroxylase System. Biochim Biophys Acta. 1977;488:225–234. doi: 10.1016/0005-2760(77)90180-1. [DOI] [PubMed] [Google Scholar]
  17. Ingelman-Sundberg M, Blanck J, Smettan G, Ruckpaul K. Reduction of Cytochrome P-450 LM2 by NADPH in Reconstituted Phospholipid Vesicles Is Dependent on Membrane Charge. Eur J Biochem. 1983;134:157–162. doi: 10.1111/j.1432-1033.1983.tb07546.x. [DOI] [PubMed] [Google Scholar]
  18. Ingelman-Sundberg M, Haaparanta T, Rydstrom J. Membrane Charge As Effector of Cytochrome P-450LM2 Catalyzed Reactions in Reconstituted Liposomes. Biochem. 1981;20:4100–4106. doi: 10.1021/bi00517a024. [DOI] [PubMed] [Google Scholar]
  19. Ingelman-Sundberg M, Hagbjork AL, Ueng YF, Yamazaki H, Guengerich FP. High Rates of Substrate Hydroxylation by Human Cytochrome P450 3A4 in Reconstituted Membranous Vesicles: Influence of Membrane Charge. Biochem Biophys Res Commun. 1996;221:318–322. doi: 10.1006/bbrc.1996.0593. [DOI] [PubMed] [Google Scholar]
  20. Ingelman-Sundberg M, Johansson I. Catalytic Properties of Purified Forms of Rabbit Liver Microsomal Cytochrome P-450 in Reconstituted Phospholipid Vesicles. Biochem. 1980;19:4004–4011. doi: 10.1021/bi00558a018. [DOI] [PubMed] [Google Scholar]
  21. Jackson ML, Schmidt CF, Lichtenberg D, Litman BJ, Albert AD. Solubilization of Phosphatidylcholine Bilayers by Octyl Glucoside. Biochem. 1982;21:4576–4582. doi: 10.1021/bi00262a010. [DOI] [PubMed] [Google Scholar]
  22. Kelley RW, Reed JR, Backes WL. Effects of Ionic Strength on the Functional Interactions Between CYP2B4 and CYP1A2. Biochem. 2005;44:2632–2641. doi: 10.1021/bi0477900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Levy D, Bluzat A, Seigneuret M, Rigaud JL. A Systematic Study of Liposome and Proteoliposome Reconstitution Involving Bio-Bead-Mediated Triton X-100 Removal. Biochim Biophys Acta. 1990;1025:179–190. doi: 10.1016/0005-2736(90)90096-7. [DOI] [PubMed] [Google Scholar]
  24. Lu AY, Strobel HW, Coon MJ. Hydroxylation of Benzphetamine and Other Drugs by a Solubilized Form of Cytochrome P-450 From Liver Microsomes: Lipid Requirement for Drug Demethylation. Biochem Biophys Res Commun. 1969;36:545–551. doi: 10.1016/0006-291x(69)90339-8. [DOI] [PubMed] [Google Scholar]
  25. Lubet RA, Mayer RT, Cameron JW, Nims RW, Burke MD, Wolff T, Guengerich FP. Dealkylation of Pentoxyresorufin: a Rapid and Sensitive Assay for Measuring Induction of Cytochrome(s) P-450 by Phenobarbital and Other Xenobiotics in the Rat. Arch Biochem Biophys. 1985;238:43–48. doi: 10.1016/0003-9861(85)90138-9. [DOI] [PubMed] [Google Scholar]
  26. Omura T, Sato R. The Carbon Monoxide-Binding Pigment of Liver Microsomes. I. Evidence for Its Hemoprotein Nature. J Biol Chem. 1964;239:2370–2378. [PubMed] [Google Scholar]
  27. Phillips AH, Langdon RG. Hepatic Triphosphopyridine Nucleotide-Cytochrome c Reductase: Isolation, Characterization, and Kinetic Studies. J Biol Chem. 1962;237:2652–2660. [PubMed] [Google Scholar]
  28. Reed JR, Hollenberg PF. Examining the Mechanism of Stimulation of Cytochrome P450 by Cytochrome B5: the Effect of Cytochrome B5 on the Interaction Between Cytochrome P450 2B4 and P450 Reductase. J Inorg Biochem. 2003;97:265–275. doi: 10.1016/s0162-0134(03)00275-7. [DOI] [PubMed] [Google Scholar]
  29. Reed JR, Kelley RW, Backes WL. An Evaluation of Methods for the Reconstitution of Cytochromes P450 and NADPH P450 Reductase into Lipid Vesicles. Drug Metab Dispos. 2006;34:660–666. doi: 10.1124/dmd.105.006825. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Schenkman JB, Jansson I. The Many Roles of Cytochrome B5. Pharmacol Ther. 2003;97:139–152. doi: 10.1016/s0163-7258(02)00327-3. [DOI] [PubMed] [Google Scholar]
  31. Stewart JC. Colorimetric Determination of Phospholipids With Ammonium Ferrothiocyanate. Anal Biochem. 1980;104:10–14. doi: 10.1016/0003-2697(80)90269-9. [DOI] [PubMed] [Google Scholar]
  32. Strobel HW, Lu AYH, Heidema J, Coon MJ. Phosphatidylcholine Requirement in the Enzymatic Reduction of Hemoprotein P-450 and in Fatty Acid, Hydrocarbon, and Drug Hydroxylation. J Biol Chem. 1970;245:4851–4854. [PubMed] [Google Scholar]
  33. Taniguchi H, Imai Y, Iyanagi T, Sato R. Interaction Between NADPH-Cytochrome P-450 Reductase and Cytochrome P-450 in the Membrane of Phosphatidylcholine Vesicles. Biochim Biophys Acta. 1979;550:341–356. doi: 10.1016/0005-2736(79)90220-7. [DOI] [PubMed] [Google Scholar]
  34. Taniguchi H, Pyerin W. Phospholipid Bilayer Membranes Play Decisive Roles in the Cytochrome P-450-Dependent Monooxygenase System. J Cancer Res Clin Oncol. 1988;114:335–340. doi: 10.1007/BF02128175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. White RE, Coon MJ. Oxygen Activation by Cytochrome P-450. Annu Rev Biochem. 1980;49:315–356. doi: 10.1146/annurev.bi.49.070180.001531. [DOI] [PubMed] [Google Scholar]

RESOURCES