Aluminum-substituted heme domain of P450BM-3 (BMP): introducing a heme-derived fluorescent probe for studies of substrate binding and protein-protein interactions in cytochromes P450.

We describe a novel technique for heme removal and replacement in the heme domain of P450BM-3 (BMP). The method was applied to obtain the aluminum-protoporphyrin IX (Al-PP) substituted derivative of BMP (Al-BMP). The overall yield of the purified Al-BMP was about 15% as related to the initial amount of the hemeprotein. Al-BMP possesses extensive fluorescence in the 550-650 nm region with excitation in the porphyrin absorbance bands. The protein was shown to bind substrates of P450BM-3 (palmitic, arachidonic, and cis-parinaric acids) with affinities similar to those of the native enzyme (3-6 µM). However, the substrate-induced changes in fluorescence of Al-PP reveal the existence of a second, low-affinity substrate-binding site, which cannot be detected by the spin shift in the native, heme-containing BMP. Using fluorescence resonance energy transfer, we have demonstrated that Al-BMP forms a complex with the flavoprotein domain of P450BM-3 labeled with 7-ethylamino-3-(4'-maleimidylphenyl)-4-methylcoumarin maleimide, revealing the affinity similar to that of native BMP (Kd = 5 µM at 0.06 M ionic strength). Therefore, aluminum-substituted BMP may serve as a valuable tool in studies on the mechanisms of interactions of P450s with their substrates and protein partners.

A single active-site mutation of P450BM-3 dramatically enhances substrate binding and rate of product formation.

Identifying key structural features of cytochromes P450 is critical in understanding the catalytic mechanism of these important drug-metabolizing enzymes. Cytochrome P450BM-3 (BM-3), a structural and mechanistic P450 model, catalyzes the regio- and stereoselective hydroxylation of fatty acids. Recent work has demonstrated the importance of water in the mechanism of BM-3, and site-specific mutagenesis has helped to elucidate mechanisms of substrate recognition, binding, and product formation. One of the amino acids identified as playing a key role in the active site of BM-3 is alanine 328, which is located in the loop between the K helix and ß 1-4. In the A328V BM-3 mutant, substrate affinity increases 5-10-fold and the turnover number increases 2-8-fold compared to wild-type enzyme. Unlike wild-type enzyme, this mutant is purified from E. coli with endogenous substrate bound due to the higher binding affinity. Close examination of the crystal structures of the substrate-bound native and A328V mutant BMPs indicates that the positioning of the substrate is essentially identical in the two forms of the enzyme, with the two valine methyl groups occupying voids present in the active site of the wild-type substrate-bound structure.

Crystal structure of inhibitor-bound P450BM-3 reveals open conformation of substrate access channel.

P450BM-3 is an extensively studied P450 cytochrome that is naturally fused to a cytochrome P450 reductase domain. Crystal structures of the heme domain of this enzyme have previously generated many insights into features of P450 structure, substrate binding specificity, and conformational changes that occur on substrate binding. Although many P450s are inhibited by imidazole, this compound does not effectively inhibit P450BM-3. Omega-imidazolyl fatty acids have previously been found to be weak inhibitors of the enzyme and show some unusual cooperativity with the substrate lauric acid. We set out to improve the properties of these inhibitors by attaching the omega-imidazolyl fatty acid to the nitrogen of an amino acid group, a tactic that we used previously to increase the potency of substrates. The resulting inhibitors were significantly more potent than their parent compounds lacking the amino acid group. A crystal structure of one of the new inhibitors bound to the heme domain of P450BM-3 reveals that the mode of interaction of the amino acid group with the enzyme is different from that previously observed for acyl amino acid substrates. Further, required movements of residues in the active site to accommodate the imidazole group provide an explanation for the low affinity of imidazole itself. Finally, the previously observed cooperativity with lauric acid is explained by a surprisingly open substrate-access channel lined with hydrophobic residues that could potentially accommodate lauric acid in addition to the inhibitor itself.

Bacillus megaterium CYP102A1 oxidation of acyl homoserine lactones and acyl homoserines.

Quorum sensing, the ability of bacteria to sense their own population density through the synthesis and detection of small molecule signals, has received a great deal of attention in recent years. Acyl homoserine lactones (AHLs) are a major class of quorum sensing signaling molecules. In nature, some bacteria that do not synthesize AHLs themselves have developed the ability to degrade these compounds by cleaving the amide bond or the lactone ring. By inactivating this signal used by competing bacteria, the degrading microbe is believed to gain a competitive advantage. In this work we report that CYP102A1, a widely studied cytochrome P450 from Bacillus megaterium, is capable of very efficient oxidation of AHLs and their lactonolysis products acyl homoserines. The previously known substrates for this enzyme, fatty acids, can also be formed in nature by hydrolysis of the amide of AHLs, so CYP102A1 is capable of inactivating the active parent compound and the products of both known pathways for AHL inactivation observed in nature. AHL oxidation primarily takes place at the omega-1, omega-2, and omega-3 carbons of the acyl chain, similar to this enzyme's well-known activity on fatty acids. Acyl homoserines and their lactones are better substrates for CYP102A1 than fatty acids. Bioassay of the quorum sensing activity of oxidation products reveals that the subterminally hydroxylated AHLs exhibit quorum sensing activity, but are 18-fold less active than the parent compound. In vivo, B. megaterium inactivates AHLs by a CYP102A1 dependent mechanism that must involve additional components that further sequester or metabolize the products, eliminating their quorum sensing activity. Cytochrome P450 oxidation of AHLs represents an important new mechanism of quorum quenching.

Interactions of substrates at the surface of P450s can greatly enhance substrate potency.

Cytochrome P450s are a superfamily of heme containing enzymes that use molecular oxygen and electrons from reduced nicotinamide cofactors to monooxygenate organic substrates. The fatty acid hydroxylase P450BM-3 has been particularly widely studied due to its stability, high activity, similarity to mammalian P450s, and presence of a cytochrome P450 reductase domain that allows the enzyme to directly receive electrons from NADPH without a requirement for additional redox proteins. We previously characterized the substrate N-palmitoylglycine, which found extensive use in studies of P450BM-3 due to its high affinity, high turnover number, and increased solubility as compared to fatty acid substrates. Here, we report that even higher affinity substrates can be designed by acylation of other amino acids, resulting in P450BM-3 substrates with dissociation constants below 100 nM. N-Palmitoyl-l-leucine and N-palmitoyl-l-methionine were found to have the highest affinity, with dissociation constants of less than 8 nM and turnover numbers similar to palmitic acid and N-palmitoylglycine. The interactions of the amino acid side chains with a hydrophobic pocket near R47, as revealed by our crystal structure determination of N-palmitoyl-l-methionine bound to the heme domain of P450BM-3, appears to be responsible for increasing the affinity of substrates. The side chain of R47, previously shown to be important in interactions with negatively charged substrates, does not interact strongly with N-palmitoyl-l-methionine and is found positioned at the enzyme-solvent interface. These are the tightest binding substrates for P450BM-3 reported to date, and the affinity likely approaches the maximum attainable affinity for the binding of substrates of this size to P450BM-3.

Obligatory intermolecular electron-transfer from FAD to FMN in dimeric P450BM-3.

Cytochromes P450 typically catalyze the monooxygenation of hydrophobic compounds resulting in the insertion of one atom of dioxygen into the organic substrate and the reduction of the other oxygen atom to water. The two electrons required for the reaction are normally provided by another redox active protein, for example cytochrome P450 reductase (CPR) in mammalian endoplasmic reticulum membranes. P450BM-3 from Bacillus megaterium is a widely studied P450 cytochrome in which the P450 is fused naturally to a diflavin reductase homologous to CPR. From the original characterization of the enzyme by Fulco's laboratory, the enzyme was shown to have a nonlinear dependence of reaction rate on enzyme concentration. In recent experiments we observed enzyme inactivation upon dilution, and the presence of substrate can diminish this inactivation. We therefore carried out enzyme kinetics, cross-linking experiments, and molecular weight determinations that establish that the enzyme is capable of dimerizing in solution. The dimer is the predominant form at higher concentrations under most conditions and is the only form with significant activity. Further experiments selectively knocking out the activity of individual domains with site-directed mutagenesis and measuring enzyme activity in heterologous dimers establish that the electron-transfer pathway in P450BM-3 passes through both protein molecules in the dimer during a single turnover, traversing from the FAD domain of one molecule into the FMN domain of the other molecule before passing to the heme domain. Analysis of our results combined with other analyses in the literature suggests that the heme domain of either monomer may accept electrons from the reduced FMN domain.

LXRs regulate the balance between fat storage and oxidation.

Despite the well-established role of liver X receptors (LXRs) in regulating cholesterol homeostasis, their contribution to lipid homeostasis remains unclear. Here we show that LXR null mice are defective in hepatic lipid metabolism and are resistant to obesity when challenged with a diet containing both high fat and cholesterol. This phenotype is dependent on the presence of dietary cholesterol and is accompanied by the aberrant production of thyroid hormone in liver. Interestingly, the inability of LXR-/- mice to induce SREBP-1c-dependent lipogenesis does not explain the LXR-/- phenotype, since SREBP-1c null mice are not obesity resistant. Instead, the LXR-/- response is due to abnormal energy dissipation resulting from uncoupled oxidative phosphorylation and ectopic expression of uncoupling proteins in muscle and white adipose. These studies suggest that, by selectively sensing the cholesterol component of a lipid-rich diet, LXRs govern the balance between storage and oxidation of dietary fat.

A new approach to the study of protein-protein interaction by FTIR: complex formation between cytochrome P450BM-3 heme domain and FMN reductase domain.

We describe a new approach to the study of protein-protein interaction using Fourier transform infrared spectroscopy (FTIR). This approach is based on the combination of FTIR technique with both protein titration experiments and the principal component analysis (factor analysis) of the IR absorption spectra in the 1500-1800 cm(-1) region for the protein mixtures. We have applied this approach to the interaction of the heme domain with the FMN domain of bacterial monooxygenase cytochrome P450BM-3 (CYP102A1). The analysis reveals that the first principal component reflects the protein-protein complex formation because the loading factors show a clear systematic dependence on the concentration of the heme domain according to a titration curve with a dissociation constant of approximately 5 microM. The spectrum of the first principal component has been assigned to structural changes in the secondary structure (increase of beta-sheet and alpha-helix and decrease of turn structures), amino acid side chains (protonation of aspartate and C-terminal COO group), and deprotonation of a propionic acid COOD group in the heme.

Sequence alignments, variabilities, and vagaries.

It seems as if the algorithms and weighting matrices for multiple sequence alignments of the highly divergent members of the P450 gene superfamily have advanced to the point that unknown proteins can be aligned to structurally known members with reasonable accuracy. As stated earlier, the alignment tends to break down at gaps in the sequence alignments, but these regions can be improved manually. This type of alignment and analysis is especially useful for extracting and analyzing the various genome databases. Variations of the conservation analysis can be used to identify charged and uncharged residues that may be important in domain/domain interactions with redox partners or effector molecules (e.g., cytochrome b5). From these alignments and with comparative analysis within families and across P450 families, one can readily obtain an estimation of those residues that might be involved in substrate binding, in redox partner interaction, and in the catalytic mechanism.

Probing the interaction of bovine cytochrome P450scc (CYP11A1) with adrenodoxin: evaluating site-directed mutations by molecular modeling.

The present study was undertaken to evaluate the role of positively charged amino acid residues proposed to reside on the proximal surface of bovine cytochrome P450 cholesterol side chain cleavage (P450scc, CYP11A1) and to determine which residues may be involved in protein-protein interactions with the electron carrier adrenodoxin (Adx). In previous studies, nine different lysine residues were identified by chemical and immunological cross-linking experiments as potentially interacting with Adx, while in the present study, two arginine residues have been identified from sequence alignments. From these 11 residues, 13 different P450scc mutants were made of which only seven were able to be expressed and characterized. Each of the seven mutants were evaluated for their ability to bind Adx, to be reduced, and for their enzymatic activity. Among these, K403Q and K405Q showed a consistent decrease in Adx binding, the ability to be reduced by Adx, and enzymatic activity, with K405Q being affected to a much greater extent. More dramatic was the complete loss of Adx binding by R426Q, while still retaining its ability to be chemically reduced and bind carbon monoxide. Independently, a homology model of P450scc was constructed and docked with the structure of Adx. Four potential sites of interaction were identified: P450scc:K403 with Adx:D76, P450scc:K405 with Adx:D72; P450scc:R426 with Adx:E73, and P450scc:K267 with Adx:E47. Thus, the biochemical and molecular modeling studies together support the hypothesis that K267, K403, K405, and R426 participate in the electrostatic interaction of P450scc with Adx.