Short Communication A PROTEOMIC APPROACH TO THE IDENTIFICATION OF CYTOCHROME P450 ISOFORMS IN MALE AND FEMALE RAT LIVER BY NANOSCALE LIQUID CHROMATOGRAPHY-ELECTROSPRAY IONIZATION-TANDEM MASS SPECTROMETRY

Nanoscale reversed-phase liquid chromatography (LC) combined with electrospray ionization-tandem mass spectrometry (ESI-MS/ MS) has been used as a method for the direct identification of multiple cytochrome P450 (P450) isoforms found in male and female rat liver. In this targeted proteomic approach, rat liver microsomes were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by in-gel tryptic digestion of the proteins present in the 48to 62-kDa bands. The resultant peptides were extracted and analyzed by LC-ESI-MS/MS. P450 identifications were made by searching the MS/MS data against a rat protein database containing 21,576 entries including 47 P450s using Sequest software (Thermo Electron, Hemel Hempstead, UK). Twentyfour P450 isoforms from the subfamilies 1A, 2A, 2B, 2C, 2D, 2E, 3A, 4A, 4F, CYP17, and CYP19 were positively identified in rat liver. The cytochromes P450 (P450s) are one of the largest known gene families and carry out a wide range of biological oxidation and reduction processes important in the metabolism of a large number of drugs, xenobiotics, and endogenous compounds. P450 enzymes are characterized into families and subfamilies by their sequence similarities; there are over 280 different families of P450s and currently more than 1925 sequenced and named isoforms (drnelson.utmem.edu/Cytochrome P450.html). In the rat, a species traditionally important in drug development studies, there are about 50 P450 genes as identified from the UniGene database (drnelson.utmem.edu/UNIGENE. RAT.html; Wheeler et al., 2004). Traditional methods for the detection of P450 proteins have relied on immunodetection of protein, activity assays, or on the detection of P450 mRNA (Patterson and Murray, 2002). These techniques have significant limitations. Immunoblotting, although very sensitive, relies on the availability of isoform-specific antibodies, and it is necessary to preselect which P450s are to be analyzed and to identify each isoform in turn. Activity assays that are geared to investigate the activity of a P450 isoform invariably require multiple analysis techniques, and different assays must be developed for different target substrates; even then, they may not be totally isoform-specific. Measurements at the mRNA level are fraught with uncertainty, since the presence and abundance of a particular type of mRNA do not necessarily indicate a similar presence and abundance of the corresponding protein (Anderson and Seilhamer, 1997; Chen et al., 2002; McFadyen et al., 2003). Mass spectrometry is an alternative method for the analysis of expressed proteins, uniquely offering the ability to detect low levels of multiple proteins in a single analytical run. To date, there have been relatively few reports of the analysis of P450 proteins by mass spectrometry; the majority of these have been activation-based, concentrating on the observation of substrates and metabolites involved in P450 reactions rather than the P450s themselves. Matrix-assisted laser desorption/ ionization (MALDI) time-of-flight (TOF) mass spectrometry has been used to determine the molecular masses of closely related CYP2B1 and 2B2 (Lewis et al., 1993). MALDI-TOF analysis of peptides generated by cyanogen bromide cleavage of CYP3A4 (He et al., 1998; Lightning et al., 2000) and trypsin proteolysis of CYP2E1 (Cai and Guengerich, 2001) have been described. With regard to multiple protein identification, two-dimensional gel electrophoresis followed by MALDI-TOF and peptide mass fingerprinting (PMF) has become a principal approach for the proteomic profiling of various in vitro and in vivo biological systems (Henzel et al., 1993). A major drawback of the two-dimensional gel electrophoresis approach is its low performance in the separation of membrane proteins including P450s (Galeva and Alterman, 2002). In recognition of this, SDS-PAGE (onedimensional gel) separation of microsomal proteins followed by MALDI-TOF and PMF was used to separate and analyze the endoplasmic reticulum proteins from rat and rabbit livers. Using this one-dimensional gel separation, up to eight P450s were positively identified (Galeva et al., 2003). However, PMF is most reliable when samples containing only a few proteins are analyzed because the presence of multiple proteins within an SDS-PAGE band may lead to spurious results. This is particularly the case when proteins are present in low abundance, and their tryptic peptide sequence coverage is low (Clauser et al., 1999; Fenyö, 2000; Huang et al., 2002). Liquid chromatography (LC)-electrospray ionization (ESI)-tandem mass spectrometry (MS/MS) has also been applied to the analysis of P450s This work is supported by a grant from the Engineering and Physical Sciences Research Council, UK. Current address for K.J.W.: University of Hull, Cottingham Road, Hull HU16 7RX, UK. 1 Abbreviations used are: P450, cytochrome P450; MALDI, matrix-assisted laser desorption/ionization; TOF, time-of-flight; PMF, peptide mass fingerprinting; PAGE, polyacrylamide gel electrophoresis; LC, liquid chromatography; ESI, electrospray ionization; MS/MS, tandem mass spectrometry; TFA, trifluoroacetic acid. Address correspondence to: Prof. Laurence H. Patterson, Department of Pharmaceutical & Biological Chemistry, The School of Pharmacy, University of London, 29-39 Brunswick Square, London WC1N 1AX. E-mail: [email protected] 0090-9556/04/3204-382–386$20.00 DRUG METABOLISM AND DISPOSITION Vol. 32, No. 4 Copyright © 2004 by The American Society for Pharmacology and Experimental Therapeutics 1288/1140769 DMD 32:382–386, 2004 Printed in U.S.A. 382 at A PE T Jornals on A ril 2, 2017 dm d.aspurnals.org D ow nladed from (Koenigs et al., 1999; Regal et al., 2000). Very recently, Kislinger et al. (2003) described a multidimensional LC-ESI-MS/MS method for the identification of mammalian proteins. They applied this method to analyze liver microsomes from female mice and were able to detect 20 P450 isoforms. The multidimensional LC-ESI-MS/MS or multidimensional protein identification technology method (Link et al., 1999) does not require prefractionation of the protein sample but rather relies on chromatographic separation of the very complex mixture of peptides generated by proteolysis of unseparated protein mixtures. Although the multidimensional protein identification technology approach may hold advantages in the unbiased analysis of protein mixtures, in the current study we wish to target our analysis to the identification of P450 isoforms. In this regard we have used SDSPAGE to separate endoplasmic reticulum proteins into discrete bands according to molecular weight and then selected the bands in the P450 molecular weight range for further analysis. Proteins within bands in the 48to 62-kDa range were digested with trypsin, and the resulting peptides were separated and analyzed by nanoscale LC-ESI-MS/MS. The MS/MS spectra provide amino acid sequence information that, in combination with peptide mass data, provides a secure identification of proteins. Male and female rats were investigated to illustrate the use of this technique in the identification of gender differences in multiple P450 protein expression. Materials and Methods Preparation of Rat Liver Microsomes. Fresh liver (9–10 g) from male and female outbred 10-week-old Wistar rats (Bantin and Kingman, Hull, UK) was washed with cold isotonic saline (0.9% sodium chloride, 4°C) to remove blood. Connective tissue was excised. Liver was homogenized using an Ultra Turrax T25 (Janke and Kunkel, IKA Labortecnik, Staufen, Germany) in 0.01 M Tris-HCl buffer, pH 7.4, containing 0.25 M sucrose, 15% glycerol, and 0.67 mM phenylmethanesulfonyl fluoride. Microsomes were prepared using differential centrifugation as follows: an initial centrifugation at 2400g for 10 min was used to sediment the cell debris, nuclei, and unbroken cells. The supernatant was centrifuged at 12,000g for 20 min at 4°C. Supernatant from this step were centrifuged at 180,000g for 1 h at 4°C. The resultant microsomal pellets were suspended in 0.1 M Tris-HCl, containing 15% glycerol and 1 mM EDTA, pH 7.4, and then recentrifuged at 180,000g for 1 h. The final pellet was resuspended in 0.1 M Tris-HCl, containing 15% glycerol and 1 mM EDTA, pH 7.4, and stored at 80°C. Microsomal protein was determined using the Bradford assay (Bradford, 1976). SDS-PAGE. One-dimensional SDS-PAGE was performed using standard methods on the Hoefer Mighty Small gel system (Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK). Microsome samples were boiled at 95°C for 8 min in a solubilization buffer (2% SDS, 20% glycerol, 0.01% bromphenol blue, and 50 mM Tris-HCl, pH 6.8) and 0.1 M dithiothreitol. Microsomal protein (20 g) was resolved on a 10% acrylamide gel. Gels were stained with 0.2% Coomassie Brilliant Blue R-250 in 30% methanol and 10% acetic acid (1 h), and destained overnight with the same solvent. In-Gel Tryptic Digestion and Peptide Extraction. The molecular weight region on the SDS-PAGE gel between 48 and 62 kDa was divided into five approximately equal bands, and each band was excised with a scalpel. Bands were washed in distilled water until the pH was neutral and completely destained using 50 mM NH4HCO3 in 40% ethanol. Bands were cut into fine pieces to increase the surface area then dried with acetonitrile and then in a SpeedVac (Thermo Savant, Holbrook, NY) for 30 min. Digestion was carried out using sequencing-grade modified trypsin (approximately 75 ng/ l) (Promega, Southampton, UK) in 25 mM NH4HCO3. Sufficient trypsin solution was added to swell the gel pieces, which were kept on ice for 30 min and then covered with 25 mM NH4HCO3 and incubated at 37°C overnight. Peptides FIG. 1. MS/MS spectrum of the tryptic peptide GTAVLTSLTSVLHDSK [M 2H] ion of m/z 815.4, identified to originate from CYP2C12. Shown in the inset is the peptide’s amino acid sequence. y and b ions are formed by peptide bond cleavage with charge retention on the C terminus and N terminus, respectively. 383 PROTEOMIC IDENTIFICATION OF CYTOCHROME P450s IN RAT LIVER at A PE T Jornals on A ril 2, 2017 dm d.aspurnals.org D ow nladed from were extracted from the gel pieces with ultrasonication, using sequential washings with a solution of 5% trifluoroacetic acid (TFA) in 50% acetonitrile. The extracts were combined and dried in a SpeedVac to complete dryness. Samples were stored at 80°C and reconstituted in 0.1% TFA before analysis. LC-ESI-MS/MS. Nanoscale LC was performed using an LC Packings UltiMate capillary high-performance liquid chromatography system with FAMOS autosampler (Dionex, Camberley, Surrey, UK). A separate UltiMate Micro Pump (Dionex) was used as a loading pump. The sample (1 l) was injected via a sample loop (using 0.1% TFA in water as carrier solvent) onto a 1 mm 300 m PepMap C18 guard column (5 m) (LC Packings, Sunnyvale, CA). The sample was washed with 0.1% TFA for 3.5 min on the guard column before being switched onto a 15 cm 75 m PepMap C18 column (3 m) (LC Packings) equilibrated with 95% mobile phase A (5% acetonitrile containing 0.1% formic acid) and 5% mobile phase B (80% acetonitrile containing 0.1% formic acid), at a flow rate of 200 nl/min. Five minutes after sample injection the proportion of mobile phase B was increased linearly to 50% over 30 min and then stepped to 95% and maintained at this level for 10 min (wash phase). The column was then re-equilibrated for 20 min with 95% mobile phase A, 5% mobile phase B. The column effluent was continuously directed into an LCQ mass spectrometer fitted with a nano-ESI source (Thermo Electron, Hemel Hempstead, UK), and spectra were recorded. Mass spectrometer conditions were optimized using in-solution tryptic digests of purified recombinant P450 isoforms 1A2, 2E1, and 3A4, obtained from PanVera Corp. (Madison, WI). ESI was performed under the following conditions: positive ionization mode; spray voltage, 1.8 kV; capillary voltage, 28 V; capillary temperature, 180°C; and no sheath or auxiliary gas was used. Data were collected in the full-scan and data-dependent MS/MS modes; three microscans were performed, with the maximum ion injection time of 200 ms. In the full-scan mode, ions were collected in the m/z range of 400 to 2000. The MS/MS collision energy was set to 35%. Protein Identification. MS/MS spectra were searched using Sequest Browser software (Thermo Electron) (Eng et al., 1994; Yates et al., 1995), against a rat protein database containing 21,576 entries including 47 P450s (Finnigan Xcalibur; Thermo Finnigan, San Jose, CA; revision 1.0 P/N XCALI64012, July 2000). The aim of the Sequest approach is to find the peptide sequence in a database that best explains the fragment ions present in a spectrum. Candidate sequences are found in the database on the basis of intact peptide masses, and the complete or partial spectra expected to result from the fragmentation of these candidate peptides are generated and compared with the experimental spectrum. The final score assigned to each candidate peptide sequence is the Xcorr, a measure of how well the theoretical spectrum cross-correlates to the observed spectrum. Proteins that were matched by two or more peptides with Xcorr values 2.5 were considered conclusively identified, provided that the peptides were unique to that protein in the database (Ducret et al., 1998). Results and Discussion Initially, it was considered necessary to evaluate the performance of the SDS-PAGE LC-ESI-MS/MS procedure for the identification of P450s. This was achieved by loading known amounts of purified recombinant P450 isoforms 1A2, 2E1, and 3A4 onto the SDS-PAGE gel and proceeding through the analytical cycle to identify the proteins. It was found that P450s could be successfully identified down to approximately 1 pmol loaded onto the gel (results not shown). A successful identification was defined as one in which two or more unique peptides were found with Sequest Xcorr values 2.5. Identification of P450s in Rat Liver Microsomes. Five bands of approximately equal size covering the molecular weight range of 48 to 62 kDa were cut out from SDS-PAGE and subjected to in-gel digestion with trypsin. The resultant peptides were extracted and analyzed by LC-ESI-MS/MS. To identify the proteins present, the MS/MS spectra were submitted to the Sequest algorithm. Sequest then identified the tryptic peptides by matching their MS/MS spectra against insilico generated theoretical spectra from the database. This is illustrated in Fig. 1, which shows the MS/MS spectrum of the doubly charged peptide of m/z 815.4. Sequest determined the amino acid sequence of the peptide to be GTAVLTSLTSVLHDSK, from the CYP2C12 protein. Using this approach, a total of 24 P450 proteins, about half of all the known rat P450s, were identified in male and/or female Wistar rat liver as shown in Table 1. Sequence coverage was found to vary from a maximum of 35% to a minimum of 8%. This variation may be due to a number of factors, including the level of protein abundance, the accessibility to trypsin digestion, and peptide ion mass spectrometric sensitivity. Nevertheless, it was possible to differentiate between closely related P450s, e.g., the highly homologous CYP2D2 and 2D3 (see Fig. 2). Previously, the use of MALDI-TOF-based PMF of onedimensional gel tryptic digests of liver microsomes uncovered four P450s (2A1, 2C11, 2D2, 2D5) in uninduced male Sprague-Dawley rats (Galeva et al., 2003). These isoforms were also found among the 24 P450s identified in male and female rat livers using our LC-ESIMS/MS methodology. MALDI-TOF-based PMF of clofibrateinduced rat liver microsomes revealed the presence of CYP2B1, 2B2, 4A1, and 4A3 (Galeva et al., 2003). Of these, CYP4A3 was the only isoform detected in our male and female livers, suggesting that noninduced levels of CYP4A1, 2B1, and 2B2 are low. However, we did detect CYP2B3 in female rat liver. It should be emphasized that although the MALDI-TOF PMF approach is excellent for the identification of separated proteins and simple protein mixtures (Jensen et al., 1997; Zhang and McElvain, 2000), when proteins are present in low abundance or as part of complex mixtures, as may be the case with one-dimensional gel bands, it is necessary to obtain more information (Fenyö, 2000). The most common method to achieve this is by TABLE 1 CYPs identified by MS/MS to be present in rat liver microsomes Other proteins, which were identified with good sequence coverage ( 15%), are: probable protein disulfide isomerase ER-60 precursor (42%), kinesin (38%), aldehyde dehydrogenase, microsomal (28%) rat ATP synthase chain (26%), glutamate dehydrogenase precursor (24%), mitochondrial precursor (22%), microsomal epoxide hydrolase (21%), UDPglucuronosyltransferase 2B12 precursor (16%), UDP-glucuronosyltransferase 2B3 precursor (18%), UDP-glucuronosyltransferase 2B2 precursor (19%), and F1-ATPase subunit. Cytochromes P450 Identified SWISS-PROT Accession Numbers Number of Matched Peptides (percentage of sequence coverage by amino acid count)