PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Biochemistry. Author manuscript; available in PMC 2010 May 19.
Published in final edited form as:
PMCID: PMC2693120
NIHMSID: NIHMS106404

Steroid and Protein Ligand Binding to Cytochrome P450 46A1 as Assessed by Hydrogen – Deuterium Exchange and Mass Spectrometry

Abstract

Cytochrome P450 46A1 (CYP46A1) is a key enzyme responsible for cholesterol elimination from the brain. This P450 can interact with different steroid substrates and protein redox partners. We utilized hydrogen-deuterium (H-D) exchange mass spectrometry for investigating CYP46A1-ligand interactions. First, we tested the applicability of the H-D exchange methodology and assessed the amide proton exchange in substrate-free and cholesterol sulfate-bound P450. The results showed good correspondence to the available crystal structures and prompted investigation of the CYP46A1 interactions with the two steroid substrates cholesterol and 24S-hydroxycholesterol and the protein redox partner adrenodoxin (Adx). Compared to substrate-free P450, four peptides in cholesterol-bound CYP46A1 (65–80, 109–116, 151–164, and 351–361) and eight peptides in 24S-hydroxycholesterol-bound enzyme (50–64, 65–80, 109–116, 117–125, 129–143, 151–164, 260–270, and 364–373) showed altered deuterium incorporation. Most of these peptides constitute the enzyme active site, whereas the 351–361 peptide is from the region putatively interacting with the redox partner Adx. This also defines the proximal (presumably water) channel that opens in CYP46A1 upon substrate binding. Reciprocal studies of Adx binding to substrate-free and cholesterol sulfate-bound CYP46A1 revealed changes in the deuteration of the Adx-binding site 144–150 and 351–361 peptides, active site 225–239 and 301–313 peptides, and in the 265–276 peptide, whose functional role is not yet known. The data obtained provide structural insights into how substrate and redox partner binding are coordinated and linked to the hydration of the enzyme active site.

The initial reactions in cholesterol biotransformation to bile acids and steroid hormones are catalyzed by a group of hemoproteins called cytochrome P450s (P450s or CYPs)1 (1, 2). The focus of the present work is on CYP46A1, which catalyzes the first step in the major pathway of cholesterol elimination from the brain. CYP46A1 converts cholesterol to 24S-hydroxycholesterol. Unlike cholesterol, 24S-hydroxycholesterol can cross the blood-brain barrier and be transported to the liver for further degradation to bile acids (3, 4). The significance of CYP46A1 may not be limited to its involvement in cholesterol degradation. 24S-Hydroxycholesterol is a potent activator of the LXR receptors in cultured cells and mice (5); therefore, CYP46A1 may play a regulatory role by producing a biologically active product. It is also possible that CYP46A1 may be involved in subsequent metabolism of 24S-hydroxycholesterol. In vitro CYP46A1 was found to further oxidize 24S-hydroxycholesterol to 24,25- and 24,27-dihydroxycholesterols with the catalytic efficiency ~10-fold higher than that of cholesterol hydroxylation (6, 7). CYP46A1 activity in vitro can be reconstituted with either NADPH-cytochrome P450 oxidoreductase (CPR), the redox partner for microsomal P450s (6), or with adrenodoxin reductase (AdR) and adrenodoxin (Adx) constituting the mitochondrial electron transfer chain (Pikuleva I. A., personal communication). Thus, it is possible that in vivo CYP46A1 interacts with different steroids and redox partners. Crystal structures of the substrate-free and cholesterol sulfate-bound CYP46A1 have been recently determined and revealed that the enzyme undergoes significant conformational changes upon substrate binding (7). This conformational flexibility prompted the current study accessing the molecular dynamics of CYP46A1 in solution upon interaction with different ligands, steroid substrates and protein redox partner Adx.

Amide hydrogen-deuterium (H-D) exchange analyzed by mass spectrometry is a powerful tool for the analysis of protein conformation and dynamics in solution (812). Two ionization methods, electrospray ionization (ESI) and matrix-assisted laser desorption ionization (MALDI) are typically used in the H-D exchange experiments. The MALDI-based approach usually has a lower coverage of protein sequence and higher rate of back-exchange than the ESI-based approach but a higher speed of spectral acquisition, more straightforward peak assignment and simpler data analysis. MALDI H-) exchange mass spectrometry forD exchange has been broadly used to study conformational changes in peptides (1315), protein aggregation (16, 17), epitope mapping of a monoclonal antibody (18), protein-membrane interaction (19), thermodynamics of protein-ligand complexes (2023), and dynamics of various proteins in different states (2434). The data obtained by MALDI H-D exchange are in good agreement with the data available from crystal structures (2734). In the present work, we demonstrate that the MALDI H-D exchange is well suited to detect ligand-induced conformational changes in CYP46A1. We were able to identify and compare regions of the P450 affected by the binding of cholesterol sulfate, cholesterol, 24S-hydroxycholesterol and Adx and through these studies enhance our understanding of how substrate binding affects the interaction between the P450 and the redox partner and how the redox partner affects binding of the substrate.

EXPERIMENTAL PROCEDURES

Materials

Deuterium oxide (99.96%) was purchased from Cambridge Isotope Laboratories (Andover, MA), immobilized pepsin from Pierce Chemicals (Rockford, IL), cholesterol and 24S-hydroxycholesterol from Avanti Polar Lipids (Alabaster, AL), and ZipTip C18 tips from Millipore (Bedford, MA). All other chemicals were from Sigma-Aldrich (St. Louis, MO). Prior to digestion, beads of immobilized pepsin were pre-washed twice with 1 mL of cold 0.03% TFA (pH 2.5) and re-suspended in cold 0.03% TFA.

Proteins

The truncated form, (Δ2–50), of recombinant human CYP46A1 was expressed as described (35). To obtain substrate-bound CYP46A1, enzyme purification was carried out in the presence of a selected steroid (35), namely 50 µM cholesterol sulfate, 30 µM cholesterol, or 50 µM 24S-hydroxycholesterol. After the last purification step, CYP46A1 was diluted and concentrated several times to decrease the concentration of NaCl present in the buffer to 25 mM and adjust the ratio of steroid to CYP46A1 to 1:1. Recombinant bovine adrenodoxin was expressed and purified as described (36).

H-D Exchange

To initiate deuterium incorporation into substrate-free and steroid-bound CYP46A1, 5 µL of 40 µM protein (200 pmol) was mixed with 95 µL of the labeling buffer, 50 mM potassium phosphate (pD 7.2, in 95% D2O). At certain time points, 10 µL of solution (20 pmol of CYP46A1) was removed and quenched with 10 µL of chilled 0.45% TFA. The final pH of the quenched sample was 2.5 and the sample was immediately frozen on dry ice/ethanol and stored at −80 °C until further analysis. Before the analysis, the quenched sample (20 µL) was thawed on ice for 5 min, mixed with 4 µL of pre-washed immobilized pepsin slurry and digested for 6 min on ice with occasional mixing. Beads of immobilized pepsin were then removed by centrifugation at 10,000 × g for 20 s and the supernatant was applied to the ZipTip C18 tip. The tip was washed with 0.03%TFA and peptides were eluted with 20% acetonitrile then 50% acetonitrile, each containing 0.03%TFA. Each fraction was mixed 1:1 (v:v) with a chilled matrix solution containing 6 mg/mL α-cyano-4-hydroxycinnamic acid, 40% acetonitrile and 0.03% TFA. The mixture was quickly spotted onto a chilled stainless steel target plate. After 1 min, when peptides-matrix crystals precipitated, extra liquid was absorbed from the edge of each spot using tissues. The plate was immediately loaded into the mass spectrometer and data were acquired without delay. All manipulations beginning from sample thawing to spotting the peptides onto the target plate were carried out at 4 °C in a cold room.

To probe the CYP46A1 dynamics upon Adx binding, 200 pmol of cholesterol sulfate-bound or substrate-free CYP46A1 and 2 nmol (10-fold molar excess) of Adx were incubated in a total volume of 10 µL for 10 min at room temperature. To initiate deuterium incorporation, the reaction mixture was then diluted 20-fold into 50 mM potassium phosphate (pD 7.2, in 95% D2O). Sample collection, quenching, pepsin digestion, ZipTip C18 separation, and spotting onto the target plate were performed exactly the same as described above.

MALDI-TOF MS

The MALDI-TOF mass spectra were acquired using a 4700 Proteomics Analyzer (Applied Biosystems, Framingham, MA). For initial assignment of peptic fragments, a peptic digest of CYP46A1 was prepared using the same digestion and handling conditions that were used for deuterium exchange experiments. The following identification of peptic fragments was performed based on exact mass and MS/MS analysis. MS-mode acquisitions consisted of 2,400 laser shots averaged from 30 sample positions in each spot. In the MS/MS-mode acquisitions, 6,000 laser shots were averaged from 60 sample positions in each spot. The signal intensity threshold was set to a signal-to-noise ratio of 10. The precursor ion selection window was set at a resolution of 150, which is a typical setting for this instrument. Automated acquisition of MS and MS/MS data was controlled by 4000 Series Explorer software 3.0. MS spectra were internally calibrated in a two-point procedure using reference peptides with m/z [M+H]+ values 976.5210 and 1891.0588. These peptides had been previously identified by MS/MS analysis. After calibration, MS spectra were manually analyzed based on the matching the observed masses with the theoretical masses of CYP46A1 peptides generated by FindPept software (http://us.expasy.org/tools/findpept.html) for pepsin digestion (pH>2) with 4 missed cleavages. The observed m/z [M+H]+ values deviated from the theoretical m/z [M+H]+ values by no more than 10 ppm. Each identified peptide was also systematically interrogated by MS/MS analysis, and each MS/MS spectrum was manually interpreted to validate the peptide.

The deuterium incorporation was determined by calculating the centroid of each isotopic peak cluster. The centroids were computed using the DEX software package (37); the centroid is a weighted average of the m/z values of the peaks in the isotopic envelope, with the weights equal to the corresponding peak areas. The deconvolution was performed using the natural isotopic distribution.

The back-exchange was estimated by measuring the deuterium incorporation in a model peptide: KAQYARSVLLEKDAEPDILELATGYR (peptide B, NIST Reference Material 2397). This peptide was deuterated for 3 hours and digested using the same handling conditions that were used for the CYP46A1 H-D exchange experiments. The ratio between deuterium incorporation and the total number of exchangeable backbone amide hydrogens for two peptic fragments of 1028.51 and 1034.59 Da was calculated and represents the back-exchange. The average of three experiments showed that the deuterium loss during the analysis was 48 ± 2% for both peptic fragments. Since all experiments were carried out under identical experimental conditions, the data obtained were not adjusted for the back-exchange (10).

Data Analysis

The full set of data used to determine if a peptide experienced a significant change in deuterium incorporation upon substrate or Adx binding is in the Supporting Information. Significance was not determined by a specific statistical test. Instead, we attempted to apply the same rules, described below, to each determination in an unbiased fashion.

Variation due to the ZipTip C18 separation and sample/matrix spotting sometimes caused a replicate experiment to give a low intensity or obscured mass spectrum for a particular peptide at one or more time points. The relative deuterium incorporation at these time points could either not be determined, or deuterium uptake curve was distorted. Uptake curves with few time points were discarded. Peptides with significant differences between the control and treatment groups were then identified as follows. First, among the three replicate experiments for each peptide, the two replicate experiments with the most similar deuterium uptake curves were selected. Next, the mean difference in deuterium incorporation between the control time points and the treatment time points were plotted for each peptide (Figures S1 and S2 in Supporting Information). Specifically, for a given peptide, the difference between each control/treatment pair in both replicate experiments was determined, and then the mean of these differences was calculated. These plots served as a guide for selecting the peptides that experienced a significant change in deuterium incorporation upon treatment. Generally, peptides with a mean difference greater than 0.5 were selected. However, to make a final decision for each peptide, the deuterium uptake curves for both replicate experiments were compared (Figures S3 to S17 in Supporting Information). Generally, both replicate experiments had to indicate a significant difference between the control and treatment to classify the peptide as experiencing a change in deuterium incorporation upon substrate or Adx binding. The exceptions were peptides 260–270 and 364–373 in 24S-hydroxycholesterol treatment and peptides 301–313 and 351–361 in Adx treatment. These exceptions are discussed in the Supporting Information.

Analysis of the DEX output was performed using the R programming language (38), Microcal Origin 6.0 (OriginLab, Northhampton, MA), and PyMOL (DeLano Scientific LLC, South San Francisco, CA).

RESULTS

Peptide Mapping and Sequence Coverage of CYP46A1

To improve sequence coverage, the peptic peptides were first fractionated on ZipTip C18 tips by using a two-step elution with 20% acetonitrile followed by 50% acetonitrile. Subsequent accurate mass measurements of peptide ions and sequence information from the MS/MS fragmentation made it possible to assign monoisotopic masses to specific CYP46A1 peptides in each fraction. A total of 53 peptide fragments whose isotopic envelopes were well-resolved after the H-D exchange were considered. These peptides cover most secondary structural elements in CYP46A1 and represent 83% of the protein sequence (Figure 1). The regions of CYP46A1 that were lost and not analyzed include residues 81–108 (the β1–2 sheet/B helix), 281–293 (H-I loop), and 406–425 (K” helix) (Figure 1).

Figure 1
Sequence coverage of human CYP46A1 by peptic digestion. The peptides considered suitable for obtaining H-D exchange data are indicated by arrows and cover 83% of the (Δ2–50) CYP46A1 sequence. The secondary structural elements are shown ...

H-D Exchange of Cholesterol Sulfate-Bound CYP46A1

The deuterium incorporation into substrate-bound CYP46A1 was compared to that of the substrate-free CYP46A1. Eleven peptides were found to have a change in the deuterium uptake upon cholesterol sulfate binding (Figure 2). Six of these peptides (65–80, 109–116, 117–125, 225–239, 362–373, 364–373) contain residues (F80, M108, Y109, R110, A111, L112, F121, N227, A367, W368, and F371) that interact with cholesterol sulfate in the crystal structure of cholesterol sulfate-bound CYP46A1(7). The other five peptides (50–64, 129–143, 151–164, 351–361 and 445–455) are away from the active site. All eleven peptides that showed altered H-D exchange in solution represent the regions in CYP46A1 that were affected by cholesterol sulfate binding in the enzyme crystal structure (Figure 3). Also, the largest occurrences of deuteration in solution were observed for the most dynamic elements of the CYP46A1 crystal structure (peptides 50–64, 65–80, 109–116, and 225–239). These results suggested that the MALDI-based H-D exchange provides meaningful data and that the conformational differences between the cholesterol sulfate-bound and substrate-free crystal structures of CYP46A1 (7) indeed reflect the substrate-induced conformational changes occurring in the protein in solution. These studies with cholesterol sulfate established the applicability of the H-D exchange methodology for elucidation of the molecular dynamics of CYP46A1 in solution and gave impetus to investigation of cholesterol and 24S-hydroxycholesterol binding. Crystal structures of cholesterol- and 24S-hydroxycholesterol-bound CYP46A1 are not yet available.

Figure 2
Peptides showing altered H-D exchange upon binding of cholesterol sulfate. The scale of the y axis shows the expected number of deuterated amide groups in each peptide. Adjustment for the back-exchange was not performed; therefore, the difference in deuterium ...
Figure 3
Regions in the CYP46A1 structure showing altered deuterium incorporation upon binding of cholesterol sulfate. Substrate-free CYP46A1 is in orange, cholesterol sulfate-bound is in cyan, cholesterol sulfate is in yellow, and heme is in pink. The 230–239 ...

H-D Exchange of Cholesterol- and 24S-Hydroxycholesterol-Bound CYP46A1

As in the case with cholesterol sulfate, the deuterium incorporation was compared to that of the substrate-free CYP46A1. Only four peptides (65–80, 109–116, 151–164, and 351–361) showed altered H-D exchange upon cholesterol binding (Figure 4). These peptides also exhibited changes in H-D exchange in cholesterol sulfate-bound CYP46A1. Yet in cholesterol-bound CYP46A1 the deuterium incorporation was smaller than that in the cholesterol sulfate-bound enzyme.

Figure 4
Peptides showing altered H-D exchange upon binding of cholesterol. The scale of the y axis shows the expected number of deuterated amide groups in each peptide. Adjustment for the back-exchange was not performed; therefore, the difference in deuterium ...

Interaction with 24S-hydroxycholesterol elicited H-D exchange similar to that observed upon binding of cholesterol sulfate. Eight peptides were identified (Figure 5). Seven of them are the same as those that showed altered deuterium uptake in cholesterol sulfate-bound CYP46A1 (50–64, 65–80, 109–116, 117–125, 129–143, 151–164, and 364–373); however, one peptide (260–270) was different. This peptide may reflect a distinct feature of 24S-hydroxycholesterol binding to CYP46A1. Also, interaction with 24S-hydroxycholesterol did not affect deuteration of the 351–361 peptide, which showed changes in H-D exchange in cholesterol- and cholesterol sulfate-bound CYP46A1.

Figure 5
Peptides showing altered H-D exchange upon binding of 24S-hydroxycholesterol. The scale of the y axis shows the expected number of deuterated amide groups in each peptide. Adjustment for the back-exchange was not performed; therefore, the difference in ...

Figure 2, Figure 4 and Figure 5 show mainly a decrease in H-D exchange suggesting a reduction in the flexibility of the protein backbone and/or shielding of the selected backbone protons by the substrate. Substrate-increased proton exchange of CYP46A1 was rare and observed only for the 364–373 peptide upon 24S-hydroxycholesterol-binding (Figure 5).

H-D Exchange of CYP46A1 Complex with Adx

Adx is a small 13 kDa soluble protein that transfers electrons to CYP46A1 in the in vitro reconstituted system. We used Adx as a model to gain insight into how substrate binding affects interaction of CYP46A1 with the redox partner. Unfortunately, we could not investigate the complex formation of with CPR, the CYP46A1 natural redox partner. The sequence coverage of this complex was low in MALDI-based H-D exchange because both (Δ2–50) CYP46A1 and CPR are high molecular weight proteins (52 kDa and 78 kDa, respectively).

Substrate-free and cholesterol sulfate-bound forms of CYP46A1 were used in the experiments with Adx. The results are summarized in Figure 6. Interaction with Adx affected the deuteration of five peptides in substrate-free CYP46A1 (144–150, 225–239, 265–276, 301–313, and 351–361). Three of these peptides (144–150, 265–276, and 301–313) also showed altered H-D exchange upon Adx binding in cholesterol sulfate-bound CYP46A1. The 225–239 peptide that had the largest decrease in the deuterium incorporation in substrate-free CYP46A1-Adx complex was unaffected in cholesterol sulfate-bound CYP46A1. As for the 351–361 peptide, unlike experiments with substrate-free CYP46A1, we could not find this peptide in cholesterol sulfate-bound CYP46A1. Based on our mutagenesis studies the 351–361 peptide represents a portion of the Adx-binding site in CYP46A1 (Pikuleva I.A., personal communication).

Figure 6
Peptides showing altered H-D exchange upon binding of Adx to substrate-free and cholesterol sulfate-bound CYP46A1. The 225–239 peptide in cholesterol sulfate-bound CYP46A1 is shown for comparison. The scale of the y axis shows the expected number ...

In contrast to substrate binding that mainly led to a decrease in H-D exchange, Adx binding increased the H-D exchange in three of the five peptides identified in substrate-free CYP46A1 and all three peptides in cholesterol sulfate-bound CYP46A1.

DISCUSSION

In the present work we utilized H-D exchange mass spectrometry to study CYP46A1, an important enzyme that controls cholesterol elimination from the brain. We capitalized on the availability of substrate-free and substrate (cholesterol sulfate)-bound crystal structures of CYP46A1 (7) and comparative analysis of these structures indicating that the enzyme active site is conformationally flexible. We first compared deuteration of cholesterol sulfate-bound and substrate-free CYP46A1 in solution and identified peptides showing different deuterium incorporation. These peptides were then mapped on the crystal structures of CYP46A1 and found to coincide with the regions of structural changes occurring upon interaction with cholesterol sulfate (Figure 3). Such good correspondence to the crystallographic data suggested that the MALDI H-D exchange methodology can indeed be used for studies of CYP46A1-ligand interactions and justified subsequent experiments with cholesterol, 24-hydroxycholesterol and Adx. A summary of the peptides exhibiting a change in deuterium uptake upon binding of each of the tested ligands is shown in Table 1.

Table 1
A summary of peptides showing significant change in deuterium uptake upon ligand binding. X indicates significance.

Cholesterol binds weaker to CYP46A1 than cholesterol sulfate, and in vitro has an apparent Kd [asymptotically equal to]10 times higher than cholesterol sulfate (7). The catalytic efficiency of cholesterol hydroxylation by CYP46A1 is also lower in the in vitro reconstituted system (7). Likely because of weaker binding, only four peptides in CYP46A1 were found to have altered deuteration upon interaction with cholesterol (shown in red in Figure 7). Two of these peptides form a part of the CYP46A1 active site. They are located at the entrance (the 65–80 peptide, which forms a part of the helix A) and in the middle of the substrate binding cavity (the 109–116 peptide, which forms a part of the B’ helix) (Figure 7). The other two peptides are outside the active site and seem to be associated with the binding of the CYP46A1 redox partner Adx. The 351–361 peptide (helix K) contains K358, whose mutation to alanine abolishes cholesterol hydroxylation when the redox partner is Adx but has no effect on cholesterol hydroxylation when the redox partner is CPR (Pikuleva I.A., personal communication). The 351–361 peptide also defines a side of the proximal channel (presumably water channel) that is formed upon binding of cholesterol sulfate and filled with a network of hydrogen-bonded water molecules (39). Our H-D exchange data indicate that cholesterol binding alters the deuteration of this region as well. As for the 151–164 peptide, residues within this peptide (that form a part of helix D) have not yet been mutated but the neighboring R147 in the same helix was. As with the K358A mutation, the R147A substitution had no effect on cholesterol hydroxylation when the redox partner was CPR, but reduced the rate of cholesterol hydroxylation [asymptotically equal to]3-fold when the redox partner was Adx (Pikuleva I.A., personal communication). R147, V151, M154, N158, E162, and E166 in the D helix form a network of hydrogen bonds via their backbone atoms. Consequently, cholesterol-induced changes in the 151–164 peptide may affect the position of the side chain of R147 and thus interaction with Adx.

Figure 7
Regions in substrate-free CYP46A1 exhibiting altered deuterium incorporation upon binding of cholesterol (in red). Corresponding peptides in cholesterol sulfate-bound are in cyan. Unaffected regions in substrate-free CYP46A1 are in orange. Cholesterol ...

A number of peptides having altered deuterium incorporation upon cholesterol sulfate binding did not show changes in the deuterium uptake upon cholesterol binding (Table 1). The lack of changes in deuteration of the 117–125 (B’-C loop) and 364–373 (β4–1 sheet) peptides that contain F121, V126 and A367, which restrain the side chain of cholesterol sulfate, suggests a different position of the steroid side chain in cholesterol-bound CYP46A1 (Figure 7). Also, the lack of changes in the 225–239 peptide (forms a part of the F–G loop) that contains N227 hydrogen bonded to the sulfate anion in cholesterol sulfate indicates that the cholesterol 3β-hydroxyl does not interact with the residues in the F-G loop. Finally, cholesterol binding did not affect the deuteration of the 445–455 peptide, one of the two peptides that form proximal channel in cholesterol sulfate-bound CYP6A1 (39), but reduced the deuterium uptake of the other 351–361 peptide. It possible that the hydration of the CYP46A1 active site in cholesterol-bound and cholesterol sulfate-bound enzymes is different. Cholesterol seems to induce smaller conformational changes in CYP46A1 than cholesterol sulfate, suggesting that the shape of the active site in cholesterol-bound CYP46A1 is more similar to that of the substrate-free enzyme than that of the cholesterol sulfate-bound form (Figure 7). The major differences from the substrate-free CYP46A1 appear to be at the entrance and in the middle the substrate access channel where the cholesterol 3β-hydroxyl and steroid nucleus interact with the residue(s)/backbone of CYP46A1. The similarity with cholesterol sulfate-bound CYP46A1 includes conformational changes in or close to the sites of the redox partner binding and around the water channel. Identification of these conformational changes is an important finding for the P450 field because it provides first structural insight into how substrate binding affects interaction with the redox partner and hydration of the P450 active site.

24S-Hydroxycholesterol binds tighter to CYP46A1 than cholesterol and is a better substrate for CYP46A1 than cholesterol (6, 7). The tighter binding of 24S-hydroxycholesterol is probably due to the additional interaction(s) provided by the 24S-hydroxyl group. Probably due to these additional interaction(s), the two peptides, 117–125 and 364–373, that contain F121 and A367 which define the position of the steroid side chain (Figure 8), and are unaffected by cholesterol binding, had altered H-D exchange upon 24S-hydroxycholesterol binding. The 117–125 peptide showed a decreased deuterium uptake, whereas the deuteration of the 364–373 peptide was increased. These results suggest that the side chain of 24S-hydroxycholesterol will be closer to the 117–125 peptide but farther from the 364–373 peptide than the side chain of cholesterol. Unlike cholesterol sulfate and cholesterol, 24S-hydroxycholesterol did not elicit changes in the deuteration of the 351–361 peptide but reduced deuterium uptake of the 260–270 peptide. The former is putatively involved in the interaction with Adx and also forms the entrance to the proximal channel; the function of the latter (which forms the C-terminal part of helix G) is unknown. These distinct features of 24S-hydroxycholesterol binding, lack of changes in the H-D exchange of the 351–361 peptide and reduced deuteration of the 260–270, may explain our previous observation pertinent to substrate dependence of the CYP46A1 redox partner interactions. We found that the catalytic efficiency of cholesterol hydroxylation is [asymptotically equal to]3-fold higher with Adx as a redox partner than with CPR, whereas for 24S-hydroxycholesterol hydroxylation, CPR is a much better redox partner than Adx (Pikuleva I.A., personal communication). The H-D exchange behavior of 24S-hydroxycholesterol-bound CYP46A1 suggests that 24S-hydroxycholesterol does not “prepare” CYP46A1 for efficient binding of Adx or maybe even deteriorate interaction with this redox partner.

Figure 8
Regions in substrate-free CYP46A1 exhibiting altered deuterium incorporation upon binding of 24S-hydroxycholesterol (in light blue). The unaffected 351–361 peptide is in orange. Heme is in pink, A367 and F121 are in blue, and the shape of the ...

We also tested how binding of one of the CYP46A1 substrates, cholesterol sulfate, affects interaction with Adx. Two peptides were affected, 225–239 and 351–361. The 225–239 peptide forms a part of the F-G loop which is disordered in substrate-free CYP46A1 but becomes more stabilized in cholesterol sulfate-bound CYP46A1 because N227 forms two hydrogen bonds with the sulfate anion (Figure 9). Binding of Adx made the F-G loop less dynamic in substrate-free CYP46A1 but had no effect on this region in the cholesterol sulfate-bound enzyme which is already stabilized by the interaction with the substrate. Also, interaction with Adx reduced deuteration of the 351–361 peptide in substrate-free CYP46A1, but we could not recover this peptide in all three experiments with the cholesterol sulfate-bound enzyme. Reduced deuterium uptake of the 351–361 peptide in substrate-free P450 is consistent with our mutagenesis studies suggesting that K358 in this peptide interacts with Adx. The lack of this peptide in the experiments with cholesterol sulfate-bound CYP46A1 is more difficult to explain. Probably, simultaneous binding of cholesterol sulfate and Adx affected the CYP46A1 cleavage in this region, and the 351–361 peptide was not generated. Substrate-free and substrate-bound forms of CYP46A1 also showed similarities in the H-D exchange in the presence of Adx. Reduced deuteration of the 144–150 peptide (the N-terminal part of the F helix) is consistent with our mutagenesis data suggesting the involvement of R147 in binding of Adx. Altered deuterium uptake of the 265–276 (the G–H loop) and 301–313 (a part of the I helix) peptides is a novel finding. The 265–276 peptide is spatially close to the 144–150 peptide (Figure 9). It is possible that interaction of Adx with R147 also affects the neighboring 265–276 peptide. The 301–313 peptide constitutes a part of the active site with I301, A302, and T306 defining the position of the side chain of cholesterol sulfate over the heme iron, the site of catalysis (7). In the crystal structure of CYP46A1, C24 and C25 of cholesterol sulfate, the primary and secondary sites of hydroxylation, are positioned at a 5.7 Å distance from the heme iron, which is ~1–1.5 Å greater than expected for hydroxylation of the oxyferryl intermediate during turnover (40). The reduced H-D exchange of the 301–313 peptide suggests that binding of the redox partner makes the substrate side chain or the whole steroid molecule move closer to the I helix and consequently the heme iron.

Figure 9
Regions in CYP46A1 affected by the binding of Adx. Peptides in substrate-free CYP46A1 are shown in orange, and in cholesterol sulfate-bound P450 are in cyan. Cholesterol sulfate is in yellow, heme is in pink, R147, N227, I301, A302, T306, and K358 are ...

In summary, by using H-D exchange and mass spectrometry we identified unique and common regions in CYP46A1 undergoing conformational changes upon binding of steroid substrates and the protein redox partner Adx. We gained structural insight into how substrate binding may affect the hydration of the P450 active site and why redox partner preferences of CYP46A1 depend on the steroid used. Results of the H-D exchange mass spectrometry become particularly informative when combined with the data obtained by other methods.

Supplementary Material

01

SUPPORTING INFORMATION AVAILABLE:

Figures S1 to S2 show the mean difference in deuterium incorporation between the control and treatment time points for the substrate and Adx binding experiments. Figures S3–S17 show selected duplicates of the deuterium uptake curves for the peptides presented in Table 1. This material is available free of charge via the Internet at http://pubs.acs.org.

Acknowledgments

This work was supported in part by the grants from the National Institutes of Health GM062882 and AG024336 (to I.A.P.) and by the Jules and Doris Stein Professorship from the Research to Prevent Blindness Foundation (to I.A.P.).

Certain commercial materials, instruments, and equipment are identified in this manuscript in order to specify the experimental procedure as completely as possible. In no case does such identification imply a recommendation or endorsement by the National Institute of Standards and Technology (NIST) nor does it imply that the materials, instruments, or equipment identified is necessarily the best available for the purpose.

Footnotes

1Abbreviations used: CYP, cytochrome P450; CYP46A1, cytochrome P450 46A1; CPR, NADPH-cytochrome P450 oxidoreductase; AdR, adrenodoxin reductase; Adx, adrenodoxin; H-D, hydrogen-deuterium; ESI, electrospray ionization; MALDI, matrix assisted laser desorption ionization; MS, mass spectrometry; TOF, time-of-flight; TFA, trifluoroacetic acid

REFERENCES

1. Pikuleva IA. Cytochrome P450s and cholesterol homeostasis. Pharmacol. Therapeut. 2006;112:761–773. [PubMed]
2. Pikuleva IA. Cholesterol-metabolizing cytochromes P450. Drug Metab. Disposition. 2006;34:513–520. [PubMed]
3. Lund EG, Guileyardo JM, Russell DW. cDNA cloning of cholesterol 24-hydroxylase, a mediator of cholesterol homeostasis in the brain. Proc. Natl. Acad. Sci. U. S. A. 1999;96:7238–7243. [PubMed]
4. Bjorkhem I, Andersson U, Ellis E, Alvelius G, Ellegard L, Diczfalusy U, Sjovall J, Einarsson C. From brain to bile. Evidence that conjugation and ω-hydroxylation are important for elimination of 24S-hydroxycholesterol (cerebrosterol) in humans. J. Biol. Chem. 2001;276:37004–37010. [PubMed]
5. Janowski BA, Grogan MJ, Jones SA, Wisely GB, Kliewer SA, Corey EJ, Mangelsdorf DJ. Structural requirements of ligands for the oxysterols liver X receptors LXRα and LXRβ Proc. Natl. Acad. Sci. USA. 1999;96:266–271. [PubMed]
6. Mast N, Norcross R, Andersson U, Shou M, Nakayama K, Bjorkhem I, Pikuleva IA. Broad substrate specificity of human cytochrome P450 46A1 which initiates cholesterol degradation in the brain. Biochemistry. 2003;42:14284–14292. [PubMed]
7. Mast N, White MA, Bjorkhem I, Johnson EF, Stout CD, Pikuleva IA. Crystal structure of substrate-bound and substrate-free cytochrome P450 46A1, the principal cholesterol hydroxylase in the brain. Proc. Natl. Acad. Sci. U.S.A. 2008;105:9546–9551. [PubMed]
8. Busenlehner LS, Armstrong RN. Insights into enzyme structure and dynamics elucidated by amide H/D exchange mass spectrometry. Arch. Biochem. Biophys. 2005;433:34–46. [PubMed]
9. Engen JR. Analysis of protein complexes with hydrogen exchange and mass spectrometry. Analyst. 2003;128:623–628. [PubMed]
10. Wales TE, Engen JR. Hydrogen exchange mass spectrometry for the analysis of protein dynamics. Mass Spectrom. Rev. 2006;25:158–170. [PubMed]
11. Hoofnagle AN, Resing KA, Ahn NG. Protein analysis by hydrogen exchange mass spectrometry. Annu. Rev. Biophys. Biomol. Struct. 2003;32:1–25. [PubMed]
12. Tsutsui Y, Wintrode PL. Hydrogen/deuterium exchange-mass spectrometry: a powerful tool for probing protein structure, dynamics and interactions. Curr. Med. Chem. 2007;14:2344–2358. [PubMed]
13. Figueroa ID, Russell DH. Matrix-assisted laser desorption ionization hydrogen/deuterium exchange studies to probe peptide conformational changes. J. Am. Soc. Mass Spectrom. 1999;10:719–731. [PubMed]
14. Kipping M, Schierhorn A. Improving hydrogen/deuterium exchange mass spectrometry by reduction of the back-exchange effect. J. Mass Spectrom. 2003;38:271–276. [PubMed]
15. Oliva A, Llabres M, Farina JB. Application of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry and hydrogen exchange combined with enzymatic digestion for the structural characterization of antimalaric Apf66 peptide. Talanta. 2007;72:1192–1198. [PubMed]
16. Nazabal A, Reis SD, Bonneu M, Saupe SJ, Schmitter J-M. Conformational transition occurring upon amyloid aggregation of the HET-s prion protein of Podospora anserine analyzed by hydrogen/deuterium exchange and mass spectrometry. Biochemistry. 2003;42:8852–8861. [PubMed]
17. Nazabal A, Schmitter J-M. Hydrogen-deuterium exchange analyzed by matrix-assisted laser desorption-ionization mass spectrometry and the HET-s prion model. Methods Enzymol. 2006;413:167–181. [PubMed]
18. Baerga-Ortiz A, Hughes CA, Mandell JG, Komives EA. Epitope mapping of a monoclonal antibody against human thrombin by H/D-exchange mass spectrometry reveals selection of a diverse sequence in a highly conserved protein. Protein Sci. 2002;11:1300–1308. [PubMed]
19. Man P, Montagner C, Vernier G, Dublet B, Chenal A, Forest E, Forge V. Defining the interacting regions between apomyoglobin and lipid membrane by hydrogen/deuterium exchange coupled to mass spectrometry. J. Mol. Biol. 2007;368:464–472. [PubMed]
20. Powell KD, Fitzgerald MC. High-throughput screening assay for the tunable selection of protein ligands. J. Comb. Chem. 2004;6:262–269. [PubMed]
21. Ma L, Fitzgerald MC. A new H/D exchange- and mass spectrometry-based method for thermodynamic analysis of protein-DNA interactions. Chemistry & Biology. 2003;10:1205–1213. [PubMed]
22. Tang L, Roulhac PL, Fitzgerald MC. H/D exchange and mass spectrometry-based method for biophysical analysis of mutidomain proteins at the domain level. Anal. Chem. 2007;79:8728–8739. [PubMed]
23. Roulhac PL, Weaver KD, Adhikari P, Anderson DS, DeArmond PD, Mietzner TA, Crumbliss AL, Fitzgerald MC. Ex vivo analysis of synergistic anion binding to FbpA in Gram-negative bacteria. Biochemistry. 2008;47:4298–4305. [PubMed]
24. Johnson RS, Walsh KA. Mass spectrometric measurement of protein amide hydrogen exchange rates of apo- and holo-myoglobin. Protein Sci. 1994;3:2411–2418. [PubMed]
25. Mandell JG, Falick AM, Komives EA. Measurement of amide hydrogen exchange by MALDI-TOF mass spectrometry. Anal. Chem. 1998;70:3987–3995. [PubMed]
26. Mandell JG, Baerga-Ortiz A, Akashi S, Takio K, Komives EA. Solvent accessibility of the thrombin-thrombomodulin interface. J. Mol. Biol. 2001;306:575–589. [PubMed]
27. Turner BT, Jr, Maurer MC. Evaluating the roles of thrombin and calcium in the activation of coagulation factor XIII using H/D exchange and MALDI-TOF MS. Biochemistry. 2002;41:7947–7954. [PubMed]
28. Koeppe JR, Komives EA. Amide H/2H exchange reveals a mechanism of thrombin activation. Biochemistry. 2006;45:7724–7732. [PMC free article] [PubMed]
29. Anand GS, Hotchko M, Brown SHJ, Ten Eyck LF, Komives EA, Taylor SS. R-subunit isoform specificity in protein kinase A: distinct features of protein interfaces in PKA types I and II by amide H/2H exchange mass spectrometry. J. Mol. Biol. 2007;374:487–499. [PMC free article] [PubMed]
30. Sabo TM, Brasher PB, Maurer MC. Perturbations in factor XIII resulting from activation and inhibition examined by solution based methods and detected by MALDI-TOF MS. Biochemistry. 2007;46:10089–10101. [PubMed]
31. Gu Z, Zitzewitz JA, Matthews CR. Mapping the structure of folding cores in TIM barrel proteins by hydrogen exchange mass spectrometry: the roles of motif and sequence for the indole-3-glycerol phosphate synthase from Sulfolobus solfataricus. J. Mol. Biol. 2007;368:582–594. [PMC free article] [PubMed]
32. Laine O, Streaker ED, Nabavi M, Fenselau CC, Beckett D. Allosteric signaling in the biotin repressor occurs via local folding coupled to global dampening of protein dynamics. J. Mol. Biol. 2008;381:89–101. [PMC free article] [PubMed]
33. Alverdi V, Mazon H, Versluis C, Hemrika W, Esposito G, van den Heuvel R, Scholten A, Heck AJR. cGMP-binding prepares PKG for substrate binding by disclosing the C-terminal domain. J. Mol. Biol. 2008;375:1380–1393. [PubMed]
34. Schuster MC, Ricklin D, Papp K, Molnar KS, Coales SJ, Hamuro Y, Sfyroera G, Chen H, Winters MS, Lambris JD. Dynamic structural changes during complement C3 activation analyzed by hydrogen/deuterium exchange mass spectrometry. Mol. Immunol. 2008;45:3142–3151. [PMC free article] [PubMed]
35. White MA, Mast N, Bjorkhem I, Johnson EF, Stout CD, Pikuleva IA. Use of complementary cation and anion heavy-atom salt derivatives to solve the structure of cytochrome P450 46A1. Acta Cryst. 2008;D64:487–495. [PMC free article] [PubMed]
36. Pikuleva IA, Tesh K, Waterman MR, Kim Y. The tertiary structure of full-length bovine adrenodoxin suggests functional dimers. Arch. Biochem. Biophys. 2000;373:44–55. [PubMed]
37. Hotchko M, Anand GS, Komives EA, Ten Eyck LF. Automated extraction of backbone deuteration levels from amide H/2H mass spectrometry experiments. Protein Sci. 2006;15:583–601. [PubMed]
38. R Development Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 2008. ISBN 3-900051-07-0, URL http://www.R-project.org.
39. Pikuleva IA. Cholesterol-metabolizing cytochromes P450: implications for cholesterol lowering. Expert Opin. Drug Metab. Toxicol. 2008;4:1403–1414. [PMC free article] [PubMed]
40. Poulos TL, Johnson EF. In: Structures of Cytochrome P450 Enzymes. 3rd Ed. Ortiz de Montellano PR, editor. New York: Kluwer; 2005. pp. 87–114.