Investigation on drug-binding in heme pocket of CYP2C19 with UV–visible and resonance Raman spectroscopies

Accepted Manuscript Investigation on drug-binding in heme pocket of CYP2C19 with UV–visible and resonance Raman spectroscopies

Sayed M. Derayea, Hirofumi Tsujino, Yukiko Oyama, Yoshinobu Ishikawa, Taku Yamashita, Tadayuki Uno PII: DOI: Reference:

S1386-1425(18)30978-8 https://doi.org/10.1016/j.saa.2018.10.045 SAA 16551

To appear in:

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy

Received date: Revised date: Accepted date:

23 July 2018 23 October 2018 25 October 2018

Please cite this article as: Sayed M. Derayea, Hirofumi Tsujino, Yukiko Oyama, Yoshinobu Ishikawa, Taku Yamashita, Tadayuki Uno , Investigation on drug-binding in heme pocket of CYP2C19 with UV–visible and resonance Raman spectroscopies. Saa (2018), https://doi.org/10.1016/j.saa.2018.10.045

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Investigation on drug-binding in heme pocket of CYP2C19 with UV-visible and resonance Raman spectroscopies Sayed M. Derayea1,2, Hirofumi Tsujino1, Yukiko Oyama3, Yoshinobu Ishikawa4, Taku Yamashita5, and Tadayuki Uno1

Graduate School of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka,

PT

1

Suita 565-0871, Japan, 2

Analytical Chemistry Department, Faculty of Pharmacy, Minia University, Minia,

Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1

Oehonmachi, Kumamoto 862-0973, Japan

Graduate School of Integrated Pharmaceutical and Nutritional Sciences, University

NU

4

SC

3

RI

Egypt

of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan 5

School of Pharmacy and Pharmaceutical Sciences, Mukogawa Women’s University,

MA

11-68 Koshien-Kyubancho, Nishinomiya 663-8179, Japan

D

Running Head: Resonance Raman and UV spectroscopy of CYP2C19

PT E

Correspondence author to: Sayed M. Derayea, current address is; Faculty of Pharmacy, Minia University, Minia 61519, Egypt. Tel./ fax: (+20) 86-2369075,: mobile : (+20) 114-33-52507.

AC

CE

E-mail: [email protected], [email protected]

ACCEPTED MANUSCRIPT Abstract Cytochrome P450 (CYP) is a class of heme-containing enzymes which mainly catalyze a monooxygenation reaction of various chemicals, and hence CYP plays a key role in the drug metabolism. Although CYP2C19 isoform is a minor hepatic CYP, it metabolizes clinically important drugs such as omeprazole and S-mephenytoin. In this work, the interaction of purified CYP2C19 WT (CYP2C19) with seven drugs

PT

(phenytoin, S-mephenytoin, omeprazole, lansoprazole, cimetidine, propranolol, and warfarin) was investigated using spectroscopic methods. The binding of each drug

RI

and the induced structural change in the heme distal environment were evaluated. Ferric form of CYP2C19 was revealed to contain a six-coordinate low-spin heme with

SC

a water molecule as a sixth ligand in a distal site, and the addition of each drug caused varied minor fraction of five-coordinate heme. It was suggested that the ligated water

NU

molecule was partly moved away from the heme distal environment and that the degree of water removal was dependent on the type of drugs. The effect on the

MA

coordination was varied with the studied drugs with wide variation in the dissociation constants from 2.6 µM for lansoprazole to 5400 µM for warfarin. Phenytoin and Smephenytoin showed that binding to CYP2C19 occurred in a stepwise manner and

D

that the coordination of a water molecule was facilitated in the second binding step. In

PT E

the ferrous CO-bound state, (Fe-CO) stretching mode was clearly observed at 471 cm-1 in the absence of drugs. The Raman line was greatly up-shifted by omeprazole (487 cm-1) and lansoprazole (477 cm-1) but was minimally affected by propranolol,

CE

phenytoin, and S-mephenytoin. These results indicate that slight chemical

AC

modification of a drug greatly affects the heme distal environments upon binding.

ACCEPTED MANUSCRIPT Key words CYP2C19; Raman spectroscopy, Spectrophotometry; Drug binding; Substrate recognition. 1. INTRODUCTION Cytochromes P450 (CYPs) are heme-thiolate monooxygenases which metabolize a

PT

vast array of endogenous and exogenous substrates and hence play a central role in drug metabolism and detoxification [1-3]. Based on the sequence homology, the

RI

multiple CYP enzymes are classified into families (the first number), subfamilies (the second letter), and isoforms [4]. More than fifty CYPs are contained in human, and

SC

five CYPs (1A2, 2C9, 2C19, 2D6, and 3A4) have been shown to be responsible for the metabolism of more than 90% of drugs in clinical use [5, 6]. Unlike most classical

NU

enzymes with their strict substrate selectivity, however, these microsomal CYPs can each metabolize a number of substrates that are different in size, shape, and

MA

stereochemistry [7]. Furthermore, inter-individual and ethnic differences in the drug metabolism are a clinically important problem in the use of many drugs, and the differences mainly stem from genetic polymorphisms of CYP genes [8-10]. Thus, it is

D

crucial to elucidate properties of the CYP isoforms that determine the selectivity and

PT E

metabolism of drugs for the personalized drug usage. In the mammalian CYP2 family enzymes, six potential substrate recognition sites (SRS) have been predicted [11]. Recently, crystal structures of clinically significant

CE

human CYPs, 1A2 [12], 2C9 [13, 14], 2C19 [15], 2D6 [16], and 3A4 [17-19], have been solved in drug-free and -bound forms, and the SRSs have been revealed to locate close to the drugs in the structures of drug-bound forms (Fig. 1). At the tertiary

AC

structural level, CYP enzymes exhibit a similar overall fold with a well conserved heme-binding core [20, 21]. However, recent results demonstrate that microsomal CYPs exhibit striking conformational diversity and plasticity [22-24]. Not only do different CYPs exhibit different conformations, but an individual CYP can adopt multiple conformations in response to various drugs. This plasticity of CYP enzymes makes it difficult to model protein-drug interactions even when the atomic structure of the CYP is known [25]. CYP2C19 is a member of the human CYP2C subfamily, which includes four structurally related enzymes [26]. CYP2C9 and 2C19 are the most highly conserved

ACCEPTED MANUSCRIPT of these forms showing 91% structural identity, but have very distinctive substrate specificities. Although CYP2C19 isoform is a minor hepatic CYP, it is responsible for the metabolism of a wide variety of clinically important drugs. For example, CYP2C19 is the principal enzyme responsible for the stereoselective 4'-hydroxylation of S-mephenytoin [27, 28] and is highly selective for the 5-hydroxylation of the popular anti-ulcer drug omeprazole [28] and lansoprazole [29]. The structurally

PT

related CYP2C9 exhibits little activity toward these substrates, but exhibits phenytoin pro-S phenyl ring hydroxylation [30, 31] and S-warfarin 7-hydroxylation [26]. These two drugs are also metabolized by CYP2C19, but both the pro-S and pro-R rings of

RI

phenytoin [30, 31], and both S- and R-warfarin at a larger number of sites (mainly 7

SC

and 8 positions) [32], are hydroxylated with a lower catalytic efficiency. CYP2C19 is partially responsible for the metabolism of -blocker propranolol, through oxidation

NU

of its side chain [33]. The chemical structures of the previously mentioned drugs are presented in Fig. 2.

MA

In the present study an Escherichia coli over-expression system of CYP2C19 was constructed, in which the enzyme is expressed in a soluble form. The drug binding properties of these seven drugs with CYP2C19 was evaluated using absorption and

D

resonance Raman spectroscopic techniques. The highly purified and soluble enzyme

PT E

without any detergents enabled the investigation of the molecular basis on how this enzyme recognizes and metabolizes particular drugs in a regio- and stereo-specific manner. A role of key residues in the substrate specificity of CYP2C19 will be

CE

discussed. 2. Experimental:

AC

2.1.Chemicals and Reagents: Lansoprazole (L-8533-1G) and phenytoin sodium (D-4505) were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). Omeprazole (150-02091), S-mephenytoin (581-20371), propranolol hydrochloride (161-11591), warfarin sodium (239-02171), cimetidine, (034-16312) and other chemicals were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All these drugs were used without further purification. 2.2. Protein Expression and Purification:

ACCEPTED MANUSCRIPT The CYP2C19 WT gene was constructed previously. The CYP2C9 WT gene was designed, synthesized, and ligated into a pET3a-based expression vector (pBEX).as described previously [34]. Escherichia coli strain BL21 Gold (DE3) was transformed with the plasmids, 2C19 WT was expressed and purified as previously reported [35].The purity of the prepared 2C19 enzyme was evaluated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). A single band was observed

PT

in the gel as previously reported for CYP2C19 WT [35]. The enzyme concentrations were evaluated from the heme content, which was measured by a pyridine hemochrome assay. The protein solutions were flash frozen with liquid nitrogen and

RI

stored at -80 C until use.

SC

2.3. Absorption Spectroscopy

The absorption spectra of the purified CYP2C19 were recorded on a DU640

NU

spectrophotometer (Beckman) at 25 C. Drug titrations of the ferric CYP2C19 were performed in buffer A (100 mM potassium phosphate (pH 7.4) and 20% glycerol).

MA

Stock solutions of cimetidine, phenytoin sodium, propranolol hydrochloride and warfarin sodium were prepared in buffer A, while S-mephenytoin lansoprazole and omeprazole were dissolved in dimethyl sulfoxide, (DMSO). Small aliquots of the

D

stock solutions were added via a syringe to ~8 µM CYP2C19 solutions. The DMSO

PT E

content was limited to less than 10% of the total volume in the cuvette. The absorbance changes at Soret band region were traced and analyzed. 2.4. Resonance Raman Spectroscopy

CE

The spectra were recorded using a triple monochromator (Jasco NR-1800) with a slit width of 6 cm-1, following excitation by a krypton ion laser (406.7-nm line, Coherent

AC

I-302) or by a He-Cd laser (441.6-nm line, Kimmon IK4121R-G). A CCD detector (ANDOR, DU420BU) was used, and the frequencies were calibrated with indene. A spinning Raman cell was used throughout the measurements to minimize local heating and CO photo-dissociation. The samples contained 100 µM enzyme in 100 mM potassium phosphate (pH 7.4) and 20% glycerol. Ferrous CO-bound proteins were prepared by exposing the samples to 1 atm CO for 2-3 min then by reducing with sodium dithionite. The fluorescence background of the spectra was flattened, if necessary, by subtraction of a smooth polynomial curve that had been fitted to the baseline of a spectrum. Corrected spectra were compared with the raw data to ensure that no artifacts had been generated.

ACCEPTED MANUSCRIPT 3. RESULTS AND DISCUSSION 3.1. Absorption Spectra In Fig. 3A, the change in the absorption spectrum of CYP2C19 upon addition of Smephenytoin is shown as a representative example. In the absence of the drug, a Soret maximum is observed at 416 nm and Q bands are apparent at 534 and 569 nm. The spectral profile is quite similar to that reported for substrate-free cytochrome P450cam

PT

(CYP101) [36]. Upon addition of the drug (74, 148, 293 and 850 µM), the Soret maximum decreased with concomitant appearance of a shoulder around 390 nm. A

RI

single set of isosbestic points was clearly observed, indicating that the binding of S-

terms of a simple equilibrium as shown below: CYP2C19 + Drug ⇄ CYP2C19-Drug

NU

Kd

SC

mephenytoin is a one-step equilibrium process. Thus, the binding was analyzed in

Kd = [CYP2C19][Drug]/[CYP2C19-Drug]

(1) (2)

MA

where Kd is the dissociation constant for the drug. The total concentration of CYP2C19 ([CYP2C19]0) is written as

[CYP2C19]0 = [CYP2C19] + [CYP2C19-Drug].

(3)

D

Defining the molar fraction of CYP2C19-Drug complex () as

PT E

 = [CYP2C19-Drug]/[CYP2C19]0

(4)

and using eqs. 2 and 3,  is rearranged to give  = 1/(1 + Kd/[Drug]).

(5)

CE

The molar fraction of drug-bound CYP2C19 is calculated from absorbance changes at the Soret maximum and is plotted against the logarithm of free drug concentrations.

AC

The same titration and data plotting were performed for all the studied drugs. Fig. 3B shows the obtained curves for lansoprazole, phenytoin, omeprazole, S-mephenytoin, propranolol, cimetidine; and warfarin. The theoretical curve (eq. 5) provides a satisfactory fit to the titration data for all drugs, indicating that one mol of a drug binds to one mole of CYP2C19. The obtained Kd values were 2.6, 15, 30, 131, 1010, 1610, and 5400, for the studied drugs, respectively (Table I). It is surprising that Kd values are variable more than three orders of magnitude, in spite of the fact that the six drugs (except for cimetidine) are reported to be metabolized by CYP2C19. Lansoprazole shows the highest affinity, while the Kd value for warfarin is larger than that for an inhibitor, cimetidine.

ACCEPTED MANUSCRIPT The absorption spectral change accompanied the drug binding to CYP2C19 indicates that the binding affects the electronic structure of the heme, and hence the spectral change was studied in detail. In Fig. 4, difference absorption spectra are shown for the seven drugs. A single set of isosbestic points is again clear for the binding of each drug. For the substrate drugs, a peak and trough are observed around 390 nm and 420 nm, respectively, and this type of difference spectra has been named

PT

"type I" [37, 38]. On the other hand, the spectral pattern is reversed for the inhibitor cimetidine, showing the peak and trough around 427 nm and 397 nm, respectively (type

II) (Fig. 4H). Although phenytoin is a substrate for CYP2C19, it gave a

RI

"reverse type I" spectral change (Fig. 4G). Thus, the phenytoin binding was closely

SC

examined at very low drug concentration with enlarged absorbance scale. Although the difference spectra are very small in the presence of dilute phenytoin, type I

NU

spectral pattern was observed, similar to the other five substrate drugs. Thus, it seems that substrate drugs always give type I difference spectra. Since phenytoin showed two-step binding to CYP2C19, chemically homologous S-mephenytoin (Fig. 2) may

MA

bind similarly. Thus, the binding of S-mephenytoin was further examined by the use of difference absorption spectroscopy (Fig. 4C). Although the measurement was

D

limited by the drug solubility, the peak at about 390 nm decreased while the trough became shallow by further addition of the drug. Therefore, it is concluded that

PT E

phenytoin and S-mephenytoin similarly provide type I spectra at first, and the binding of second drug molecule reverses the spectra.

CE

3.2. Resonance Raman Spectra

In order to gain insight into the heme structural change upon binding of drugs,

AC

resonance Raman spectra of ferric CYP2C19 were measured (Fig. 5). Resonance Raman spectroscopy is a powerful tool for revealing the oxidation states, spin states, and coordination numbers of heme iron [39, 40]. In the drug-free resting form, the enzyme showed Raman lines at 1584, 1501, and 1372 cm-1 which are assignable to the 2, 3, and 4 modes, respectively. These frequencies are typical of six-coordinate ferric low-spin hemes, suggesting that a water molecule occupies the iron sixth coordination site. This idea is supported by the crystal structure of CYP101 [41] which shows an absorption spectrum similar to CYP2C19 in the absence of substrates [36]. The Raman spectrum for drug-free form was essentially unchanged by the addition of drugs. In the presence of propranolol, however, a weak but distinct peak is

ACCEPTED MANUSCRIPT apparent at 1485 cm-1 with concomitant increase of a peak around 1622 cm-1 (Fig. 5A). Similar peaks are apparent in the presence of S-mephenytoin and warfarin (Fig. 5, B and C). The 1485 cm-1 line can be assigned to the 3 mode, indicating the presence of five-coordinate ferric low-spin heme. Thus, the aqua ligand was partly removed by the binding of propranolol, S-mephenytoin, and warfarin. The three drugs (propranolol, S-mephenytoin, and warfarin) showed relatively

PT

large absorption change in the difference spectra (Fig. 4). Thus, the peak-to-trough maximal difference in the absorbance was estimated and expressed as millimolar

RI

extinction coefficient (∆, Table I). Since the spectral pattern is reversed in the case of cimetidine and phenytoin, a negative sign is given to the values. It is clear that the ∆

SC

value is in parallel to the peak intensity at 1485 cm-1 (Fig. 5). In other words, the ∆ value is an index of the removal of the iron aqua ligand. On the other hand, the

NU

content of aqua complex increased by the addition of large amount of phenytoin, giving rise to reverse type I difference spectra. This suggests that the aqua ligand is

MA

stabilized by the binding of second phenytoin molecule. In the case of cimetidine, the aqua ligand may be replaced by the drug to give similar six-coordinate heme iron, since it gave type II difference spectra. It should be noted here that the drug

D

concentration used for the measurement is more than about ten-times of the Kd value

PT E

(Table I) and hence more than 90% of CYP2C19 is saturated by the drug. However, in the case of S-mephenytoin, the degree of drug saturation is limited to about 80% (Fig. 3B) to eliminate the effect of second drug binding.

CE

Since the drug binding was revealed to affect the water coordination at the iron, the heme distal environment was investigated with a CO probe. The (Fe-CO) stretching

AC

frequency is sensitive to the polarity of the residues around the bound CO and is therefore, an excellent probe of the distal environment [42-44]. In Fig. 6, resonance Raman spectra of ferrous CO-bound CYP2C19 are shown. In the absence of drugs, a strong peak is observed at 471 cm-1 which is assignable to the (Fe-CO) stretching mode (Fig. 6H). The frequency is larger than that observed for the substrate-free CYP101 from Pseudomonas putida (464 cm-1) [45] but is the same as that for cytochrome P450BM3 (CYP102) from Bacillus megaterium (471 cm-1) [46]. Thus, the distal polarity in human CYP2C19 is quite similar to that in bacterial CYP102. Upon addition of some drugs, the (Fe-CO) stretch was greatly affected. Especially noted is the omeprazole-bound form, in which the (Fe-CO) stretching frequency is

ACCEPTED MANUSCRIPT observed at 487 cm-1 (Fig. 6D). This frequency is as high as that reported for androstenedione-bound (485 cm-1) and 19-aldehyde-androstenedione-bound (490 cm1

) forms of cytochrome P450arom (CYP19) [47]. In the presence of lansoprazole which

is chemically homologous to omeprazole (Fig. 2), the (Fe-CO) stretch was up-shifted to 477 cm-1, while it was slightly up-shifted to 474 cm-1 by warfarin. However, the (Fe-CO) stretching frequency was minimally affected in the presence of propranolol, S-mephenytoin, phenytoin and cimetidine. It is also observed that the band width for

PT

the (Fe-CO) stretch is affected by the drug binding. Omeprazole, lansoprazole and

RI

propranolol induced the broadening of the band, while S-mephenytoin did not. The (Fe-CO) stretching frequencies and the band width (full-width at half height) are

SC

summarized in Table 2. The broad (Fe-CO) stretching band may imply multiple FeCO geometry, suggesting the flexible orientation of the drug bound in the heme distal

NU

side. 3.3. Water Removal:

MA

In the absence of drugs, an aqua ligand occupies the iron sixth coordination site in CYP2C19. However, the water occupancy is not 100%, based on the observation that phenytoin further facilitated the water coordination in the second binding step (Fig.

Phenytoin and S-mephenytoin, which are chemically

PT E

difference spectra.

D

4G). In the first step, phenytoin slightly removed the axial aqua ligand to afford type I

homologous, are hydroxylated similarly at the phenyl ring (Fig. 2), indicating that the phenyl moiety is directed to the heme iron to be monooxygenated. It is thus

CE

reasonable to assume that the close location of the hydrophobic phenyl ring destabilized the water coordination in their first binding.

AC

The phenyl ring of S-mephenytoin, however, minimally affected the Fe-CO vibration (Fig. 6B), although the drug is supposed to have saturated the enzyme to ~80% level (Fig. 3B). Thus, most of the bound drug locates not so close to the iron center. It should be noted here that the water removal is incomplete by the binding of the studied substrate drugs, and the most part of CYP2C19 remains to be six-coordinated (Fig 6). A small part of bound S-mephenytoin may locate close to the iron to remove the aqua ligand and is hydroxylated. The substrate binding pocket in CYP2C19 is suggested to be large enough to accommodate the second drug molecule, and even larger phenytoin clearly facilitated the water coordination in the second binding step

ACCEPTED MANUSCRIPT (Fig. 4G). In this case, the hydantoin moiety may have stabilized the aqua coordination possibly through the formation of hydrogen bonding network. The aqua ligand is also partly removed by the binding of omeprazole, lansoprazole, warfarin, and propranolol which are CYP2C19 substrates. These drugs are relatively large, and one mol of the drug binds to one mol of CYP2C19 over the examined concentration range. The hydrophobic moieties of these four drugs are

PT

hydroxylated and hence are expected to be directed toward the heme iron, removing the aqua ligand. The hydrocarbon chain of propranolol removed the water most effectively (large ∆, Table I), but less affected the (Fe-CO) stretching frequency

RI

(Table 2). However, the (Fe-CO) band is asymmetric and broadened by the

SC

propranolol binding (Fig. 6A), indicating a minor but distinct perturbation of CO environment produced by the drug. In the case of warfarin, the aqua ligand is removed

NU

to the extent similar to S-mephenytoin (Table I), and the (Fe-CO) stretch is slightly up-shifted (Table 2). Since warfarin is hydroxylated at multiple sites (Fig. 2), it may

MA

be flexible in the distal substrate binding pocket of the CYP2C19. Omeprazole and lansoprazole similarly removed the aqua ligand, and it is reasonable to see that they are chemically homologous. These drugs affected the (Fe-

D

CO) stretch greatly. Omeprazole up-shifted the frequency by 16 cm-1 while,

PT E

lansoprazole by only 6 cm-1 (Fig. 6, D and E). This difference should be attributed to the methoxyl moiety of omeprazole, since nearby position 5 of the benzimidazole ring is hydroxylated. The direct interaction between the methoxyl group and the bound

CE

CO, however, may not be possible, since negatively polar methoxyl oxygen should have made the carbonyl oxygen more positive, reducing the (Fe-CO) stretching

AC

frequency [48]. Thus, positively polar residues and/or water hydrogens may have been rearranged to direct to the bound CO moiety, through the interaction with the drug methoxyl moiety. The positive residue might be His99, since I99H mutation of CYP2C9 has been reported to confer omeprazole 5-hydroxylase activity to this isoform [49]. The benzimidazole moiety as well as the possible polar residue, however, are slightly remote from the heme and therefore, do not greatly affect the aqua ligand in the ferric state (Fig. 4). Cimetidine is an inhibitor of CYP2C19, and its binding facilitated the six-coordinate state (Fig. 4H). Crystal structures of some inhibitor-bound CYPs are now available. In fluconazole-bound CYP51 from Mycobacterium tuberculosis [50], ketoconazole-

ACCEPTED MANUSCRIPT bound P450eryF (CYP107) from Saccharopolyspora erythraea [51], and bifonazolebound rabbit CYP2B4 [25], azole nitrogens directly coordinate the heme iron, and type II difference spectra are induced [25, 51]. In the case of fluconazole-bound CYP121 from Mycobacterium tuberculosis [52], however, fluconazole does not displace the aqua ligand but occupies a position that allows formation of a direct hydrogen bond to the aqua ligand. In CYP121, the Soret maximum for the

PT

fluconazole complex is centered at 421 nm slightly shorter than that for azole complexes (~425 nm), supporting an altered ligation mode that may not involve exclusively direct iron-nitrogen ligation. In our cimetidine-CYP2C19 complex, a peak

RI

at 427 nm and a trough at 397 nm are induced in the difference spectra (Fig. 4H).

SC

They are apparently shorter than those for ketoconazole-CYP107 complex (433 and 413 nm) [51] and bifonazole-CYP2B4 complex (433 and 411 nm) [25] but are almost

NU

identical with those for ketoconazole-CYP121 complex (430 and 396 nm) [53]. Thus, type II spectral profile may be classified into two sub-groups. The inhibitors fluconazole, ketoconazole, and bifonazole have no steric restraints on the atom

MA

neighboring the coordinating nitrogen, while cimetidine has methyl group and long chain next to the nitrogens on the imidazole ring (Fig. 2). These substituents may

D

prevent the imidazole moiety from coordinating the heme iron directly. Alternatively, the nitrogens on the long chain may interact with the aqua ligand. In this case,

geometry (Fig. 6G).

PT E

however, the long side chain may be flexible and minimally affect the bound CO

3.4. Enzyme Reaction

CE

The aqua ligation to heme iron is one of the key determinant of CYP hydroxylation reaction. For example, the binding of the substrate camphor to CYP101 almost

AC

removes the aqua ligand [54], shifts the spin equilibrium toward the high-spin form, and also shifts the redox potential from about -300 to -170 mV [55]. A linear free energy relationship between redox potential and spin equilibrium has been established [55]. The great increase in the potential facilitates electron transfer from redox partners to CYP enzymes and hence subsequent catalytic reactions. This elevation is invoked by very slight change as large as 0.4 Å in the location of substrates [54]. In the binding of substrate drugs examined in the present study, the aqua ligand was partly displaced from the heme iron, inducing type I difference spectra (Fig. 4). Thus, a part of drugs bound in the heme distal side that removed the aqua ligand may effectively be hydroxylated. Although the second binding step was apparent for

ACCEPTED MANUSCRIPT phenytoin, this step may be clinically less important since therapeutic concentrations of protein-free phenytoin are estimated to be ≤ 5 µM in plasma [56]. S-warfarin, the active enantiomer of warfarin, is metabolized by CYP2C9 [57, 58] while R-warfarin is metabolized by CYP2C19. In the crystal structure of S-warfarinCYP2C9 complex, this drug lies in a predominantly hydrophobic pocket lined by residues Arg97, Gly98, Ile99, Phe100, Leu102, Ala103, Val113, Phe114, Asn217,

PT

Thr364, Ser365, Leu366, Pro367 and Phe476 [13]. These residues are completely conserved between CYP2C9 and 2C19 except for Ile99 (His99 in CYP2C19). Assuming that the protein structure is conserved between these two CYPs as well, the

RI

amino acid at position 99 may be a key determinant of the enantio-selective

SC

hydroxylation of warfarin, as well as omeprazole 5-hydroxylation [49]. This is the sole residue in SRS-1 that differs between the two CYPs. SRS-2 and SRS-6 are

NU

completely conserved, while SRS-3 (Leu/Val at 237) and SRS-5 (Ile/Leu at 362) are substituted synonymously at single position. Predominant differences are observed in SRS-4 in which five residues are substituted. Thus, SRS-4 should also play a key role

MA

in discriminating drug specificities between CYP2C9 and 2C19. 4. CONCLUSION

D

The UV spectral changes that occurred upon the addition of drug to CYP2C19 was

PT E

utilized for estimation of the binding constant (Kd) of the studied drugs. Lansoprazole shows the highest affinity, while the Kd value for warfarin is larger than that for an inhibitor, cimetidine. Raman resonance and also UV spectroscopic studies revealed

CE

that the water ligand from the heme was partly removed with different degrees upon binding of substrate drugs. In addition, it was found that slight chemical modifications

AC

of the drug should affect the drug specificities of CYP2C19. Omeprazole and lansoprazole are chemically homologous, but the Kd value and (Fe-CO) stretching frequency are greatly affected, indicating distinct drug-protein interactions. In the case of S-mephenytoin and phenytoin, the Kd value and the degree of aqua coordination was greatly affected, too. These findings further suggest that a slight modification of the drug may improve/impair its metabolism which is clinically applied. Comprehensive studies on the CYP-drug interactions as well as drug-drug interactions by the use of various spectroscopic techniques will elucidate mechanisms of drug recognition by CYPs at molecular level. Acknowledgment

ACCEPTED MANUSCRIPT This work was supported by Grant-in-Aid for Science Research (15390015 and 18390013 to T. U.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and by The Mitsubishi Foundation. Sayed M. Derayea

was

financially supported from Egyptian government through scholarship. 5. REFERENCES

AC

CE

PT E

D

MA

NU

SC

RI

PT

[1] S. Rendic, Summary of information on human CYP enzymes: human P450 metabolism data, Drug metabolism reviews, 34 (2002) 83-448. [2] T.D. Porter, M.J. Coon, Cytochrome P-450. Multiplicity of isoforms, substrates, and catalytic and regulatory mechanisms, The Journal of biological chemistry, 266 (1991) 13469-13472. [3] I.G. Denisov, T.M. Makris, S.G. Sligar, I. Schlichting, Structure and chemistry of cytochrome P450, Chemical reviews, 105 (2005) 2253-2277. [4] D.R. Nelson, L. Koymans, T. Kamataki, J.J. Stegeman, R. Feyereisen, D.J. Waxman, M.R. Waterman, O. Gotoh, M.J. Coon, R.W. Estabrook, I.C. Gunsalus, D.W. Nebert, P450 superfamily: update on new sequences, gene mapping, accession numbers and nomenclature, Pharmacogenetics, 6 (1996) 1-42. [5] S. Rendic, F.J. Di Carlo, Human cytochrome P450 enzymes: a status report summarizing their reactions, substrates, inducers, and inhibitors, Drug metabolism reviews, 29 (1997) 413-580. [6] D.A. Smith, B.C. Jones, Speculations on the substrate structure-activity relationship (SSAR) of cytochrome P450 enzymes, Biochemical pharmacology, 44 (1992) 2089-2098. [7] E.F. Johnson, Mapping determinants of the substrate selectivities of P450 enzymes by site-directed mutagenesis, Trends in pharmacological sciences, 13 (1992) 122-126. [8] F.J. Gonzalez, D.W. Nebert, Evolution of the P450 gene superfamily: animal-plant 'warfare', molecular drive and human genetic differences in drug oxidation, Trends Genet., 6 (1990) 182-186. [9] M. Ingelman-Sundberg, Pharmacogenetics of cytochrome P450 and its applications in drug therapy: the past, present and future, Trends in pharmacological sciences, 25 (2004) 193-200. [10] W.E. Evans, H.L. McLeod, Pharmacogenomics--drug disposition, drug targets, and side effects, The New England journal of medicine, 348 (2003) 538-549. [11] O. Gotoh, Substrate recognition sites in cytochrome P450 family 2 (CYP2) proteins inferred from comparative analyses of amino acid and coding nucleotide sequences, The Journal of biological chemistry, 267 (1992) 83-90. [12] S. Sansen, J.K. Yano, R.L. Reynald, G.A. Schoch, K.J. Griffin, C.D. Stout, E.F. Johnson, Adaptations for the oxidation of polycyclic aromatic hydrocarbons exhibited by the structure of human P450 1A2, The Journal of biological chemistry, 282 (2007) 14348-14355. [13] P.A. Williams, J. Cosme, A. Ward, H.C. Angove, D. Matak Vinkovic, H. Jhoti, Crystal structure of human cytochrome P450 2C9 with bound warfarin, Nature, 424 (2003) 464-468. [14] M.R. Wester, J.K. Yano, G.A. Schoch, C. Yang, K.J. Griffin, C.D. Stout, E.F. Johnson, The structure of human cytochrome P450 2C9 complexed with flurbiprofen at 2.0-A resolution, The Journal of biological chemistry, 279 (2004) 35630-35637.

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

[15] R.L. Reynald, S. Sansen, C.D. Stout, E.F. Johnson, Structural Characterization of Human Cytochrome P450 2C19: Active Site Differences Between P450s 2C8, 2C9, and 2C19, The Journal of biological chemistry, 287 (2012) 44581-44591. [16] P. Rowland, F.E. Blaney, M.G. Smyth, J.J. Jones, V.R. Leydon, A.K. Oxbrow, C.J. Lewis, M.G. Tennant, S. Modi, D.S. Eggleston, R.J. Chenery, A.M. Bridges, Crystal structure of human cytochrome P450 2D6, The Journal of biological chemistry, 281 (2006) 7614-7622. [17] P.A. Williams, J. Cosme, D.M. Vinkovic, A. Ward, H.C. Angove, P.J. Day, C. Vonrhein, I.J. Tickle, H. Jhoti, Crystal structures of human cytochrome P450 3A4 bound to metyrapone and progesterone, Science, 305 (2004) 683-686. [18] J.K. Yano, M.R. Wester, G.A. Schoch, K.J. Griffin, C.D. Stout, E.F. Johnson, The structure of human microsomal cytochrome P450 3A4 determined by X-ray crystallography to 2.05-A resolution, The Journal of biological chemistry, 279 (2004) 38091-38094. [19] M. Ekroos, T. Sjogren, Structural basis for ligand promiscuity in cytochrome P450 3A4, Proceedings of the National Academy of Sciences of the United States of America, 103 (2006) 13682-13687. [20] M. Otyepka, J. Skopalik, E. Anzenbacherova, P. Anzenbacher, What common structural features and variations of mammalian P450s are known to date?, Biochimica et biophysica acta, 1770 (2007) 376-389. [21] T.L. Poulos, Structural and functional diversity in heme monooxygenases, Drug metabolism and disposition: the biological fate of chemicals, 33 (2005) 10-18. [22] E.F. Johnson, C.D. Stout, Structural diversity of human xenobiotic-metabolizing cytochrome P450 monooxygenases, Biochemical and biophysical research communications, 338 (2005) 331-336. [23] E.E. Scott, Y.A. He, M.R. Wester, M.A. White, C.C. Chin, J.R. Halpert, E.F. Johnson, C.D. Stout, An open conformation of mammalian cytochrome P450 2B4 at 1.6-A resolution, Proceedings of the National Academy of Sciences of the United States of America, 100 (2003) 13196-13201. [24] E.E. Scott, M.A. White, Y.A. He, E.F. Johnson, C.D. Stout, J.R. Halpert, Structure of mammalian cytochrome P450 2B4 complexed with 4-(4chlorophenyl)imidazole at 1.9-A resolution: insight into the range of P450 conformations and the coordination of redox partner binding, The Journal of biological chemistry, 279 (2004) 27294-27301. [25] Y. Zhao, M.A. White, B.K. Muralidhara, L. Sun, J.R. Halpert, C.D. Stout, Structure of microsomal cytochrome P450 2B4 complexed with the antifungal drug bifonazole: insight into P450 conformational plasticity and membrane interaction, The Journal of biological chemistry, 281 (2006) 5973-5981. [26] J.A. Goldstein, S.M. de Morais, Biochemistry and molecular biology of the human CYP2C subfamily, Pharmacogenetics, 4 (1994) 285-299. [27] J.A. Goldstein, M.B. Faletto, M. Romkes-Sparks, T. Sullivan, S. Kitareewan, J.L. Raucy, J.M. Lasker, B.I. Ghanayem, Evidence that CYP2C19 is the major (S)mephenytoin 4'-hydroxylase in humans, Biochemistry, 33 (1994) 1743-1752. [28] X.Q. Li, T.B. Andersson, M. Ahlstrom, L. Weidolf, Comparison of inhibitory effects of the proton pump-inhibiting drugs omeprazole, esomeprazole, lansoprazole, pantoprazole, and rabeprazole on human cytochrome P450 activities, Drug metabolism and disposition: the biological fate of chemicals, 32 (2004) 821-827. [29] R.E. Pearce, A.D. Rodrigues, J.A. Goldstein, A. Parkinson, Identification of the human P450 enzymes involved in lansoprazole metabolism, The Journal of pharmacology and experimental therapeutics, 277 (1996) 805-816.

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

[30] M. Bajpai, L.K. Roskos, D.D. Shen, R.H. Levy, Roles of cytochrome P4502C9 and cytochrome P4502C19 in the stereoselective metabolism of phenytoin to its major metabolite, Drug metabolism and disposition: the biological fate of chemicals, 24 (1996) 1401-1403. [31] J.O. Miners, D.J. Birkett, Cytochrome P4502C9: an enzyme of major importance in human drug metabolism, British journal of clinical pharmacology, 45 (1998) 525538. [32] F. Jung, K.J. Griffin, W. Song, T.H. Richardson, M. Yang, E.F. Johnson, Identification of amino acid substitutions that confer a high affinity for sulfaphenazole binding and a high catalytic efficiency for warfarin metabolism to P450 2C19, Biochemistry, 37 (1998) 16270-16279. [33] S.J. Gardiner, E.J. Begg, Pharmacogenetics, drug-metabolizing enzymes, and clinical practice, Pharmacological reviews, 58 (2006) 521-590. [34] T. Yamashita, Y. Hoashi, K. Watanabe, Y. Tomisugi, Y. Ishikawa, T. Uno, Roles of heme axial ligands in the regulation of CO binding to CooA, The Journal of biological chemistry, 279 (2004) 21394-21400. [35] T.Z. Attia, T. Yamashita, M. Miyamoto, A. Koizumi, Y. Yasuhara, J.-i. Node, Y. Erikawa, Y. Komiyama, C. Horii, M. Yamada, Comparison of cytochrome p450 mediated metabolism of three central nervous system acting drugs, Chem. Pharm. Bull., 60 (2012) 1544-1549. [36] J.H. Dawson, L.A. Andersson, M. Sono, Spectroscopic investigations of ferric cytochrome P-450-CAM ligand complexes. Identification of the ligand trans to cysteinate in the native enzyme, The Journal of biological chemistry, 257 (1982) 3606-3617. [37] Y. Imai, R. Sato, Substrate interaction with hydroxylase system in liver microsomes, Biochemical and biophysical research communications, 22 (1966) 620626. [38] H. Remmer, J. Schenkman, R.W. Estabrook, H. Sasame, J. Gillette, S. Narasimhulu, D.Y. Cooper, O. Rosenthal, Drug interaction with hepatic microsomal cytochrome, Molecular pharmacology, 2 (1966) 187-190. [39] T.G. Spiro, X.-Y. Li, Biological Applications of Raman Spectroscopy, (1988) (Spiro, T. G., ed.) Vol. 3, pp. 1–38, Wiley-Interscience, New York. [40] J.R. Kincaid, Academic Press, San Diego, 2000. [41] T.L. Poulos, B.C. Finzel, A.J. Howard, Crystal structure of substrate-free Pseudomonas putida cytochrome P-450, Biochemistry, 25 (1986) 5314-5322. [42] T. Uno, Y. Nishimura, M. Tsuboi, R. Makino, T. Iizuka, Y. Ishimura, Two types of conformers with distinct Fe-C-O configuration in the ferrous CO complex of horseradish peroxidase. Resonance Raman and infarared spectroscopic studies with native and deuteroheme-substituted enzymes, The Journal of biological chemistry, 262 (1987) 4549-4556. [43] G.B. Ray, X.-Y. Li, J.A. Ibers, J.L. Sessler, T.G. Spiro, How Far Can Proteins Bend the FeCO Unit? Distal Polar and Steric Effects in Heme Proteins and Models, J. Am. Chem. Soc., 116 (1994) 162-176. [44] N.-T. Yu, E.A. Kerr, Biological Applications of Raman Spectroscopy, (1988) (Spiro, T. G. ed.) Vol. 3, pp. 39-95, Wiley-Interscience, New York. [45] T. Uno, Y. Nishimura, R. Makino, T. Iizuka, Y. Ishimura, M. Tsuboi, The resonance Raman frequencies of the Fe-CO stretching and bending modes in the CO complex of cytochrome P-450cam, The Journal of biological chemistry, 260 (1985) 2023-2026.

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

[46] T.J. Deng, L.M. Proniewicz, J.R. Kincaid, H. Yeom, I.D. Macdonald, S.G. Sligar, Resonance Raman studies of cytochrome P450BM3 and its complexes with exogenous ligands, Biochemistry, 38 (1999) 13699-13706. [47] T. Tosha, N. Kagawa, T. Ohta, S. Yoshioka, M.R. Waterman, T. Kitagawa, Raman evidence for specific substrate-induced structural changes in the heme pocket of human cytochrome P450 aromatase during the three consecutive oxygen activation steps, Biochemistry, 45 (2006) 5631-5640. [48] T.K. Das, S. Dewilde, J.M. Friedman, L. Moens, D.L. Rousseau, Multiple active site conformers in the carbon monoxide complexes of trematode hemoglobins, The Journal of biological chemistry, 281 (2006) 11471-11479. [49] G.C. Ibeanu, B.I. Ghanayem, P. Linko, L. Li, L.G. Pederson, J.A. Goldstein, Identification of residues 99, 220, and 221 of human cytochrome P450 2C19 as key determinants of omeprazole activity, The Journal of biological chemistry, 271 (1996) 12496-12501. [50] L.M. Podust, T.L. Poulos, M.R. Waterman, Crystal structure of cytochrome P450 14alpha -sterol demethylase (CYP51) from Mycobacterium tuberculosis in complex with azole inhibitors, Proceedings of the National Academy of Sciences of the United States of America, 98 (2001) 3068-3073. [51] J.R. Cupp-Vickery, C. Garcia, A. Hofacre, K. McGee-Estrada, Ketoconazoleinduced conformational changes in the active site of cytochrome P450eryF, Journal of molecular biology, 311 (2001) 101-110. [52] H.E. Seward, A. Roujeinikova, K.J. McLean, A.W. Munro, D. Leys, Crystal structure of the Mycobacterium tuberculosis P450 CYP121-fluconazole complex reveals new azole drug-P450 binding mode, The Journal of biological chemistry, 281 (2006) 39437-39443. [53] K.J. McLean, M.R. Cheesman, S.L. Rivers, A. Richmond, D. Leys, S.K. Chapman, G.A. Reid, N.C. Price, S.M. Kelly, J. Clarkson, W.E. Smith, A.W. Munro, Expression, purification and spectroscopic characterization of the cytochrome P450 CYP121 from Mycobacterium tuberculosis, Journal of inorganic biochemistry, 91 (2002) 527-541. [54] R. Raag, T.L. Poulos, The structural basis for substrate-induced changes in redox potential and spin equilibrium in cytochrome P-450CAM, Biochemistry, 28 (1989) 917-922. [55] M.T. Fisher, S.G. Sligar, Control of Heme Protein Redox Potential and Reduction Rate: Linear Free Energy Relation between Potential and Ferric Spin State Equilibrium, J. Am. Chem. Soc., 107 (1985) 5018-5019. [56] T. Iwamoto, Y. Kagawa, Y. Naito, S. Kuzuhara, M. Okuda, Clinical evaluation of plasma free phenytoin measurement and factors influencing its protein binding, Biopharmaceutics & drug disposition, 27 (2006) 77-84. [57] H. Takahashi, H. Echizen, Pharmacogenetics of warfarin elimination and its clinical implications, Clinical pharmacokinetics, 40 (2001) 587-603. [58] L.S. Kaminsky, Z.Y. Zhang, Human P450 metabolism of warfarin, Pharmacology & therapeutics, 73 (1997) 67-74.

ACCEPTED MANUSCRIPT Table I. Parameters for the drug-binding to CYP2C19. peak/trough a

∆/mM-1 cm-1

1010 ± 130

384/418 (I)

51.4

131 ± 24

386/419 (I)

30.2

5400 ± 110

385/416 (I)

22.6

Omeprazole

30 ± 3

386/419 (I)

18.2

Lansoprazole

2.6 ± 0.2

385/416 (I)

17.0

Phenytoin c

15 ± 2

416/387 (rI)

Cimetidine

1610 ± 120

427/397 (II)

S-Mephenytoin b Warfarin

a

RI

Propranolol

PT

Kd/ µM

Drug

-13.5

The peak and trough wavelength (nm) are given. Spectral type is given in

SC

parentheses: I, type I; rI, reverse type I; II, type II. Parameters for the first binding step are given.

c

Parameters for the second binding step are given.

NU

b

MA

Table 2. Effect of drug-binding on the (Fe-CO) band of CYP2C19. (Fe-CO)/cm-1

W1/2h (cm-1) a

None

471

22

Propranolol

471

S-Mephenytoin

472

PT E

Warfarin Omeprazole

Cimetidine

CE

Lansoprazole Phenytoin

D

Drug

29 20

474

24

487

35

477

32

471

24

471

25

W1/2h is full width at half height of the (Fe-CO) stretching band.

AC

a

-9.7

ACCEPTED MANUSCRIPT Highlights



Although cytochrome P450 2C19 (CYP2C19) isoform is a minor hepatic CYP, it metabolizes clinically important drugs.



In this work, the interaction of purified CYP2C19 WT with seven drugs was



PT

investigated using spectroscopic methods. The binding of these drugs induced a variety of structural change in the heme

RI

distal environment with different degree of water removal dependent on the type of drugs

CE

PT E

D

MA

NU

affect the drug specificities of the enzyme.

SC

In addition, it was found that slight chemical modifications of the drug could

AC



Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6