Communication

Cite This: Biochemistry 2017, 56, 5476-5480 pubs.acs.org/biochemistry

Structural Basis of Single-Nucleotide Polymorphisms in Cytochrome

P450 2C9
Keiko Maekawa,*,, Motoyasu Adachi, Yumiko Matsuzawa, Qinghai Zhang, Ryota Kuroki,,
Yoshiro Saito, and Manish B. Shah*,#,@

Department of Analytical Chemistry, Faculty of Pharmaceutical Sciences, Doshisha Womens College of Liberal Arts, Kodo,
Kyotanabe, Kyoto 610-0395, Japan

Division of Medicinal Safety Science, National Institute of Health Sciences, Kamiyoga 1-18-1, Setagaya, Tokyo 158-8501, Japan

National Institutes for Quantum and Radiological Science and Technology, 2-4 Shirane Shirakata, Tokai-mura, Ibaraki 319-1106,
Japan

Department of Integrative Structural and Computational Biology, The Scripps Research Institute, 10550 North Torrey Pines Road,
La Jolla, California 92037, United States

Japan Atomic Energy Agency, 2-4 Shirane Shirakata, Tokai-mura, Ibaraki 319-1195, Japan
#
Department of Pharmaceutical Sciences, University of Connecticut, Storrs, Connecticut 06267, United States
@
Department of Pharmaceutical Sciences, Albany College of Pharmacy and Health Sciences, 106 New Scotland Avenue, Albany, New
York 12208, United States
*
S Supporting Information

Over 700 alleles of CYPs in various ethnic populations have

ABSTRACT: Single-nucleotide polymorphisms in drug- been identied and examined via genetic testing, with many
metabolizing cytochrome P450 (CYP) enzymes are having a signicant impact on enzyme activity and drug
important contributors to interindividual dierences in metabolism.4,5
drug metabolism leading to adverse drug reactions. The highly polymorphic CYP2C9 enzyme metabolizes up to
Despite their extensive characterization and importance 1518% of the drugs that include warfarin, tolbutamide and
in pharmacogenetics of clinical drugs, the structural basis glimepiride, phenytoin, urbiprofen and diclofenac, and
of CYP polymorphisms has remained scant. Here we losartan.68 Sixty CYP2C9 alleles have been reported until
report the crystal structures of human CYP2C9 and its now, and many exhibited altered activities compared to that of
polymorphic variants, *3 (I359L) and *30 (A477T), with the wild-type (WT) enzyme (Table S1).9 CYP2C9 *3 (I359L)
an antihypertensive drug losartan. The structures show was associated with reduced catalytic activities toward several
distinct interaction and occupation of losartan in the active substrates such as warfarin, tolbutamide, and losartan,10,11
site, the access channel, and the peripheral binding site. whereas the CYP2C9 *30 (A477T) variant demonstrated loss
The I359L substitution located far from the active site of antihypertensive eects for losartan.12,13 Both CYP2C9 *3
remarkably altered the residue side chains near the active and *30 variants showed 77 and 99% reduced activity,
site and the access channel, whereas the T477 substitution respectively, with losartan compared to that of the WT enzyme
illustrated hydrogen-bonding interaction with the reor- in vitro.
iented side chain of Q214. The results yield structural Despite the huge amount of eort to understand genetic
insights into the reduced catalytic activity of the CYP2C9 variations and dierences in drug response, and with more than
variants and have important implications for under-
100 structures of multiple human CYPs published in the
standing genetic polymorphisms in CYP-mediated drug
presence or absence of various ligands, the structural basis of
metabolism.
SNPs illustrating the eect of such amino acid substitutions is
still lacking. Allelic variants are often dicult to purify and
crystallize, largely because the level of expression, yield, or
A dverse drug reactions result in more than half a million
injuries and deaths and cost millions of dollars per year in
the United States alone.1 Identifying human genetic poly-
stability is lower than that of their WT counterparts. Moreover,
structurally it might be dicult to comprehend the diminished
in vitro activity of expressed variants because the amino acid
morphisms responsible for pharmacokinetic and pharmacody- changes in the protein occur in the region not directly located
namic dierences between individuals could lead to develop- near the active site or the substrate recognition site. In this
ment of safer medication specic to an individuals DNA trait, study, we report the crystal structures of CYP2C9 and its two
which is often termed personalized medicine. Single-nucleotide important allelic variants, *3 and *30, in complex with an
polymorphisms (SNPs) in drug-metabolizing cytochrome P450
(CYP) enzymes, which constitute the major enzyme family in Received: August 16, 2017
drug metabolism with increasing importance in pharmacoge- Revised: September 16, 2017
netics, are key determinants of variability in drug response.2,3 Published: October 3, 2017

2017 American Chemical Society 5476 DOI: 10.1021/acs.biochem.7b00795

Biochemistry 2017, 56, 54765480
Biochemistry Communication

angiotensin II receptor antagonist losartan (Table S2). where a palmitic acid is forming hydrogen bond in the same
Losartan, a prodrug used mainly as an antihypertensive agent, region of residues from P227 to T229 as shown in Figure
is primarily oxidized in the liver by CYP2C9 and CYP3A4 to a S3A.17 The conserved residues FPGT from position 226 to 229
more potent E-3174 (Figures S1 and S2).14 illustrate a signature sequence in human CYP2C8, -2C9, and
The CYP2C9 WT and the *30 variant demonstrated binding -2C19 and observation of ligands bound at this site in multiple
of three molecules of losartan, with one at the peripheral site, CYP2C structures suggest a potential role of such a site in
another in the active site, and the third in the access channel. substrate recognition. Across the larger CYP2 family of
Whereas the *3 variant, with only two molecules bound, lacked enzymes, the residue at position 226 is variable hydrophobic,
the losartan in the access channel and exhibited an altered at 227 is conserved proline, at position 228 is either glycine or
binding mode in the active site (Figure 1). The losartan binding alanine, and position 229 includes either a hydrophobic or
hydrophilic substitution. In addition, the peripheral binding site
of losartan is near the surface binding site of progesterone in
CYP3A4, located in the Phe cluster of residues in the FG
cassette (Figure S3B).18 The F and G helices are longer in
CYP2C9, which lacks such a hydrophobic cluster.
The active site losartan in all the CYP2C9 structures make
two important polar interactions with the residue side chains of
R108 and N204. The presence of R108 in the active site is
important for the binding anity of the potent CYP2C9
inhibitor benzbromarone and other ionized phenols.19 R108,
known to be crucial in the formation of 4-hydroxydiclofe-
nac,20,21 rotates by 90 into the active site in the current
CYP2C9 complexes compared with the previously determined
CYP2C9 structures (Figure S4A).22 The guanidino group of
R108 forms a hydrogen bond with the tetrazole ring and the
hydroxyl of the imidazole ring in losartan, thereby stabilizing
substrate binding near the heme iron. The active site losartan is
signicantly altered in the *3 complex; however, the polar
Figure 1. Structures of CYP2C9 and allelic variants in complex with contacts between losartan and the side chains of R108 and
losartan. (A) Structure of CYP2C9 WT (yellow) complexed with three N204 are maintained in a similar fashion as observed with the
molecules of losartan (orange sticks). (B) Structure of CYP2C9 *3 WT and *30 complexes. The biphenyl rings of losartan in the
(orange) with two molecules of losartan (cyan sticks). (C) Structure of
*3 active site ip by 90, which is now 7 from the heme iron
CYP2C9 *30 (green) in complex with three losartan molecules (pink
sticks) in an orientation and location similar to that of the WT in place of the 5 distance in the other two losartan complexes.
complex. Heme is shown as red sticks. The location of each losartan is In addition to these polar contacts, the residue side chains of
represented by red numbers: 1, peripheral site; 2, active site; and 3, V113, F114, V237, and V292 were involved in hydrophobic
access channel. interactions with the active site losartan in all the three
structures as observed with other ligand-bound structures of
site at the periphery in all three complexes and the access CYP2C9.2224 Importantly, the site of hydroxylation of losartan
channel in the WT and *30 complexes are located at previously in the active site of CYP2C9, irrespective of WT, *3, and *30, is
predicted substrate recognition sites (SRS) 3, and 5 and 6, positioned away from the heme iron, in an orientation not
respectively, in the CYP2 family of enzymes.15,16 consistent with the demonstrated metabolism. Such an altered
The structural overlay of all the three complexes [root-mean- pose of the substrate has been observed in numerous
square deviation (RMSD) of 0.20.25 ] revealed the crystallographic structures of CYP enzymes.25,26 Because
residues located within 5 of losartan bound on the periphery substrate binding occurs during the rst step in the catalytic
(Figure 2). The tetrazole ring of losartan is located between cycle, it is possible for the substrate to reorient in subsequent
residues F226, P227, G228, and T229, whereas the Cl of the steps during the electron transfer from NADPH-cytochrome
losartan interacts with the side chain of K232. Interestingly, this P450 reductase (CPR). Moreover, the crystallization experi-
peripheral site is similar to that observed in a CYP2C8 structure ments are performed in the absence of NADPH or CPR.
Additional studies are warranted to elucidate the role of these
redox partners in reorienting the ligand in a conformation that
is suitable for metabolism. The water molecule is ordered by its
interactions with the heme iron in all three structures.
In contrast to the *3 complex, the structures of CYP2C9 WT
and *30 illustrated binding of an additional losartan in the
substrate access channel. The third losartan in the WT and the
*30 complex was observed in two similar orientations that
could be superimposed on each other. In each of the two
structures, the polar Q214 and N218 residue side chains are
positioned near the imidazole ring of losartan, while the
tetrazole of losartan contacts the polar side chain of T364
Figure 2. Structural overlay of CYP2C9losartan complexes (Figure S4B) and main chain oxygens of S365 and F476. Apart
representing interactions of peripheral losartan (sticks) with the from the residues mentioned above, a majority of residues
amino acid residue side chains (sticks) located within 5 . surrounding the losartan are hydrophobic. In particular, the side
5477 DOI: 10.1021/acs.biochem.7b00795
Biochemistry 2017, 56, 54765480
Biochemistry Communication

chain of F100 and F476 -stacks with the imidazole and the
phenyl ring, respectively, of the access channel losartan found
in the WT and *30 complexes. Such -stacking interactions by
F100 and F476 side chains are also seen in the warfarin
complex,27 where the only molecule bound in the structure was
in this access channel region.
Most importantly, the eect of SNP that results in amino
acid substitution from isoleucine to leucine at position 359
located on the K helix in the CYP2C9 *3 complex is transduced
on the I helix from residue 307 to 311 (Figure 3). The side

Figure 4. SNP in CYP2C9 *30. The *30 complex (green) associated

with the A477T substitution on the 4 loop is superimposed on the
CYP2C9 WT complex (yellow) with losartan. The Q214 side chain in
the WT complex interacts with the imidazole ring of losartan and
rotates 180 in the *30 complex to hydrogen bond (yellow dashes)
with the polar threonine substitution at position 477. Losartans
(spheres) are numbered in red.

Interestingly, the same side chain in the *30 variant rotates

more than 90 from the losartan activation site to hydrogen
bond with the hydrophilic substitution T477 (Figure S5B). The
Figure 3. SNP in CYP2C9 *3 and its inuence on substrate binding. results suggest the possible role of Q214 that may be crucial for
Overlay of CYP2C9 WT (yellow) and *3 (orange) complexed with activation of a prodrug losartan by CYP2C9. Indeed,
losartan. The eect of the distal I359L substitution is transduced to the CYP2C9*28 (Q214L) exhibited 87% decreased activity for
neighboring I helix residues. The Y308 side chain is displaced by >3 losartan oxidation compared with that of the WT.12 Of note, in
(black dashes), subsequently aecting the 4 loop that is crucial for the previous structures, residue Q214 was either located in the
substrate interaction near the active site. Losartans shown as spheres region containing engineered amino acids for crystallization or
are numbered in red. Part of the protein image has been omitted for not modeled because of disordered electron density.22,27,31,32
the sake of clarity. Moreover, the structures of CYP2C9 in complex with
losartan are signicantly dierent from the previously reported
chain of Y308 with the greatest inuence is pushed toward the CYP2C9 structures.22,27,31,32 All the published CYP2C9
4 loop by >3 , thereby aecting the secondary structural structures until now (August 2017) either contained multiple
elements and the orientation of multiple residues. Such an stabilizing amino acid substitutions introduced into the FG
eect of I359L substitution more than 15 from the active site loop for crystallization purposes or had disordered or missing
signicantly aects the important 4 loop involved in substrate electron density of crucial amino acids in that region near the
interaction in the other losartan complexes. The disorder of the access channel (Figure S6). Although these residues are outside
4 loop leads to an altered orientation of multiple residue side the active site, it is evident that such engineering or poorly
chains. Most importantly, F476, demonstrated to be important ordered residues aect ligand binding and protein conforma-
in -stacking with losartan and various substrates,28,29 rotates tions in this FG region that were previously attributed to
180 and protrudes into the access channel where the third substrate specicity and enzymatic activity.33 Furthermore, the
losartan is bound in the WT and *30 complexes. The *3 allele structural overlay of the CYP2C9losartan complex with the
is the most functionally impaired of all the CYP2C9 genetic published urbiprofen (RMSD 0.41 ) and warfarin (RMSD
variants and confers up to 80 and 77% decreases in catalytic 0.47 ) complexes of CYP2C9 revealed dierences and
eciency toward glimepiride and losartan, respectively.12 The similarities of ligand binding. The active site urbiprofen
signicant structural change transduced from the I359L superimposes with losartan in the active site, whereas warfarin
substitution on the K helix to the rotation of F476 on the 4 that -stacks with F476 overlays with the access channel
loop that blocks the access channel (Figure S5A) may help us losartan in the CYP2C9 complex (Figure S6A,B). This is
understand the loss of activity of not only losartan but also consistent with previous ndings that the interactions of
many other important substrates, including urbiprofen and dierent ligands with CYP2C9 involve multiple substrate
warfarin, that are not metabolized by this defective allele. binding modes and sites.34,35
Furthermore, a recent study demonstrated that the loss of Overall, the structures are the rst representation of allelic
CYP2C9 activity in these individuals is not compensated by variants in the CYP superfamily of enzymes that illustrate the
switching to another CYP or metabolism pathway, further implications of polymorphism upon drug binding and provide a
complicating clearance of the drug.30 useful framework to infer altered drug metabolism by genetic
Structural analysis of the CYP2C9 *30losartan complex variants. These structures elucidate the eects of *3 and *30
with a hydrophobic alanine to polar threonine substitution at and also provide crucial insight into interpreting the *4
position 477 on the 4 loop revealed signicant reorientation of (I359T), *18 (I359L, D397A), and *28 (Q214L) variants that
an adjacent Q214, which promotes the formation of a hydrogen demonstrated reduced activity toward various substrates.3638
bond between these two residue side chains (Figure 4). The It is likely that the I359T substitution in *4 perturbs the
side chain of Q214 in the CYP2C9 WT structure interacts with residues on the I helix and the eect is transduced to the active
the imidazole ring of the access channel losartan near the site in a fashion similar to that in *3 with I359L. The *18
hydroxyl moiety, the potential activation site of losartan. variant includes amino acid substitution as in *3, whereas *28
5478 DOI: 10.1021/acs.biochem.7b00795
Biochemistry 2017, 56, 54765480
Biochemistry Communication

with the hydrophobic substitution of leucine at position 214 (4) Sim, S. C., and Ingelman-Sundberg, M. (2010) Hum. Genomics 4,
may preclude or dier in interactions with the substrate 278281.
compared to the polar glutamine as seen in the CYP2C9 (5) de Leon, J., Susce, M. T., and Murray-Carmichael, E. (2006) Mol.
losartan WT complex. Moreover, the ndings of multiple Diagn. Ther. 10, 135151.
(6) Rettie, A. E., and Jones, J. P. (2005) Annu. Rev. Pharmacol.
substrate binding sites in each of these structures suggest the
Toxicol. 45, 477494.
possible role of cooperativity or allosteric regulation by human (7) Zanger, U. M., Turpeinen, M., Klein, K., and Schwab, M. (2008)
drug-metabolizing enzymes. It remains to be elucidated Anal. Bioanal. Chem. 392, 10931108.
whether losartan bound at the peripheral site triggers the (8) Miners, J. O., and Birkett, D. J. (1998) Br. J. Clin. Pharmacol. 45,
conformational change and move to the active site or if the 525538.
binding occurs in a sequential fashion with one losartan binding (9) Sim, S. C., and Ingelman-Sundberg, M. (2013) Methods Mol. Biol.
at the peripheral site followed by another in the active site via 987, 251259.
the access channel. Indeed, the structures will facilitate (10) Goldstein, J. A. (2001) Br. J. Clin. Pharmacol. 52, 349355.
computational approaches for predictions of their eects in (11) Joy, M. S., Dornbrook-Lavender, K., Blaisdell, J., Hilliard, T.,
drug clearance. Such information would be very useful for not Boyette, T., Hu, Y., Hogan, S. L., Candiani, C., Falk, R. J., and
only obtaining mechanistic insights into how the allelic Goldstein, J. A. (2009) Eur. J. Clin. Pharmacol. 65, 947953.
(12) Maekawa, K., Harakawa, N., Sugiyama, E., Tohkin, M., Kim, S.
variations lead to altered catalytic activities toward one drug
R., Kaniwa, N., Katori, N., Hasegawa, R., Yasuda, K., Kamide, K.,
but also evaluating their eects toward drugdrug interactions Miyata, T., Saito, Y., and Sawada, J. (2009) Drug Metab. Dispos. 37,
especially when designing candidates in drug discovery as part 18951903.
of the lead optimization process.

(13) Yin, T., Maekawa, K., Kamide, K., Saito, Y., Hanada, H.,
Miyashita, K., Kokubo, Y., Akaiwa, Y., Otsubo, R., Nagatsuka, K.,
ASSOCIATED CONTENT Otsuki, T., Horio, T., Takiuchi, S., Kawano, Y., Minematsu, K.,
*
S Supporting Information Naritomi, H., Tomoike, H., Sawada, J., and Miyata, T. (2008)
The Supporting Information is available free of charge on the Hypertens. Res. 31, 15491557.
ACS Publications website at DOI: 10.1021/acs.bio- (14) Sica, D., Gehr, T., and Ghosh, S. (2005) Clin. Pharmacokinet. 44,
797814.
chem.7b00795. (15) Gotoh, O. (1992) J. Biol. Chem. 267, 8390.
Experimental details, gures, tables, and methods (PDF) (16) Berka, K., Hendrychova, T., Anzenbacher, P., and Otyepka, M.

(2011) J. Phys. Chem. A 115, 1124811255.

AUTHOR INFORMATION (17) Schoch, G. A., Yano, J. K., Wester, M. R., Griffin, K. J., Stout, C.
D., and Johnson, E. F. (2004) J. Biol. Chem. 279, 94979503.
Corresponding Authors (18) Williams, P. A., Cosme, J., Vinkovic, D. M., Ward, A., Angove,
*E-mail: manish.shah@acphs.edu (structural analysis). H. C., Day, P. J., Vonrhein, C., Tickle, I. J., and Jhoti, H. (2004) Science
*E-mail: kmaekawa@dwc.doshisha.ac.jp (protein isolation, 305, 683686.
crystallization). (19) Locuson, C. W., 2nd, Wahlstrom, J. L., Rock, D. A., Rock, D. A.,
ORCID and Jones, J. P. (2003) Drug Metab. Dispos. 31, 967971.
(20) Ridderstrom, M., Masimirembwa, C., Trump-Kallmeyer, S.,
Manish B. Shah: 0000-0001-8593-6446 Ahlefelt, M., Otter, C., and Andersson, T. (2000) Biochem. Biophys.
Funding Res. Commun. 270, 983.
This work was supported by Grants KAKENHI22590054 and (21) Dickmann, L. J., Locuson, C. W., Jones, J. P., and Rettie, A. E.
KAKENHI25293128 from the Japan Society for the Promotion (2004) Mol. Pharmacol. 65, 842850.
of Science to K.M. and National Institutes of Health Grant R01 (22) Wester, M. R., Yano, J. K., Schoch, G. A., Yang, C., Griffin, K. J.,
GM098538 to Q.Z. The research was performed under the Stout, C. D., and Johnson, E. F. (2004) J. Biol. Chem. 279, 35630
35637.
approval of the Photon Factory Program Advisory Committee
(23) Haining, R. L., Jones, J. P., Henne, K. R., Fisher, M. B., Koop, D.
(Proposal 2013G597). R., Trager, W. F., and Rettie, A. E. (1999) Biochemistry 38, 32853292.
Notes (24) Tracy, T. S., Hutzler, J. M., Haining, R. L., Rettie, A. E.,
The authors declare no competing nancial interest. Hummel, M. A., and Dickmann, L. J. (2002) Drug Metab. Dispos. 30,
385390.
Deceased.

(25) Ekroos, M., and Sjogren, T. (2006) Proc. Natl. Acad. Sci. U. S. A.
ACKNOWLEDGMENTS 103, 13682.
(26) DeVore, N. M., and Scott, E. E. (2012) J. Biol. Chem. 287,
The authors are grateful to Dr. James R. Halpert at the 2657626585.
University of Connecticut, United States, for supporting our (27) Williams, P. A., Cosme, J., Ward, A., Angove, H. C., Matak
collaboration.

Vinkovic, D., and Jhoti, H. (2003) Nature 424, 464468.

(28) Melet, A., Assrir, N., Jean, P., Pilar Lopez-Garcia, M., Marques-
ABBREVIATIONS Soares, C., Jaouen, M., Dansette, P. M., Sari, M. A., and Mansuy, D.
WT, wild-type; CYP, cytochrome P450; RMSD, root-mean- (2003) Arch. Biochem. Biophys. 409, 8091.
(29) Mosher, C. M., Hummel, M. A., Tracy, T. S., and Rettie, A. E.
square deviation; CPR, cytochrome P450 reductase; SNP, (2008) Biochemistry 47, 1172511734.
single-nucleotide polymorphism.

(30) Flora, D. R., Rettie, A. E., Brundage, R. C., and Tracy, T. S.

(2017) J. Clin. Pharmacol. 57, 382393.
REFERENCES (31) Branden, G., Sjogren, T., Schnecke, V., and Xue, Y. (2014) Drug
(1) Classen, D. C., Pestotnik, S. L., Evans, R. S., Lloyd, J. F., and Discovery Today 19, 905911.
Burke, J. P. (1997) JAMA, J. Am. Med. Assoc. 277, 301306. (32) Skerratt, S. E., Andrews, M., Bagal, S. K., Bilsland, J., Brown, D.,
(2) Meyer, U. A., Zanger, U., Skoda, R., and Grant, D. M. (1989) Bungay, P. J., Cole, S., Gibson, K. R., Jones, R., Morao, I., Nedderman,
Psychopharmacol Ser. 7, 141147. A., Omoto, K., Robinson, C., Ryckmans, T., Skinner, K., Stupple, P.,
(3) Daly, A. K. (2004) Curr. Top. Med. Chem. 4, 17331744. and Waldron, G. (2016) J. Med. Chem. 59, 1008410099.

5479 DOI: 10.1021/acs.biochem.7b00795

Biochemistry 2017, 56, 54765480
Biochemistry Communication

(33) Wada, Y., Mitsuda, M., Ishihara, Y., Watanabe, M., Iwasaki, M.,
and Asahi, S. (2008) J. Biochem. 144, 323333.
(34) Korzekwa, K. R., Krishnamachary, N., Shou, M., Ogai, A., Parise,
A., Rettie, A. E., Gonzalez, F. J., and Tracy, T. S. (1998) Biochemistry
37, 41374147.
(35) Locuson, C. W., Rock, D. A., and Jones, J. P. (2004)
Biochemistry 43, 69486958.
(36) Imai, J., Ieiri, I., Mamiya, K., Miyahara, S., Furuumi, H., Nanba,
E., Yamane, M., Fukumaki, Y., Ninomiya, H., Tashiro, N., Otsubo, K.,
and Higuchi, S. (2000) Pharmacogenetics 10, 8589.
(37) DeLozier, T. C., Lee, S. C., Coulter, S. J., Goh, B. C., and
Goldstein, J. A. (2005) J. Pharmacol. Exp. Ther. 315, 10851090.
(38) Maekawa, K., Fukushima-Uesaka, H., Tohkin, M., Hasegawa, R.,
Kajio, H., Kuzuya, N., Yasuda, K., Kawamoto, M., Kamatani, N.,
Suzuki, K., Yanagawa, T., Saito, Y., and Sawada, J. (2006)
Pharmacogenet. Genomics 16, 497514.

5480 DOI: 10.1021/acs.biochem.7b00795

Biochemistry 2017, 56, 54765480