Health Articles

 

Bio.gnu.kr

nature publishing group
The CYP3A4*18 Genotype in the Cytochrome P450 3A4 Gene, a Rapid Metabolizer of Sex Steroids, Is Associated With Low Bone Mineral Density
YS Kang1, SY Park1, CH Yim1, HS Kwak1, P Gajendrarao2, N Krishnamoorthy2, S-C Yun3,
KW Lee2 and KO Han1
Osteoporosis is influenced by genetic factors. The interindividual variability in the activity of CYP3A, the metabolic
enzyme of sex hormones, may result from genetic polymorphisms. In a study of 2,178 women of ages 40–79 years,
the presence of the CYP3A4*18 variant was found to be significantly associated with low bone mass. In vitro functional
analyses indicate that CYP3A4*18 is a gain-of-function mutation in sex steroid metabolism, resulting in rapid oxidation
of estrogens and testosterone; in vivo pharmacokinetics using midazolam (MDZ) verify the altered activity of the
CYP3A4*18, showing lower metabolic turnover in the mutant than in the wild type. Molecular modeling reveals the
structural changes in the substrate recognition sites of CYP3A4*18 that can cause changes in enzymatic activity and that
potentially account for the difference between the catalytic activities of estrogen and MDZ, depending on the genotype.
The results indicate that a genetic variation in the CYP3A4 gene—as a gain-of-function mutation in the metabolism of
certain CYP3A substrates, including sex steroids—may predispose individuals to osteoporosis.
Osteoporosis is a multifactorial disease with a strong genetic endogenous compounds, but genetic factors are also among the component. Genetic factors influence bone mass, bone size, most plausible mechanisms.
bone quality, and bone turnover, and they may modulate the The CYP3A activity of the adult human liver is the sum risk of osteoporosis.1 Many candidate genes have thus far been activity of at least two CYP3A family members: CYP3A4 and suggested, but none has yet been supported strongly and consist- CYP3A5. To date, approximately 40 allelic variants in the ently by subsequent studies.
CYP3A4 gene have been reported as showing marked ethnic The members of the cytochrome P450 3A (CYP3A) subfamily differences in allele frequencies.6,7 CYP3A5, the second-most are the major enzymes in the nicotinamide adenine dinucleotide important CYP3A protein in the liver, has characteristic poly- phosphate-oxidase–dependent oxidative metabolism of vari- morphic expression caused by genetic variation; certain genetic ous endogenous and exogenous compounds, including sex variations, such as CYP3A5*3 and CYP3A5*6, give rise to an hormones. A wide interindividual variability in the expres- aberrantly spliced mRNA with a premature stop codon, which sion and catalytic activity of CYP3A has been reported in the produces a nonfunctioning protein.8,9 general population.2 The interindividual variation, exceeding We therefore hypothesized that genetic variations of CYP3A 30-fold in some populations, may influence the circulating lev- proteins, the important metabolizing enzymes of estrogen, els of endogenous sex steroids and thereby mediate the risk of might be among the major determinants in the development certain estrogen-associated diseases such as osteoporosis.3–5 of osteoporosis. To identify the candidate genetic variations in This variation is, at least partly, caused by multiple environ- the CYP3A4 gene, we sequenced the entire coding region and mental factors, including induction by drugs, chemicals, and performed detailed structural and functional studies, including The first three authors contributed equally to this work.
1Department of Internal Medicine, Cheil General Hospital and Women's Healthcare Center, Kwandong University College of Medicine, Seoul, Korea; 2Division of Applied Life Sciences (BK21 Program), Environmental Biotechnology, National Core Research Center, Gyeongsang National University, Jinju, Korea; 3Department of Preventive Medicine, University of Ulsan College of Medicine, Seoul, Korea. Correspondence: KO Received 19 May 2008; accepted 11 September 2008; advance online publication 19 November 2008. VOLUME 85 NUMBER 3 MARCH 2009 www.nature.com/cpt
table 1 Clinical characteristics and bone density data of the subjects (n = 2,178)
Age at menopause (years) Spine BMD (g/cm2) Values are means ± SD.
BMD, bone mineral density; BMI, body mass index; YSM, years since menopause.
aP value adjusting for age, BMI, and CYP3A5. bP value adjusting for age, BMI, and CYP3A4.
both in vitro and in vivo analyses for candidate genotypes. We also assessed the CYP3A5*3 al eles that are known to contribute to the reduction of CYP3A5 activity. We analyzed the associa- tion of CYP3A genotypes with bone mineral density (BMD) in Korean women.
CYP3A4 and CYP3A5 genotyping analysis
A screening search for base changes in all 13 exons in the CYP3A4 gene in 225 Koreans identified two mutations: a T → Lumbar spine BMD (g/cm C point mutation (L293P) at codon 293, named CYP3A4*18, and a silent mutation (L295L) at codon 295. Among 2,178 Korean women of ages ranging from 40 to 79 years, CYP3A4*18 was detected in 53 women, with an allelic frequency of 1.2%: 2,125 were wild type (WT), 52 were heterozygotes, and 1 was a Years since menopause homozygote. For CYP3A5, the genotyping revealed that 62.7% of the subjects were the CYP3A5*3/*3 genotype and the allelic frequency of CYP3A5*3 was 79.5%. Significant linkage disequi- librium was shown between the two genotypes in our study (Dʹ = 0.80, P < 0.001), as presented in the previous data.10 CYP3A4 polymorphism and BMD
The clinical characteristics of the study groups in relation to CYP3A4 and CYP3A5 genotypes are listed The groups were all balanced with regard to clinical variables. Subjects with the CYP3A4*18 genotype were significantly associated with low lumbar spine BMD after adjusting for age, body mass index CYP3A4*1 CYP3A4*18 (BMI), and CYP3A5 genotype (P = 0.014). However, in analy- ses of BMD according to CYP3A5 genotype, no BMD difference Figure 1 The association of CYP3A4 genotype with lumbar and femur bone
was observed in women with deficient CYP3A5 activity as com- mineral density (BMD). (a) Relationship of lumbar spine BMD to years since
menopause (YSM), relative to CYP3A4 genotype, in 2,178 women of ages
pared to those with whole CYP3A5 activity, thereby indicating 40–79 years. The slope relating to the CYP3A4*18 genotype did not differ from that CYP3A5 has no significant influence on bone metabolism that of the CYP3A4*1 (wild type) genotype (P = 0.437). YSM values Subsequent analyses to assess the differences between <6 months were arbitrarily set to YSM = 0. (b) Comparison of femur BMD by
the CYP3A4 genotypes, adjusted for age, years since menopause, CYP3A4 genotype in 1,353 women in whom femur BMD measurements were BMI, or CYP3A5 genotype, did not elucidate any significant inter- performed using a QDR-2000. BMD comparisons were adjusted for age, body mass index, and CYP3A5 genotype.
action effects between the CYP3A4 genotype and each covariate. The relationship between lumbar spine BMD and years since subjects. Their BMD was measured by dual-energy X-ray absorp- menopause in subjects with the CYP3A4 genotype is shown in tiometry using the QDR-2000 Adjusted BMD for No significant differences in slope were found between the total hip was significantly lower in CYP3A4*18 subjects than genotypes. To assess the femoral BMD, we analyzed 1,353 in CYP3A4*1 subjects (P = 0.027). We also observed a tendency ClInICAl PhArMACOlOgY & TherAPeuTICs VOLUME 85 NUMBER 3 MARCH 2009
toward association of CYP3A4*18 with a low adjusted BMD at netic tests: metabolic turnover for MDZ was significantly lower the femoral neck, trochanter, and Ward's triangle; however, this in CYP3A4*18 than in CYP3A4 WT.
tendency did not reach statistical significance.
Molecular modeling construction for the CYP3A4
In vitro functional analysis of CYP3A4 enzyme activity
To compare CYP3A4 enzyme activity for metabolism of sex hor- To investigate the structural behavior of the proteins, molecu- mones in vitro, we obtained both types of recombinant enzymes lar modeling studies and molecular dynamic simulation were using CYP3A4*1 or CYP3A4*18 prepared in a baculovirus system. carried out, for the WT and CYP3A4*18 proteins. As shown Western blotting and CO spectra indicated that both were present both proteins have centrally located hemes; how- in the form of their holoproteins (data not shown). The levels of ever, there is a structural difference between the WT and the individual metabolites (16α-OHE1, 2-OHE1, and 4-OHE1) in the oxidations of estrone (E1) tended to be higher with the mutant enzyme, CYP3A4*18, than with the CYP3A4*1 enzyme ( The oxidative metabolite of testosterone, 6β-hydroxytestosterone, CYP3A4*1 CYP3A4*18 was present in significantly higher quantities by the CYP3A4*18 enzyme These results suggest that CYP3A4*18 has a higher catalytic efficiency for both estrogen and testosterone metabolism than the CYP3A4 WT enzyme does.
In vivo pharmacogenetics for CYP3A4 polymorphism
Midazolam concentration (ng/ml) Midazolam (MDZ) is the drug that is most widely used as an in vivo probe for phenotyping CYP3A activity.11,12 We carried out a pharmacokinetic test to compare MDZ metabolic activity in 13 CYP3A4*18+ subjects and 26 CYP3A4*1+ subjects, each matched for age, sex, BMI, and CYP3A5 genotypeThe P < 0.001 concentration–time plot revealed that the mean plasma concen- trations of MDZ after a 7.5 mg oral administration of MDZ tended to be lower in subjects with the CYP3A5*3 genotype, but that this difference was not statistical y significant (data not shown). On the other hand, plasma concentrations were significantly higher in CYP3A4*18+ subjects. The CYP3A4*18+ subjects showed diminished MDZ clearance and increased MDZ area under the Clearance (ml/min/kg) plasma concentration curve as compared to CYP3A4*1+ sub- Midazolam AUC (ng·h/ml) jects (P = 0.029, P < 0.001, respectively), thereby suggesting that CYP3A4*18 is associated with decreased catalytic activity for MDZ. To confirm the contradictory data resulting from the MDZ pharmacokinetic tests, we tested the in vitro MDZ kinetics with CYP3A4*1 CYP3A4*18 recombinant enzymes The maximum reaction veloc- ity value for 1-OH MDZ in the case of CYP3A4*1 was greater than the maximum reaction velocity value for CYP3A4*18. These results were consistent with the results from in vivo pharmacoki- table 2 estrone and testosterone hydroxylase activity with
CYP3A4*1 CYP3A4*18 Metabolite formation rate (min [Midazolam] (µmol/l) Figure 2 In vivo pharmacokinetics and in vitro enzyme kinetics to assess the
differences in midazolam (MDZ) oxidation relative to the CYP3A4 genotype. (a) The mean MDZ plasma concentration and (b) MDZ clearance and area under
the plasma concentration curve after a 7.5-mg oral dose of MDZ indicated that CYP3A4*18 is associated with decreased metabolism of MDZ as compared to CYP3A4*1. *P < 0.05, P values by linear mixed-effects model. (c) Kinetic
characteristics of MDZ biotransformation in vitro. Kinetics of 1-hydroxy (OH)
Values are means ± SD of three experiments. Concentration is expressed as ng/ml. MDZ was determined using recombinant CYP3A4*1 and CYP3A4*18. Assays Substrate concentrations were as follows: estrone 200 μmol/l and testosterone 250 μmol/l.
were performed at substrate concentrations between 10 and 250 μmol/l for MDZ. Data were analyzed using the Michaelis–Menten equation.
6β-OHT, 6β-hydroxytestosterone; E1, estrone.
VOLUME 85 NUMBER 3 MARCH 2009 www.nature.com/cpt
occupied by negative (E374) and polar (S119) residues. Although a more refined study using molecular dynamic sim- ulation would be necessary to confirm this issue, our results showed that the docking mode of MDZ bonding to the active site is different from that of testosterone, suggesting compat- ibility with our enzyme kinetic studies.
This study shows that the CYP3A4*18 variant of the CYP3A4 gene is associated with low BMD in Korean women, causing the conformational changes in the SRS regions of the mutant protein that lead to change in enzymatic activity.
The major CYP3A family members expressed in the adult human liver, CYP3A4 and CYP3A5, were targeted in our study as candidate genes for osteoporosis. Large race-related differ- ences in the CYP3A4 gene have been reported. For instance, CYP3A4*3 (M445T) and CYP3A4*17 (F189S) are found only in Caucasians, CYP3A4*15 (R162Q) is found only in Africans, and CYP3A4*10 (D174H) is found in both Caucasians and Africans.14,15 We have therefore screened all the exons of the CYP3A4 gene in Koreans and identified two single-nucleotide polymorphisms: L293P and the silent mutation, L295L, which is a novel single-nucleotide polymorphism detected first in our study. CYP3A4*18 has been identified only in some Asian populations, including Japanese and Chinese people, but not in Caucasians or African Americans. The certain mutations Figure 3 Molecular dynamic simulation and molecular docking results for
reported in Chinese or Japanese people, such as CYP3A4*3 CYP3A4*1 (wild type) and CYP3A4*18. (a) Secondary structural details of
(M445T), CYP3A4*16 (F189S), CYP3A4*4 (I118V), CYP3A4*5 CYP3A4*1 (illustration in gray) where the heme is shown as ruby-colored (P218R), and CYP3A4*6 (A17776), were not identified in our sticks. (b) Snapshots taken from the simulations at 2 ns, CYP3A4*1 (orange)
screening.10,16 These findings might be explained by the pres- superimposed with CYP3A4*18 (green) (RMSD = 2.13Å). Comparison of the ence of ethnic differences even among Asians.
secondary structural changes on helix I between CYP3A4*1 (lower box) and CYP3A4*18 (upper box) are presented in an enlarged view in the small boxes. CYP3A4*18 has shown a significantly higher turnover Docking mode of (c) midazolam (blue) and (d) testosterone (aqua green) in
activity for both testosterone and insecticide chlorpyrifos the CYP3A4*1 active site cavity. Residues are represented as gray sticks.
in vitro.14 However, the functional significance of CYP3A4*18 in drug metabolism or disease pathogenesis has not been clari- CYP3A4*18 proteins. Codon 293 is located at the start of the fied in vivo. The data we present in this study suggest, for the highly conserved helix I, which has been known to play an first time in the field of osteoporosis genetics, that gene muta- essential role in substrate specificity 12 The most tion possibly contributes to disease vulnerability, and that the important change in the L293P secondary structural elements CYP3A4*18 genotype is one of the genetic risk factors for low was observed in the conserved helix I: a long, straight α helix bone mass. For CYP3A5, the second CYP3A family member in in the WT was modified into two small α helices separated by a the adult human liver, we focused on the CYP3A5*3 genotype short loop in the final conformation of CYP3A4*18 because the CYP3A5*3/*3 homozygote is common in Koreans17 small boxes). This change reduces stability, and the consequent and is not able to make functioning CYP3A5 by alternative splic- fluctuation on the separated helices could lead to changes in ing. No BMD difference was observed between women with the other substrate recognition sites (SRSs) through SRS4 of whole CYP3A5 activity and those with deficient CYP3A5 activ- helix I. SRS3 and SRS1 were closer to SRS4 in the CYP3A4*18 ity, thereby indicating that CYP3A5 has no significant influence than in the WT protein. Furthermore, the conformational on bone metabolism.
shifts on the G and F helices observed in CYP3A4*18 could The molecular modeling strongly supports our hypoth- affect all the SRS regions except SRS6. To uncover the pos- esis, showing the significant secondary structural change in sible conformation of substrate–enzyme binding in the active CYP3A4*18. Surprisingly, change of a single amino acid, L293P, site, molecular docking was carried out using GOLD, version at the beginning of helix I has an influence on the overall pro- 3.1.1 (CCDC Software Ltd., Cambridge, UK), with the MDZ tein structure and leads to the modification of the arrangement and testosterone structures The active of SRS regions, the important sites for substrate recognition, site pocket is mostly hydrophobic with neutral residues. and substrate access to the active site. Our in vitro and in vivo One side of the substrate is brokered by nonpolar residues pharmacokinetic studies using a conventional probe drug, (I301, F304, A305, I369, and A370), and the other side is MDZ, also indicate that CYP3A4*18 is a functional mutation. ClInICAl PhArMACOlOgY & TherAPeuTICs VOLUME 85 NUMBER 3 MARCH 2009
In our in vivo pharmacokinetic study, subjects with CYP3A4*18 The genotype-specific differences in BMD were shown first exhibited lower enzyme activity in MDZ hydroxylation as com- in our study involving the CYP3A4*18 protein. The results of pared to the CYP3A4*1 group matched for the CYP3A5 geno- an additional search for the kinetics of metabolism and con- type. This result was confirmed by our in vitro enzyme kinetic struction of enzyme structure suggest a possible mechanism to assay with different doses of MDZ. The data obtained from both explain the effect of CYP3A4*18 on bone mass. This finding in vitro and in vivo experiments with MDZ are contradictory suggests that the CYP3A4 polymorphism may be a predictor to ours and to previous data using sex steroids as substrates. for osteoporosis in some Asian populations. Genetic markers in Molecular docking studies with the WT structure showed that estrogen metabolism could be clinically useful for identifying the pattern of MDZ bonding to the active site is different from subjects at risk for osteoporosis and also for estrogen-related that of testosterone. Therefore, structural changes and variable diseases such as breast cancer.
docking patterns detected in our modeling studies suggest that the conformational change in CYP3A4*18 may lead to the MethoDs
alteration of metabolic activity, depending on substrate types. subjects and measurement of BMD. The subjects in our study were a
CYP3A4*18, therefore, seems to be a two-faced mutation in hospital-based series of 2,178 healthy women of ethnic Korean back- metabolic activity: it acts as a rapid metabolizer of sex steroids, ground who had visited the hospital for a general checkup between 1994 and 2004. The protocol for the study was approved by the Chell but, on the other hand, it is a poor metabolizer of some drugs, General Hospital Institutional Review Board, and informed con- such as MDZ.
sent was obtained from the participants. We enrolled women aged CYP3A4 is thought to be responsible for the phase I metabo- 40–79 years, and those who had a history of chronic medical dis- lism of numerous structural y diverse exogenous and endogenous ease or were taking medications that could affect bone and calcium molecules, including steroids, fatty acids, prostaglandins, and metabolism were excluded from the study. Each patient was clini- cally examined, and routine biochemical tests were performed to lipid-soluble vitamins.18 As the critical role of estrogen in the exclude underlying diseases.
maintenance of skeletal health has been well demonstrated, and BMD at the lumbar spine (L2–L4) was measured by dual-energy the degree of estrogen exposure is a determining factor for bone X-ray absorptiometry using a QDR-2000 (Hologic, Bedford, MA), or mass,19,20 we evaluated whether CYP3A4*18 could affect estro- XR-36 (Norland, Fort Atkinson, WI). The precision errors (the coeffi- gen metabolism. Estrogens are metabolized to a large number of cients of variation for the in vivo BMD measurements) were 0.65 and 0.7%, respectively. The scores measured by XR-36 were converted to oxidated (2-, 4-, 6α-, 6β-, 12β-, 15α-, 16α-, and 16β-hydroxylated) those of QDR-2000 in accordance with the conversion equation used metabolites, mainly by CYP3A enzymes.21,22 CYP3A4 has a cat- in a previous study25: QDR-2000 = (0.876 × XR-36) + 0.124. Data were alytic activity predominantly for the 2-hydroxylation (which is obtained in 1,353 women for BMD at the proximal femur, measured by devoid of estrogen activity) rather than for the 4-hydroxylation dual-energy X-ray absorptiometry using the QDR-2000. The precision of estrogen.23,24 In our study, CYP3A4 had high catalytic activity error was 1.2%.
for the oxidation of both estrogen and testosterone. Although it DnA genotyping in CYP3A genes. To determine the presence of
is difficult to obtain statistically significant changes in estrogen genetic variations in the CYP3A4 gene, the 596 bp 5′-upstream and metabolism by CYP3A4*18 due to the smal and multiple peaks all 13 exons were amplified and sequenced in DNA from 225 ran- of metabolites, the oxidation of major metabolites of estrone by domly selected, unrelated Koreans. The identified single-nucleotide CYP3A4*18 had a tendency to be greater than that associated polymorphism, CYP3A4*18 (L293P), was detected by PCR and with the WT protein. As previous data show,14 the production the restriction fragment length polymorphism method, using for- ward primer (5′-TGATGCCCTACATTGAT CTGA-3′) and reverse of 6β-hydroxytestosterone by CYP3A4*18 from the metabolism primer (5′-GTGGTGAGGAGGCATTTTTG-3′) and restriction of testosterone was significantly higher than that produced by enzyme Msp I (NEB, Beverly, MA). On the basis of the published CYP3A5 variant alleles, we developed specific PCR–restriction Based on the data from our experiments, the plausible expla- fragment length polymorphism tests for CYP3A5*3 (6986A>G). nation for low BMD in CYP3A4*18+ women might be that a The sequences of the primers are as follows: forward primer (5′-TGGCATAGGAGATACCCACG-3′) and reverse primer (5′-GT gain-of-function mutation for sex steroids on codon 293 in the GGTCCAAACAGGGAAGAAATA-3′).
CYP3A4 gene results in the rapid metabolic clearance of sex hor- mones, including estrogens, leading to a relative sex-hormone CYP3A4 enzyme activity in vitro. The human CYP3A4*1 cDNA in
deficiency and consequent rapid bone turnover. To confirm our the vector pUV1 was a generous gift from Dr Gonzales (National hypothesis more clearly, it would be helpful to assess the cir- Cancer Institute, National Institutes of Health). A mutant contain- culating estrogen concentrations, estrogen metabolites, or the ing CYP3A4*18 was made using the GeneEditor in vitro site-directed skeletal responsiveness (such as bone turnover and BMD) to the mutagenesis kit (Promega, Madison, WI). Recombinant CYP3A4*1 hormone therapy, in relation to the various genotypes in a large and CYP3A4*18 were expressed using a baculovirus expression sys- tem purchased from Clontech Laboratories (Mountain View, CA). population. Our study involving a less common genotype found For western blot analysis, sodium dodecyl sulfate–polyacrylamide gel only in Asians might have limitations in its clinical implications, electrophoresis was used to separate the recombinant proteins, after but our pharmacogenetic approach focusing on metabolizing which the proteins were transferred onto nitrocellulose membranes. enzymes was able to identify one of the oligogenic determinants The membranes were incubated with anti-CYP3A4 primary anti- in osteoporosis. Searching for and grouping of oligogenic deter- body for 1 h at room temperature. An enhanced chemiluminescent kit (Pierce, Rockford, IL) was used for immunodetection. Enzyme minants in serial metabolic pathways of candidate molecules can content was monitored by means of the reduced CO spectrum, using a be a powerful tool to predict future osteoporosis.
DW-2000 Spectrophotometer. Protein concentration was determined VOLUME 85 NUMBER 3 MARCH 2009 www.nature.com/cpt
by the Bradford method.26 Next, we compared the catalytic activities (A080016), and the MOST/KOSEF for the Environmental Biotechnology of CYP3A4*1 and CYP3A4*18 for estrone, testosterone, and MDZ in National Core Research Center (R15-2003-012-02001-0).
accordance with previously described methods.14 Metabolites were analyzed using a liquid chromatographic–tandem mass spectrometric ConFliCt oF inteRest
system (API 2000; MDS Sciex, Concord, Ontario, Canada). Formation The authors declared no conflict of interest.
data for 1-hydroxy (OH)-MDZ for MDZ were fitted to a Michaelis– Menten model. All experiments were performed in triplicate.
2008 American Society for Clinical Pharmacology and Therapeutics In vivo pharmacokinetics using MDZ as a probe. Thirteen subjects
1. Ralston, S.H. & de Crombrugghe, B. Genetic regulation of bone mass and with the CYP3A4*18 allele and 26 normal healthy volunteers matched susceptibility to osteoporosis. Genes Dev. 20, 2492–2506 (2006).
for age (±2 years), sex, and CYP3A5 genotype were enrol ed in the 2. Daly, A.K., Cholerton, S., Gregory, W. & Idle, J.R. Metabolic polymorphisms. in vivo pharmacokinetic study using MDZ as a phenotypic probe Pharmacol. Ther. 57, 129–160 (1993).
drug. After an overnight fast, all subjects received a single 7.5-mg oral 3. Westlind, A., Löfberg, L., Tindberg, N., Andersson, T.B. & Sundberg, M. dose of MDZ (time 0). Blood samples were obtained before the drug Interindividual differences in hepatic expression of CYP3A4: relationship was administered and at 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, and 24 h afterward. to genetic polymorphism in the 5′-upstream regulatory region. Biochem. Pharmacokinetic parameters included MDZ concentration in plasma, Biophys. Res. Commun. 259, 201–205 (1999).
area under the plasma concentration curve, and total clearance. 4. Rogers, J.F., Rocci, M.L. Jr., Haughey, D.B. & Bertino, J.S. Jr. An evaluation of the suitability of intravenous midazolam as an in vivo marker for hepatic Weight-normalized oral MDZ clearance was calculated by dividing cytochrome P4503A activity. Clin. Pharmacol. Ther. 73, 153–158 (2003).
total clearance by body weight (in kilograms).
5. Wilkinson, G.R. Drug metabolism and variability among patients in drug response. N. Engl. J. Med. 352, 2211–2221 (2005).
Protein structure preparation for molecular modeling and molecular
6. Human Cytochrome P450 (CYP) Allele Nomenclature Committee dynamics simulations. Molecular modeling studies were used for
<http://www.imm.ki.se/CYPalleles/> (2006). Accessed 27 June 2006.
investigating the structure of WT CYT3A4 and the effects of its 7. Burk, O. & Wojnowski, L. Cytochrome P450 3A and their regulation. Naunyn mutation (CYP3A4*18) on structural stability. For this study, the Schmiedebergs Arch. Pharmacol. 369, 105–124 (2004).
three-dimensional coordinates of the CYP3A4 protein were obtained 8. Kuehl, P. et al. Sequence diversity in CYP3A promoters and characterization of from the Protein Data Bank (PDB ID: 1TQN).27 The X-ray crystal the genetic basis of polymorphic CYP3A5 expression. Nat. Genet. 27, 383–391
structure of the WT protein was resolved, and the missing part was recovered using the HOMOLOGY module in the INSIGHTII pro- 9. Floyd, M.D. et al. Genotype-phenotype associations for common CYP3A4 and CYP3A5 variants in the basal and induced metabolism of midazolam in gram (Insight II, version 2005.3L; Accelrys., San Diego, CA). The European- and African-American men and women. Pharmacogenetics 13,
CYP3A4*18 mutation model was constructed by the INSIGHTII 595–606 (2003).
BIOPOLYMER module.
10. Fukushima-Uesaka, H. et al. Haplotypes of CYP3A4 and their close linkage with The molecular dynamic simulations were performed using CYP3A5 haplotypes in a Japanese population. Hum. Mutat. 23, 100 (2004).
GROMACS Simulation Software, version 3.3.1, (a web-based software 11. Watkins, P.B. Noninvasive tests of CYP3A enzymes. Pharmacogenetics 4,
available at http://www.gromacs.org/)28 to study protein structural 171–184 (1994).
behavior in the polar environment. To compare the binding confor- 12. Streetman, D.S., Bertino, J.S. Jr. & Nafziger, A.N. Phenotyping of drug- mation of MDZ and testosterone, the study of the molecular docking metabolizing enzymes in adults: a review of in-vivo cytochrome P450 of the substrates to the active site of the protein was carried out using phenotyping probes. Pharmacogenetics 10, 187–216 (2000).
13. Khan, K.K., He, Y.Q., Domanski, T.L. & Halpert, J.R. Midazolam oxidation by GOLD, version 3.1.1. (CCDC Software Ltd., Cambridge, UK)29 cytochrome P450 3A4 and active-site mutants: an evaluation of multiple binding sites and of the metabolic pathway that leads to enzyme inactivation. statistical analysis. The data were presented as mean values ± SD,
Mol. Pharmacol. 61, 495–506 (2002).
and compared using Student's unpaired t- or Mann–Whitney U-tests 14. Dai, D. et al. Identification of variants of CYP3A4 and characterization of their and one-way analysis of variance or Kruskal–Wallis test as appro- abilities to metabolize testosterone and chlorpyrifos. J. Pharmacol. Exp. Ther. priate. Multiple linear regression analysis was used to adjust BMD 299, 825–831 (2001).
for confounding factors such as age, years since menopause, BMI, 15. Lamba, J.K. et al. Common allelic variants of cytochrome P4503A4 and and CYP3A5 genotype. The interactions between genotype and their prevalence in different populations. Pharmacogenetics 12, 121–132
several covariates, such as age, years since menopause, BMI, and 16. Hsieh, K.P. et al. Novel mutations of CYP3A4 in Chinese. Drug Metab. Dispos. 29,
CYP3A5 genotype, were examined using linear regression analysis. 268–273 (2001).
Comparison of MDZ clearance and area under the plasma concentra- 17. Park, S.Y., Kang, Y.S., Jeong, M.S., Yoon, H.K. & Han, K.O. Frequencies of CYP3A5 tion curve between the CYP3A4*18 and CYP3A4*1 proteins was per- genotypes and haplotypes in a Korean population. J. Clin. Pharm. Ther. 33,
formed using a linear mixed-effects model. Underlying assumptions 61–65 (2008).
regarding linear regression and mixed-effects models were checked by 18. Handschin, C. & Meyer, U.A. Induction of drug metabolism: the role of nuclear residual plot, normal probability plot of the residuals, and the absolute receptors. Pharmacol. Rev. 55, 649–673 (2003).
residual plot, and no relevant violations were found. A P value <0.05 19. Leelawattana, R. et al. The oxidative metabolism of estradiol conditions was considered to indicate a significant difference. Statistical analysis postmenopausal bone density and bone loss. J. Bone Miner. Res. 15,
was performed using commercially available software, SPSS 11.0 for 2513–2520 (2000).
20. Nguyen, T.V., Jones, G., Sambrook, P.N., White, C.P., Kelly, P.J. & Eisman, J.A. Windows (SPSS, Chicago, IL) and the Statistical Analysis System pro- Effects of estrogen exposure and reproductive factors on bone mineral gram, version 9.1 (SAS Institute, Cary, NC).
density and osteoporotic fractures. J. Clin. Endocrinol. Metab. 80, 2709–2714
Linkage analysis between the CYP3A4 and CYP3A5 genotypes was performed using SNPstats, a web-based software available at 21. Lee, A.J., Kosh, J.W., Conney, A.H. & Zhu, B.T. Characterization of the http://bioinfo.iconcologia.net/SNPstats. Linkage disequilibrium results NADPH-dependent metabolism of 17beta-estradiol to multiple are presented as D′ and P values. We defined a D′ ≥ 0.70 as high linkage metabolites by human liver microsomes and selectively expressed human disequilibrium and a P value <0.05 as a significant value (e.g., the calcu- cytochrome P450 3A4 and 3A5. J. Pharmacol. Exp. Ther. 298, 420–432
lated D′ value is significant).30 22. Lee, A.J., Mills, L.H., Kosh, J.W., Conney, A.H. & Zhu, B.T. NADPH-dependent metabolism of estrone by human liver microsomes. J. Pharmacol. Exp. Ther. 300, 838–849 (2002).
This study was supported by grants from the Korea Healthcare technology 23. Lee, A.J., Cai, M.X., Thomas, P.E., Conney, A.H. & Zhu, B.T. Characterization project, Ministry for Health, Welfare and Family Affairs, Republic of Korea of the oxidative metabolites of 17beta-estradiol and estrone formed by 15 ClInICAl PhArMACOlOgY & TherAPeuTICs VOLUME 85 NUMBER 3 MARCH 2009
selectively expressed human cytochrome p450 isoforms. Endocrinology 144,
27. Yano, J.K., Wester, M.R., Schoch, G.A., Griffin, K.J., Stout, C.D. & Johnson, E.F. The 3382–3398 (2003).
structure of human microsomal cytochrome P450 3A4 determined by X-ray 24. Tsuchiya, Y., Nakajima, M. & Yokoi, T. Cytochrome P450-mediated crystallography to 2.05-A resolution. J. Biol. Chem. 279, 38091–38094 (2004).
metabolism of estrogens and its regulation in human. Cancer Lett. 227,
28. Van Der Spoel, D., Lindahl, E., Hess, B., Groenhof, G., Mark, A.E. & 115–124 (2005).
Berendsen, H.J. GROMACS: fast, flexible, and free. Comput. Chem. 26,
25. Kim, S.W., Lim, C.H., Han, K.O., Jung, H.Y., Min, H.K. & Han, I.K. Standardization 1701–1718 (2005).
of dual energy X-ray absorptionmetry (DXA) in spinal BMD of Korean 29. Jones, G., Willett, P., Glen, R.C., Leach, A.R. & Taylor, R. Development and women and phantom. 10th International Congress of Endocrinology, validation of a genetic algorithm for flexible docking. J. Mol. Biol. 267,
San Francisco, CA 1996. 372.
727–748 (1997).
26. Bradford, M.M. A rapid and sensitive method for the quantitation of 30. Bento, J.L. et al. Genetic analysis of the GLUT10 glucose transporter (SLC2A10) microgram quantities of protein utilizing the principle of protein-dye binding. polymorphisms in Caucasian American type 2 diabetes. BMC Med. Genet. 6, 42
Anal. Biochem. 72, 248–254 (1976).
VOLUME 85 NUMBER 3 MARCH 2009 www.nature.com/cpt

Source: http://bio.gnu.kr/publication/pdf/CPT_2009.pdf