Allele-specific expression and gene methylation in the control of CYP1A2 mRNA level in human livers

ORIGINAL ARTICLE

Allele-specific expression and gene methylation

in the control of CYP1A2 mRNA level in human

livers

Roza Ghotbi1, Alvin Gomez2,Lili Milani3, Gunnel Tybring1,Ann-Christine Syvanen3,Leif Bertilsson1, MagnusIngelman-Sundberg2 andEleni Aklillu1

1Division of Clinical Pharmacology, Departmentof Laboratory Medicine, Karolinska UniversityHospital Huddinge, Karolinska Institutet,Stockholm, Sweden; 2Section ofPharmacogenetics, Department of Physiologyand Pharmacology, Karolinska Institutet,Stockholm, Sweden and 3Molecular Medicine,Department of Medical Sciences, UppsalaUniversity, Uppsala, Sweden

Correspondence:Dr E Aklillu, Division of Clinical Pharmacology,Department of Laboratory of Medicine,Karolinska Institutet, Karolinska UniversityHospital Huddinge, C-168, SE-141 86Stockholm, Sweden.E-mail: [email protected]

Received 7 October 2008; revised 13 January2009; accepted 3 February 2009; publishedonline 10 March 2009

The basis for interindividual variation in the CYP1A2 gene expression is notfully understood and the known genetic polymorphisms in the gene provideno explanation. We investigated whether the CYP1A2 gene expression isregulated by DNA methylation and displays allele-specific expression (ASE)using 65 human livers. Forty-eight percent of the livers displayed ASE notassociated to the CYP1A2 mRNA levels. The extent of DNA methylation of aCpG island including 17 CpG sites, close to the translation start site, inverselycorrelated with hepatic CYP1A2 mRNA levels (P¼0.018). The methylation oftwo separate core CpG sites was strongly associated with the CYP1A2 mRNAlevels (P¼0.005) and ASE phenotype (P¼0.01), respectively. The CYP1A2expression in hepatoma B16A2 cells was strongly induced by treatment with5-aza-20-deoxycytidine. In conclusion, the CYP1A2 gene expression isinfluenced by the extent of DNA methylation and displays ASE, mechanismscontributing to the large interindividual differences in CYP1A2 geneexpression.The Pharmacogenomics Journal (2009) 9, 208–217; doi:10.1038/tpj.2009.4;published online 10 March 2009

Keywords: CYP1A2; drug-metabolizing enzyme; gene expression; allele-specific expression;epigenetics; DNA methylation

Introduction

The human cytochrome P450 1A2 (CYP1A2) enzyme is expressed mainly in theliver and accounts for about 13% of the total P450 content in the liver.1 It has animportant function in the metabolism of caffeine as well as of several clinicallyimportant drugs such as clozapine,2 phenacetin,3 verapmil,4 endogenoussubstrates such as melatonin,5 estradiol,6 and procarcinogenic substancesincluding heterocyclic amines, arylamines and aflatoxin B1.7,8

The CYP1A2 gene shows large interindividual and interethnic differences inboth mRNA expression level and enzyme activity.8–10 The underlying causes forthis variation are not known and could be caused by genetic, epigenetic orenvironmental factors. Indeed, CYP1A2 is genetically polymorphic with morethan 16 variant alleles described so far (http://www.cypalleles.ki.se/cyp1a2.htm).However, the low frequency of the functionally different variant alleles in thepopulations and the results of several genotype–phenotype association studiesindicate a limited function of CYP1A2 genetic polymorphism as a cause ofinterindividual variation in CYP1A2 gene expression or activity.11,12 In line of

The Pharmacogenomics Journal (2009) 9, 208–217& 2009 Nature Publishing Group All rights reserved 1470-269X/09 $32.00

www.nature.com/tpj

this research, we have found one allele CYP1A2*1K causinglower expression of the gene but it is present in Ethiopians13

and almost absent in whites and Asians.11,12 The interethnicdifferences in CYP1A2 activity is important and by control-ling for the effects of smoking, oral contraceptive use andgenotype, we recently reported the presence of significantdifferences in enzyme activity between Swedish, Korean andSerbian populations, differences not explained by theknown CYP1A2 single nucleotide polymorphisms (SNPs)/haplotypes.12,14

Apart from genetic polymorphisms, phenotypic variationsare governed by other factors affecting gene expression,such as gene cis- and trans-acting transcriptional compo-nents, alternative splicing, RNA stability, expression ofregulatory RNAs and epigenetics such as histone modifica-tions and DNA methylation.15 Imprinted genes (parent oforigin-specific gene expression in somatic cells) and genessubject to X chromosome inactivation display monoallelicexpression. Differential gene expression between the twoalleles or allele-specific expression (ASE) in nonimprintedgenes is also common in the human genome.16–19 Thedegree of differences in the expression between the twoalleles varies from individual to individual causing pheno-typic variability. ASE reflects the presence of putative allele-specific cis-acting factors of either genetic or epigeneticorigin. Consequently, ASE can be used as a promisingquantitative phenotyping method for identifying cis-actingfactors as well as epigenetic variation influencing geneexpression because individuals heterozygous for cis-actingpolymorphism show a different level of mRNA expressionoriginating from one allele compared to the other.16,17,20

The two major mechanisms for epigenetic modificationsare DNA methylation and histone deacetylation, andinterindividual variations in the degree of these mechan-isms could form a basis for interindividual differences ingene expression. Epigenetic modifications and ASE inpharmacogenetic-related genes have been shown to affectexpression and activity. ASE for the MDR1, CYP2D6 andCYP2C9 genes has been reported recently.21 Hirota et al.22

reported the occurrence of ASE in the CYP3A4 geneexpression that correlated with the total CYP3A4 mRNAlevels in liver but no clear association between genemethylation and total mRNA levels was found. Nakajimaet al.23 indicated a function of histone deacetylation andDNA methylation in cell-specific inducibility of thehuman CYP1 family. Tissue-specific variation in the degreeof gene methylation in CYP2E1 gene is associated withCYP2E1 mRNA levels and enzyme activity that has beenreported previously.24 Furthermore, Hammons et al.25

showed an association between interindividual variation inhuman hepatic CYP1A2 expression and degree of methyla-tion of a CpG site (bp �2759) in the 50-flanking region of theCYP1A2 gene, with hypermethylation correlated withdecreased CYP1A2 expression. However, the CpG siteinvestigated by Hammons et al. is not located within aCYP1A2 CpG island and is located far away from thetranscription start site. Indeed, functional CpG islandstypically occur at or near the transcription start site of

genes26 causing transcriptional repression by inhibition ofbinding of transcription factor.

While appreciating the existence of the wide interindivi-dual variation in the constitutive CYP1A2 mRNA expressionand enzyme activity, the underlying causes remain unclear.It is generally accepted that most of the known SNPs in theCYP1A2 gene and 50-flanking regions do not determine theobserved large interindividual variability in constitutivehepatic CYP1A2 mRNA expression and enzyme activity.11

Identification of the molecular mechanism behind variationin CYP1A2 enzyme activity may have important implica-tions for drug safety/efficacy and cancer susceptibility.Because the epigenetic factors and ASE can lead tophenotypic variability in the pharmacodynamic and phar-macokinetic outcomes of drug therapy, we considered it ofinterest to investigate the function of DNA methylation anddifferential allelic expression in CYP1A2 gene regulation asdeterminants of interindividual variation in total CYP1A2mRNA expression and enzyme activity. The results indicatethe importance of two core methylation CpG sites locatednear the transcription start site to be associated withvariations in the total CYP1A2 transcript level and differ-ential allelic expression phenotype (Figure 1).

Results

Detection of ASE in CYP1A2

A total of 65 human liver samples were genotyped for themarker SNP rs2470890 and of those 23 were found to beheterozygous for this SNP. The regions encompassing thers2470890 were PCR amplified from both genomic DNA(gDNA) and mRNA (after conversion to cDNA), followed byminisequencing analysis of each allele. PCR testing forgDNA and cDNA samples was carried out using the sameconditions for all fragments, and the amplicon size wasverified by agarose gel electrophoresis. The averages of thesignal fraction (SAllele1/(SAllele1þ SAllele2)) from the cDNAwere compared with the signal fraction from the respectivegDNA as a reference to determine the presence ofdifferential allelic expression. Of the 23 heterozygous liversamples, 11 (47.8%) displayed significant differences in theirallelic expression (Po0.05, Figure 2). There was no correla-tion in these samples between the total mRNA expressionlevel and allelic expression ratio or ASE phenotype.

DNA methylation frequency in human livers and hepatic CYP1A2mRNA expression

The region covering the CpG island (Figure 3) was PCRamplified from gDNA after bisulfite treatment. DNA methy-lation frequency was determined by pyrosequencing usingsix internal sequencing primers that cover overlappingregions. Complete data concerning the DNA methylationfrequencies and mRNA quantification were obtained from48 human livers. Analysis of the methylation patternrevealed two domains with region-specific differences inthe distribution and level of CpG methylation. The firstdomain (domain 1) contained CpG sites 5, 6, 7, 13, 17, 18and 19, and the second domain (domain 2) consisted of

ASE and gene methylation of CYP1A2R Ghotbi et al

209

The Pharmacogenomics Journal

CpG sites 8, 9, 10, 11, 12, 14, 15, 16, 17 and 21. CpG site 17was common for both domains (Figure 3).

The distribution of the methylation frequency patternand level of variation in the whole CpG island, domain 1,domain 2 and CpG sites 14 and 16 in relation to the hepaticCYP1A2 mRNA expression is shown in Table 1. We found a

significant negative correlation between the overall meanmethylation frequency of the 17 CpG sites and total CYP1A2mRNA expression (P¼ 0.018); thus, a lower extent ofmethylation associated to a higher CYP1A2 mRNA expres-sion level. The mean methylation frequencies of domain 2correlated significantly with the total CYP1A2 mRNA levels(P¼0.02), but there was no correlation between the mRNAlevels and the methylation of domain 1 (P¼0.18). Wefurther compared the CYP1A2 mRNA levels with the meanmethylation frequency of each CpG site within domain 2.The highest negative correlation was found for the CpG sites

Figure 1 Schematic view of the region of the CYP1A2 gene analyzed in this study indicating the transcription start site in exon 1 (�832) and the

location of the CpG island in relation to the translation start codon ATG site in exon 2.

Figure 2 Comparison of fluorescence signal fractions (SAllele1/(SAlle-

le1þ SAllele2)) between cDNA and the respective genomic DNA in humanlivers heterozygous for the marker coding single nucleotide polymorph-

isms (SNP) (rs2470890C/T). Human liver samples showing deferential

allelic expression significantly deviated from equal molar expression or

allele-specific expression (Po0.05) are indicated above and below withthe arbitrary dashed lines. The individual human liver samples are

depicted by number.

Figure 3 Sequence of the fragment including the CYP1A2 CpG islands

in exon 2 (þ72 to 494 bp) investigated in the present study beforesodium bisulfite treatment. Each CpG site with the respective number-

ing in superscript is indicated with the letter Y. The aryl hydrocarbon

receptor (AhR) binding site is underlined. Analysis of the methylationpattern revealed two domains with region-specific differences in the

distribution and level of CpG methylation. The first domain (domain 1)

contained CpG sites 5, 6, 7, 13, 17, 18 and 19, and the second domain

(domain 2) consisted of CpG sites 8, 9, 10, 11, 12, 14, 15, 16, 17 and21. CpG site 17 was common for both domains.


210


14 (P¼ 0.006) and 16 (P¼0.009), located in middle ofdomain 2, whereas no correlation was found with the otherCpG sites.

Association of methylation frequency with ASE

There was no significant correlation between the overallmean methylation frequency of the 17 CpG sites with ASEphenotype in the 23 human livers used in this study(P¼0.11). Considering the methylation frequency of thetwo domains separately, no correlation was found withdomain 1 but a borderline correlation with mean methyla-tion frequency of domain 2 was observed (P¼ 0.054). Wefurther analyzed associations between the mean methyla-tion frequencies of each individual CpG site located indomain 2 with the ASE phenotype. The result indicated asignificant correlation between the mean methylationfrequency of CpG sites 9 (r2¼0.21, P¼0.04) and 11(r2¼0.33, P¼ 0.006) located in the upper region of domain2 with ASE phenotype in that a lower methylationfrequency predicted a higher variation in the relativetranscript level derived from the two alleles. Neither themethylation frequency at CpG sites 9 and 11 nor the ASEphenotype correlated with the total hepatic CYP1A2 mRNAlevels.

DNA hypomethylation and CYP1A2 mRNA level in hepatomaB16A2 cellsTo clarify whether DNA demethylation increases theCYP1A2 gene expression, we treated human hepatomaB16A2 cells with varying concentrations of a hypomethylat-ing agent 5-aza-20-deoxycytidine (AzaC). The CYP1A2 mRNAexpression level, determined by reverse transcriptase (RT)–PCR, linearly increased (r2¼0.82, Po0.01) with the amountof AzaC used, indicating that the gene is silenced bymethylation in the cells (Figure 4).

Correlations between CYP1A2 mRNA expression, protein contentand enzyme activity

CYP1A2 enzyme activity was measured in the 21 humanlivers using 3 different probes, caffeine, phenacetin and 7-ethoxyresorufin (EROD). There were significant correlations

in CYP1A2 enzyme activity determination between thethree probes: caffeine vs EROD (Po0.0001, r2¼0.75),caffeine vs phenacetin (P¼ 0.0001, r2¼0.86) and phenace-tin vs EROD (Po0.0001, r2¼0.86). The enzyme activityusing caffeine, EROD and phenacetin was correlatedsignificantly with mean CYP1A2 mRNA level (P¼ 0.035,P¼0.015 and P¼0.029, respectively) and protein content(Po0.0001 for all three probes). There were no significantcorrelations between mRNA level and protein content(P¼0.16, statistical power¼ 28.0%, using a two-sided testand a set at 0.05).

Correlations of CYP1A2 genotype with CYP1A2 phenotype, ASEand DNA methylation frequency

To investigate linkage of the marker SNP rs2470890 withother known CYP1A2 SNPs that may have influence on themRNA expression, we genotyped the livers for the knownCYP1A2 variants (�3858 G4A, �2467 delT, �739T4G,�729 C4T and �163 C4A). The SNP frequencies of�3860G4A, �2467delT, �163C4A and rs2470890 were5.0, 5.0, 67.5 and 62.5%, respectively. CYP1A2*1K(�729C4T and �739T4G) was not detected. Linkageanalysis using Arlequin population genetics software in-dicated significant linkage of rs2470890 with �163C4Aoccurring at a frequency of 62.5% but not with the otherSNPs. However, none of the individual CYP1A2 SNPs/haplotypes including the haplotype comprising the markerSNP rs2470890 and �163C4A showed any significantcorrelation with CYP1A2 mRNA expression level, proteincontent, enzyme activity as determined by three differentprobes (caffeine, EROD and phenacetin), ASE or variationsin DNA methylation frequencies in human livers.

Discussion

We investigated the function of gene methylation anddifferential allelic expression in CYP1A2 gene regulationwith the aim to understand the molecular basis behind theobserved wide variation in CYP1A2 gene expression inhuman liver. The results indicated ASE of the CYP1A2 genethat is independent of the total CYP1A2 mRNA expression.We were furthermore able to identify two independent coremethylation sites located close to the translation start site asdeterminants of interindividual variation in the CYP1A2mRNA levels and ASE phenotype, respectively, presenting anadditional insight into the regulation of CYP1A2 gene tocause altered expression and enzyme activity in humans.

We investigated whether the marker SNP (rs2470890) usedin the ASE and DNA methylation analysis is in linkagedisequilibrium with other known CYP1A2 SNPs that mightinfluence the mRNA expression and enzyme activity.Indeed, the rs2470890 was linked to �163C4A. However,none of the individual SNPs/haplotypes including themarker SNP plus �163C4A haplotype had significantinfluences on mRNA expression, protein content, enzymeactivity, differential expression of the alleles or variationsin DNA methylation frequencies in human livers. Recently,we reported that none of the CYP1A2 SNPs/haplotypes

Figure 4 Relative CYP1A2 mRNA expression in human hepatoma

B16A2 cells treated with various concentrations of the hypomethylating

agent 5-aza-20-deoxycytidine (AzaC) for 5 days. CYP1A2 gene expres-

sion increased linearly with the concentration of AzaC.


211


investigated for affected enzyme activity using caffeine as aprobe in Swedish and Korean populations.12 Accordingly,sequencing of the CYP1A1_CYP1A2 locus of the highest andlowest CYP1A2 subjects worldwide, who have previouslybeen CYP1A2 phenotyped, revealed that no SNP orhaplotype in the CYP1A2 gene has yet been identified thatcan unequivocally be used to predict the metabolicphenotype.11 Lack of CYP1A2 the observed genotype–phenotype relationship was indeed the basis for the presentstudy; that is, searching for additional factors such asepigenetic that may influence interindividual variations inCYP1A2 expression level in human livers.

The CYP1A2 gene is specifically expressed in the liver whyanalysis of ASE and gene methylation was analyzed in 65different human liver samples. About half of the liversinvestigated by the ASE assay displayed significant variationin the levels of transcript derived from the two alleles. Somegenes that display ASE are present on chromosomes thatcontain other imprinted genes,16 which is also warrant forthe CYP1A2 gene on chromosome 15 where other imprintedgenes in the region are present.27–29

We found no significant correlation between ASE andtotal CYP1A2 gene expression at the mRNA level. Differ-ential allelic expression is influenced mainly by cis-actingmechanisms affecting gene regulation and/or mRNA proces-sing, whereas total gene expression is influenced by both cis-and trans-acting factors. Measuring the relative allelic mRNAexpression level selectively detects cis-acting factors and/orepigenetic factors.30,31 Consequently, the observed differ-ences between ASE and CYP1A2 gene expression levelsindicate that the CYP1A2 gene is also regulated by trans-acting mechanisms that differ between individuals.

As indicated in Figure 2, the preferably expressed CYP1A2allele varied between livers and this bidirectional ASE couldresult from allelic heterogeneity of one or more cis-actingregulatory polymorphisms present in coding, intronic and/or other noncoding regulatory sequences,32 or from DNAmethylation, histone acetylation and other epigeneticfactors.16,33 Alternatively, the marker SNP used in the ASE

screen could be in linkage disequilibrium to another SNPthat is responsible for the differential expression of allelictranscripts, but that does not show unidirectional ASE dueto noncomplete linkage disequilibrium with the markerSNP.

We used the Pyrosequencing technology that is ideallysuited for the simultaneous analysis and quantification ofthe methylation degree of several CpG positions in proxi-mity. The results indicate that the overall DNA methylationfrequency of the 17 CpG sites in the CpG island locatedclose to the translation start site in exon 2 most likelydetermine individual variation in total CYP1A2 mRNAexpression levels in human livers.

Previous studies indicated that a group of non-X-linkedbona fide promoter CpG islands are densely methylated innormal somatic tissues and that DNA methylation has animportant function in silencing germ-cell-specific genes insomatic cells.34 Thus, we evaluated the effect of DNAmethylation on CYP1A2 gene expression and investigatedwhether demethylation alleviates the silencing pressure toincrease gene expression in human B16A2 cell lines.Hepatoma B1A2 cells were treated with increasing concen-trations of a hypomethylating agent AzaC and relativequantification of CYP1A2 mRNA by RT–PCR revealed alinear raise in CYP1A2 gene expression with increasingconcentration of AzaC, illustrating that DNA methylation isinvolved in the repression of CYP1A2 (Figure 4).

Comparison of mean methylation frequency of each CpGsite within domain 2, where methylation primarily corre-lated to CYP1A2 mRNA levels, indicated significant correla-tion with CpG sites 14 and 16 (located in the middle ofdomain 2) whereas no association was found with otherCpG sites. This may indicate that CpG sites 14 and 16, ascore methylation regulatory sites, are of importance indetermining variations in CYP1A2 gene expression. Inter-estingly, the level of and variance in the methylationfrequency among the human livers was the highest forCpG sites 14 and 16 compared to the other CpG sites in theisland (Table 2).

Table 1 Summary of sequencing primers, number of CpGs, sequencing length and dispensation order used for sequencing ofCpG island in CYP1A2 exon 2

Sequencing primers Number of CpGs Sequence length Dispensation order

Primer 1TTTGGTATTGTTAAGGATGAGT

3 16 TCAGCTGTCACTGACTG

Primer 2AGTTTGATTTTTAGTATAGATTTTG

4 25 GACTCTGTGTGTGCTCTGTCTGGCTG

Primer 3GAGTYGTTTGGATATTAT

2 23 TCTGTCAGTGTCGCTG

Primer 4GGTTTAGAATGTTTTTAATATTTT

1 8 TCACTCTG

Primer 5GGGAYGTTTTGTAGAT

4 36 TCGCTATGCTATCGTCGTGTGTGTGAGTCG

Primer 6AGGTTTTGGTGYGGTAGG

3 19 GTCGATCGACTAGTCG

Cytosine used as internal controls to verify completion of sodium bisulfite treatment is indicted in bold in their dispensation orders. Underlined TCs are CpG sites. Y is

either a C or T.


212


Previous studies presented evidence for variation in thedegree of methylation as a cause for differential allelicexpression.16,17,20,31 In the present study, although ASEphenotype did not correlate significantly with the overallmethylation frequency of the CpG island, region-specificcomparison of methylation frequencies indicated a signifi-cant correlation between the methylation frequency of CpGsites 9 and 11 (located in the upstream region of domain 2)with the ASE phenotype. Thus, a lower methylationfrequency indicated a higher ASE phenotype. This indicatesthat in the absence of a silencing pressure from genemethylation, the cis-acting factor responsible for the ASEphenotype exerts its effect. It is interesting to note thatsimilar to the ASE phenotype, methylation frequency ofCpG sites 9 and 11 was also not associated with the totalCYP1A2 mRNA expression level. This indicates thatalthough methylation frequency at CpG sites 9 and 11may contribute to the presence of differential allelicexpression, CYP1A2 mRNA expression levels are probablyinfluenced to a major extent by the degree of methylation ofthe whole CpG island, mainly at the CpG sites 14 and 16.Hammons et al.25 reported the function of a single site-specific DNA methylation status of the CCGG site (bp�2759) located adjacent to an AP-1 site to be responsible forthe interindividual variation of CYP1A2. This site, not anyother place 5 kb upstream of the gene, could however beidentified by us as a CpG site in a CpG island. Instead, wefound a CpG island containing 17 CpG sites, located close tothe translation start site, and considered it of importance toinvestigate further.

Silencing of one of the X chromosomes in females andallelic imprinting or monoallelic expression, depending onparental origin, is highly regulated through gene methyla-tion reprogramming that may occur in germ cells andduring gametogenesis.35,36 However, the function of CpGmethylation in nonimprinted, monoallelic expressing genesis not yet fully explored. The two identified core methyla-tion sites are located in the same domain close to each otherbut apparently having two different functions: the upperregion determines the ASE phenotype and the middle regiondetermines the total CYP1A2 mRNA expression. Searchingfor transcription binding sites in the CpG island, we found

that both core methylation sites are surrounded by tran-scription binding sites for transcription factors such asBRAC1, TFAP3A, SP1, SPI1, GATA-2 and GATA-3. Given thefact that CYP1A2 is a hepatic P450, we found that variationin the methylation frequency of the binding sites and henceblocking access to transcription factors might be importantin controlling the level of hepatic CYP1A2 gene expression.It is also of interest to note that the aryl hydrocarbonreceptor (AhR) binding site is located close to the CpG island(Figure 3) and thus variation in degree of gene methylationmight also result in differential CYP1A2 inducibility.

Methylation patterns may be cell type and/or tissuespecific and change slowly with age and in response toenvironmental effects such as diet.37,38 Variation in methy-lation frequency due to dietary related or environmentalfactors may be important in determining the observed wideethnic variation in CYP1A2 enzyme activity between whitesand Asians.12 Furthermore, previous pedigree studies de-monstrated heritability of ASE being transmitted by Mende-lian inheritance.17,39 Gene methylation patterns in somaticdifferentiated cells are also generally stable and heritable.35

Inheritance of ASE in the CYP3A4 gene expression has beenpreviously reported.22 Accordingly the observed ASE inCYP1A2 gene expression may be inherited to cause variationin enzyme activity similar to genetic polymorphisms.Several studies reported lack of genotype–phenotype rela-tionship in CYP1A2. Our study explains the molecularmechanism behind variations in CYP1A2 gene expressionand genotype–phenotype concordance to a certain extent.The magnitude of ASE as well as methylation frequencyvaries between individuals and possibly affects enzymeactivity and treatment response.

In conclusion, our study demonstrates that CYP1A2 geneexpression displays ASE and is influenced by DNA methyla-tion. We showed that only a few CpG sites and quite smallregions of the CpG island provide relevant information fordifferential methylation analysis. The present study hasgiven a new insight into the regulation of CYP1A2.However, further investigations are needed to explore theexact mechanisms involved in ASE, transregulatory mechan-isms and DNA methylation in explaining interindividualand ethnic variability in CYP1A2 expression. ASE discovery

Table 2 Distribution and level of variation in methylation frequency in the CYP1A2 CpG island, different domains, coremethylation sites (CpG 14 and 16) and their correlation with relative hepatic CYP1A2 mRNA level in 48 human livers usingregression analysis

CpG sites Mean±s.d. Variance Range Correlation with hepatic CYP1A2 mRNA levelF P

Overall 17 CpG sites 25.5±3.3 10.9 19.4–33.8 5.96 0.018Domain 1 24.8±3.8 14.7 15.6–32.1 1.88 0.177Domain 2 28.5±4.0 16.4 21.0–39.1 5.70 0.021CpG 14 30.0±5.8 32.2 18.0–46.0 8.28 0.006CpG 16 32.0±7.1 50.7 18.0–49.0 7.39 0.009CpG 14+CpG 16 31±6.1 37.2 18.0–47.0 8.70 0.005

List of CpG sites located in domains 1 and 2 is described in Figure 3.


213


coupled with elucidation of the underlying mechanismsmay provide a comprehensive view of transcriptionalregulation as well as functional markers for drug metaboliz-ing genes (for example, CYP1A2 and CYP3A4) where geneticpolymorphisms have a minor function in explaining theobserved wide interindividual and ethnic variability inenzyme activity.

Materials and methods

Study samples

A total of 65 unrelated healthy human livers of Scandina-vian origin belonging to a donor liver biobank, establishedat the Division of Clinical Pharmacology, Karolinska Uni-versity Hospital Huddinge, Sweden, were utilized for thisstudy after approval from the ethics committee at KarolinskaInstitutet. Tissue for preparation of microsomes was onlyavailable from 21 human livers.

Preparation of genomic DNA and cDNA

Genomic DNA and total RNA were isolated from the liversamples using QIAamp DNA Mini Kit and RNeasy Mini Kit(Qiagen, Hilden, Germany), respectively. Total RNA wastreated with RNase-free DNase (Promega, Madison, WI, USA)to remove any contamination from gDNA during total RNApreparation. cDNA was synthesized using the First StrandcDNA Synthesis Kit (Fermentas, Burlington, Ontario, Cana-da), following the manufacturer’s instructions using 2 mg oftotal RNA. cDNA was prepared from each liver using randomhexamer and oligo (dT)18 primer.

Detection of differential allelic expression by Tag arrayminisequencingAssay design. Several SNPs in the CYP1A2 gene wereconsidered as candidate markers for the ASE assay.Selection criteria for a marker SNP were high minor allelefrequency with no significant effect on enzyme expressionlevel or enzyme activity and location in the transcribedcoding region so that the same set of primers for PCRamplification can be used in both gDNA and mRNA. Anonfunctional silent SNP (5347T4C, rs2470890) located inexon 7 of the CYP1A2 gene was selected as marker.Previously we identified this SNP (rs2470890) inEthiopians and reported lack of influence on CYP1A2enzyme activity using caffeine as a probe.13 To examinelinkage of the marker SNP with other CYP1A2 SNPs, wegenotyped livers for known CYP1A2 variant SNPs(�3858G4A, �2467delT, �739T4G, �729C4T, �163C4A)as described previously.12

PCR primers and minisequencing primers with 50 tagsequences covering the marker SNP site were designed usingthe Autoprimer (http://www.autoprimer.com; BeckmanCoulter) software. The primers were obtained from Inte-grated DNA Technologies (Coralville, IA, USA).

PCR minisequencing and hybridization. Genomic DNA andcDNA were used for genotyping of the marker SNP using4 ng DNA by minisequencing analysis as described

previously.30,40 In brief a 164-bp PCR fragment spanningthe marker SNP site was amplified by PCR using a forwardprimer (50-ATGAAGCACGCCCGCTGT-30) and a reverseprimer (50-TTGGCTAAAGCTGCTATTTTTTA-30) from bothgDNA and cDNA. To remove remaining dNTPs and primers,we treated the PCR products (7.2ml) with 1 U ml�1 shrimpalkaline phosphatase and 0.48 U ml�1 exonuclease I (USBCorporation, Cleveland, OH, USA). The oligonucleotidescomplementary to the tag sequence attached to the 50-endof the minisequencing primer (50-AATACGCTGAATAGAGCCCTAGGCGCGGCTGCGCTTCTCCATCAA-30) wereimmobilized on the CodeLink Activated Slides (GEHealthcare, Uppsala, Sweden) in duplicate spots in eachsubarray. Cyclic minisequencing of the PCR products fromboth gDNA and cDNA, hybridization of the minisequencingreaction products to the captured oligonucleotides (cTags)on the slides and washing of the slides were performed asdescribed previously.41 Each sample (DNA and thecorresponding cDNA) was run in triplicates on the sameslide.

Signal detection and data analysis. Fluorescence signals fromthe microarrays were measured using a ScanArray Expressinstrument (PerkinElmer Life Sciences, Boston, MA, USA).The laser power was kept constant at 80% and thephotomultiplier tube gain adjusted to obtain equal signallevels from the reaction control spots in all four-laserchannels. The excitation lasers were Blue Argon 488 nmfor R110, green HeNe 543.8 nm for Tamra, yellow HeNe594 nm for Texas Red and red HeNe 632.8 nm for Cy5. Thesignal intensities were determined with the QuantArrayanalysis 3.1 software (PerkinElmer Life Sciences). The SNPgenotype were assigned using the SNPSnapper software version3.0.0.191 (http://www.bioinfo.helsinki.fi/SNPsnapper/) basedon scatter plots with the logarithm of the sum of bothfluorescence signals (SAllele1þ SAllele2) on the vertical axis andthe fluorescence signal fractions (SAllele1/(SAllele1þ SAllele2))on the horizontal axis. ASE for each heterozygous individualwas determined by calculating the fluorescence signalbetween the two alleles (SAllele1/(SAllele1þ SAllele2)) in bothcDNA and the respective gDNA as a reference for eachheterozygous liver sample. The mean signal intensity of theduplicate spots run in triplicates on one subarray wasconsidered as one replicate assay.

Determination of DNA methylation frequency by pyrosequencingAssay design. Searching for CpG islands in the CYP1A2 genecovering up to 5000 bp upstream region from the translationstart codon was carried out using Methyl Primer Expresssoftware (Applied Biosystems, Foster City, CA, USA). Wefound one CpG island close to the transcription start site inexon 2. Seventeen of the CpG sites located in the center ofthe island were selected for DNA methylation frequencyanalysis (Figure 1).

Bisulfite treatment and PCRThe gDNA isolated from human liver samples was treatedwith sodium bisulfite using the EZ DNA Methylation Kit


214


(Zymo Research Corporation, Orange, CA, USA). PCRprimers were designed to amplify the CpG island containing17 CpG sites located in exon 2. A 309-bp region spanningthe island was amplified by PCR from bisulfate-treatedgDNA samples. The PCR mix contained 2.5ml 10� PCRbuffer, 0.5 ml 25 mM MgCl2 solution, 0.125 ml Smart-Taq HotThermostable DNA Polymerase (Naxo Ltd., Tartu, Estonia),0.5ml 10 mM of forward (50-TTTGGTATTGTTAAGGATGAGTTAG-30) and biotinylated reverse (50-ACATACTCCTCCAAATAACAAAAA-30) primers (Invitrogen, Carlsbad, CA, USA),and 0.5 ml 10 mM dNTP (GE Healthcare), 1 ml bisulfite-treatedDNA and sterile water in a final volume of 25 ml. The PCRconditions were initial activation of the Smart-Taq Hotenzyme at 95 1C for 10 min followed by 50 cycles of 95 1C for30 s, 53 1C for 1 min, 72 1C for 1 min and a final extension at72 1C for 7 min on a GeneAmp PCR system 2700 (AppliedBiosystems). The PCR products were analyzed on a 1.5%agarose gel for quality assurance and verification beforepyrosequencing.

Pyrosequencing

The amplified PCR product was analyzed by pyrosequencingusing six internal sequencing primers designed to cover allthe 17 CpG sites. Sequencing primers, CpG sites andpyrosequencing dispensation orders including internal con-trol for bisulfite treatment are listed in Table 3. Pyrosequen-cing was performed on Pyro Q-CpG software (Biotage,Uppsala, Sweden) using the PyroGold SQA Reagent Kit(Biotage) as described previously.45 The software reports apeak height directly proportional to the number of mole-cules incorporated into the growing DNA chain.

AzaC treatment and total RNA isolation from cultured cellsThe human hepatoma cell lines B16A2 were cultured basedon ATCC protocols. All cell culture media and supplementswere obtained from Invitrogen (Rockville, MD, USA). Cellswere treated with AzaC (Sigma-Aldrich, Steinheim,Germany) at varying concentrations (0, 100, 500 and1000mM) for 5 days. B16A2 cells were allowed to reach totalconfluence and maintained at such level for 1 week beforeAzaC treatment was commenced. Subsequent to AzaCtreatments, total RNA was isolated using Trizol reagent

(Invitrogen) following manufacturer’s recommendations.Reverse transcription was carried out from 5 mg total RNAusing the SuperScript II (Invitrogen) system in the presenceof the ribonuclease inhibitor RNaseOUT (Invitrogen).

Quantitative real-time PCRThe quantification of the total CYP1A2 mRNA levels fromhuman livers and B16A2 cell lines treated with varyingconcentrations of AzaC was performed by real-time PCRusing 7500 Fast Real-Time PCR system. The quantificationmixture contained 7.5ml of TaqMan 2� PCR Master Mix,0.5 ml of cDNA, 0.75 ml of 20� mix of predesigned TaqManGene Expressions Assay contained primers for CYP1A2mRNA spanning sequence boundary between exons 6 and7 and probe in a final 15ml volume of reaction. Each samplewas ran in triplicates and human 18S rRNA 20� was used asendogenous control. The real-time PCR instrument, TaqManmaster mix, primers and probe were purchased from AppliedBiosystems.

Determination of CYP1A2 enzyme activity in human livermicrosomes

Microsomes were prepared from 21 human livers bank asdescribed previously.46 The pellet containing the microso-mal fraction was suspended and homogenized in 50 mM

potassium phosphate buffer, pH 7.4, to give a finalconcentration of 30–50 mg microsomal protein per ml andstored at �80 1C until use. The protein content wasdetermined by Lowry method,47 and the total cytochromeP450 content was quantified.48 The CYP1A2 enzyme activitywas estimated by the using caffeine, EROD and phenacetinas marker probes (Sigma-Aldrich, St Louis, MO, USA). Assayconditions used for CYP1A2 enzyme activity determinationin human liver microsomes using the three different probesare summarized in Table 3. Microsomes were diluted withbuffer and preincubated for 3 min at 37 1C before the probesubstrate was added. Multiple control incubations wereperformed with boiled microsomes, without microsomes,without nicotinamide adenine dinucleotide phosphate(NADPH) or by adding stop solution before NADPH.

Table 3 Summary of assay conditions used for CYP1A2 enzyme activity determination in human liver microsomes using threedifferent probes: caffeine, phenacetin and EROD

Activity measured Probeconc.(mM)

Solvent finalconc.

Protein(mg ml�1)

Buffer Cofactorfinal conc.

Finalincubation

volume (ml)

Incubationtime (min)

Analytical assay

EthoxyresorufinO-deethylation

5 0.5% DMSO 0.05 100 mM potassiumphosphate (pH 7.8)

0.25 mM

NADPH0.2 20 Fluorescence recorded

in real time42

CaffeineN-demethylation

100 — 0.5 100 mM potassiumphosphate (pH 7.4)

1 mM

NADPH0.5 15 HPLC43

PhenacetinO-deethylation

200 2.5%Acetonitrile

0.2 100 mM potassiumphosphate (pH 7.4)

1 mM

NADPH0.2 15 LC–MS44

Abbreviations: DMSO, dimethyl sulfoxide; HPLC, high-pressure liquid chromatography; LC–MS, liquid chromatography-mass spectrometry; NADPH, nicotinamide

adenine dinucleotide phosphate.


215


CYP1A2 protein content determination

Determination of CYP1A2 protein content was carried out asdescribed previously.49 In brief, a total of 5 mg protein wasseparated by SDS–polyacrylamide gel electrophoresis using a10% gel. After transfer, the Hybond-C extra membrane(Amersham Pharmacia Biotech, Uppsala, Sweden) wasblocked in Tris-buffered saline containing 0.05% (v/v)Tween 20 and 5% fat-free milk and incubated with 1:2500dilution of the primary CYP1A1/CYP1A2 polyclonal anti-body (Daiichi Pure Chemicals Co. Ltd., Ibaraki, Japan). Thiswas followed by incubation with a 1:2500 diluted goat anti-rabbit horseradish peroxidase-conjugated antibody (Pierce,Rockford, IL, USA). Detection was performed using Super-Signal West Pico Chemiluminescent Substrate (Pierce).

StatisticsA two-tailed Student’s t-test was used to compare the meanallelic signal (SAllele1/(SAllele1þ SAllele2)) between gDNA andcDNA for each heterozygous liver. Analysis of variance wasused to analyze association between CYP1A2 genotype,mRNA expression protein content and enzyme activity,using all three marker probes. Haplotype analysis andestimation of expected haplotype frequencies from the rawgenotype data were carried out using population geneticsoftware program Arlequin version 2.000. The Spearman’srank correlation test and regression analysis (after convert-ing to log values) were used to analyze association betweenmean methylation frequency with ASE phenotype and totalmRNA level.

Duality of interest

None declared.

Abbreviations

CYP1A2 cytochrome P450, family 1, subfamily A, polypeptide 2ASE allele-specific expressionCpG cytosine-guanineSNP single nucleotide polymorphismEROD 7-ethoxyresorufin

Acknowledgments

This study was financially supported by the Swedish ResearchCouncil for Medicine, the Swedish Research Council for Science andTechnology, and by Torsten and Ragnar Soderbergs Stiftelser.

References

1 Shimada T, Yamazaki H, Mimura M, Inui Y, Guengerich FP. Inter-individual variations in human liver cytochrome P-450 enzymesinvolved in the oxidation of drugs, carcinogens and toxic chemicals:studies with liver microsomes of 30 Japanese and 30 Caucasians.J Pharmacol Exp Ther 1994; 270: 414–423.

2 Bertilsson L, Carrillo JA, Dahl ML, Llerena A, Alm C, Bondesson U et al.Clozapine disposition covaries with CYP1A2 activity determined by acaffeine test. Br J Clin Pharmacol 1994; 38: 471–473.

3 Tassaneeyakul W, Birkett DJ, Veronese ME, McManus ME, Tukey RH,Quattrochi LC et al. Specificity of substrate and inhibitor probes forhuman cytochromes P450 1A1 and 1A2. J Pharmacol Exp Ther 1993;265: 401–407.

4 Kroemer HK, Gautier JC, Beaune P, Henderson C, Wolf CR, EichelbaumM. Identification of P450 enzymes involved in metabolism of verapamilin humans. Naunyn Schmiedebergs Arch Pharmacol 1993; 348: 332–337.

5 Hartter S, Ursing C, Morita S, Tybring G, von Bahr C, Christensen M etal. Orally given melatonin may serve as a probe drug for cytochromeP450 1A2 activity in vivo: a pilot study. Clin Pharmacol Ther 2001; 70:10–16.

6 Yamazaki H, Shaw PM, Guengerich FP, Shimada T. Roles ofcytochromes P450 1A2 and 3A4 in the oxidation of estradiol andestrone in human liver microsomes. Chem Res Toxicol 1998; 11:659–665.

7 McManus ME, Burgess WM, Veronese ME, Huggett A, Quattrochi LC,Tukey RH. Metabolism of 2-acetylaminofluorene and benzo(a)pyreneand activation of food-derived heterocyclic amine mutagens by humancytochromes P-450. Cancer Res 1990; 50: 3367–3376.

8 Eaton DL, Gallagher EP, Bammler TK, Kunze KL. Role of cytochromeP4501A2 in chemical carcinogenesis: implications for human variabilityin expression and enzyme activity. Pharmacogenetics 1995; 5: 259–274.

9 Schweikl H, Taylor JA, Kitareewan S, Linko P, Nagorney D, Goldstein JA.Expression of CYP1A1 and CYP1A2 genes in human liver. Pharmacoge-netics 1993; 3: 239–249.

10 Rodriguez-Antona C, Donato MT, Pareja E, Gomez-Lechon MJ, CastellJV. Cytochrome P-450 mRNA expression in human liver and itsrelationship with enzyme activity. Arch Biochem Biophys 2001; 393:308–315.

11 Jiang Z, Dragin N, Jorge-Nebert LF, Martin MV, Peter Guengerich F,Aklillu E et al. Search for an association between the human CYP1A2genotype and CYP1A2 metabolic phenotype. Pharmacogenet Genomics2006; 16: 359–367.

12 Ghotbi R, Christensen M, Roh HK, Ingelman-Sundberg M, Aklillu E,Bertilsson L. Comparisons of CYP1A2 genetic polymorphisms, enzymeactivity and the genotype-phenotype relationship in Swedes andKoreans. Eur J Clin Pharmacol 2007; 63: 537–546.

13 Aklillu E, Carrillo JA, Makonnen E, Hellman K, Pitarque M, Bertilsson Let al. Genetic polymorphism of CYP1A2 in Ethiopians affectinginduction and expression: characterization of novel haplotypes withsingle-nucleotide polymorphisms in intron 1. Mol Pharmacol 2003; 64:659–669.

14 Djordjevic N, Ghotbi R, Bertilsson L, Jankovic S, Aklillu E. Induction ofCYP1A2 by heavy coffee consumption in Serbs and Swedes. Eur J ClinPharmacol 2008; 64: 381–385.

15 Wray GA, Hahn MW, Abouheif E, Balhoff JP, Pizer M, Rockman MV et al.The evolution of transcriptional regulation in eukaryotes. Mol Biol Evol2003; 20: 1377–1419.

16 Lo HS, Wang Z, Hu Y, Yang HH, Gere S, Buetow KH et al. Allelic variationin gene expression is common in the human genome. Genome Res2003; 13: 1855–1862.

17 Pastinen T, Sladek R, Gurd S, Sammak A, Ge B, Lepage P et al. A surveyof genetic and epigenetic variation affecting human gene expression.Physiol Genomics 2004; 16: 184–193.

18 Yan H, Zhou W. Allelic variations in gene expression. Curr Opin Oncol2004; 16: 39–43.

19 Serre D, Gurd S, Ge B, Sladek R, Sinnett D, Harmsen E et al. Differentialallelic expression in the human genome: a robust approach to identifygenetic and epigenetic cis-acting mechanisms regulating gene expres-sion. PLoS Genet 2008; 4: e1000006.

20 Tao H, Cox DR, Frazer KA. Allele-specific KRT1 expression is a complextrait. PLoS Genet 2006; 2: e93.

21 Johnson AD, Zhang Y, Papp AC, Pinsonneault JK, Lim JE, Saffen D et al.Polymorphisms affecting gene transcription and mRNA processing inpharmacogenetic candidate genes: detection through allelic expressionimbalance in human target tissues. Pharmacogenet Genomics 2008; 18:781–791.

22 Hirota T, Ieiri I, Takane H, Maegawa S, Hosokawa M, Kobayashi K et al.Allelic expression imbalance of the human CYP3A4 gene and individualphenotypic status. Hum Mol Genet 2004; 13: 2959–2969.

23 Nakajima M, Iwanari M, Yokoi T. Effects of histone deacetylation andDNA methylation on the constitutive and TCDD-inducible expressions


216


of the human CYP1 family in MCF-7 and HeLa cells. Toxicol Lett 2003;144: 247–256.

24 Botto F, Seree E, el Khyari S, de Sousa G, Massacrier A, Placidi M et al.Tissue-specific expression and methylation of the human CYP2E1 gene.Biochem Pharmacol 1994; 48: 1095–1103.

25 Hammons GJ, Yan-Sanders Y, Jin B, Blann E, Kadlubar FF, Lyn-Cook BD.Specific site methylation in the 50-flanking region of CYP1A2 inter-individual differences in human livers. Life Sci 2001; 69: 839–845.

26 Saxonov S, Berg P, Brutlag DL. A genome-wide analysis of CpGdinucleotides in the human genome distinguishes two distinct classes ofpromoters. Proc Natl Acad Sci USA 2006; 103: 1412–1417.

27 Cassidy SB, Dykens E, Williams CA. Prader-Willi and Angelmansyndromes: sister imprinted disorders. Am J Med Genet 2000; 97:136–146.

28 Robinson WP, Horsthemke B, Leonard S, Malcolm S, Morton C, NichollsRD et al. Report of the Third International Workshop on HumanChromosome 15 Mapping 1996. October 25-27, 1996 in VancouverB.C., Canada. Cytogenet Cell Genet 1997; 76: 1–13.

29 Nicholls RD, Knoll JH, Butler MG, Karam S, Lalande M. Geneticimprinting suggested by maternal heterodisomy in nondeletion Prader-Willi syndrome. Nature 1989; 342: 281–285.

30 Milani L, Gupta M, Andersen M, Dhar S, Fryknas M, Isaksson A et al.Allelic imbalance in gene expression as a guide to cis-acting regulatorysingle nucleotide polymorphisms in cancer cells. Nucleic Acids Res 2007;35: e34.

31 Pinsonneault JK, Papp AC, Sadee W. Allelic mRNA expression of X-linkedmonoamine oxidase a (MAOA) in human brain: dissection of epigeneticand genetic factors. Hum Mol Genet 2006; 15: 2636–2649.

32 Hardison RC, Oeltjen J, Miller W. Long human-mouse sequencealignments reveal novel regulatory elements: a reason to sequencethe mouse genome. Genome Res 1997; 7: 959–966.

33 Jaenisch R, Bird A. Epigenetic regulation of gene expression: how thegenome integrates intrinsic and environmental signals. Nat Genet 2003;33: (Suppl) 245–254.

34 Shen L, Kondo Y, Guo Y, Zhang J, Zhang L, Ahmed S et al. Genome-wide profiling of DNA methylation reveals a class of normallymethylated CpG island promoters. PLoS Genet 2007; 3: 2023–2036.

35 Reik W, Dean W, Walter J. Epigenetic reprogramming in mammaliandevelopment. Science 2001; 293: 1089–1093.

36 Lees-Murdock DJ, Walsh CP. DNA methylation reprogramming in thegerm line. Epigenetics 2008; 3: 5–13.

37 Bird A. DNA methylation patterns and epigenetic memory. Genes Dev2002; 16: 6–21.

38 Waterland RA, Jirtle RL. Early nutrition, epigenetic changes attransposons and imprinted genes, and enhanced susceptibility to adultchronic diseases. Nutrition 2004; 20: 63–68.

39 Yan H, Yuan W, Velculescu VE, Vogelstein B, Kinzler KW. Allelic variationin human gene expression. Science 2002; 297: 1143.

40 Lindroos K, Sigurdsson S, Johansson K, Ronnblom L, Syvanen AC.Multiplex SNP genotyping in pooled DNA samples by a four-colourmicroarray system. Nucleic Acids Res 2002; 30: e70.

41 Lovmar L, Fredriksson M, Liljedahl U, Sigurdsson S, Syvanen AC.Quantitative evaluation by minisequencing and microarrays revealsaccurate multiplexed SNP genotyping of whole genome amplifiedDNA. Nucleic Acids Res 2003; 31: e129.

42 Burke MD, Mayer RT. Ethoxyresorufin: direct fluorimetric assay of amicrosomal O-dealkylation which is preferentially inducible by 3-methylcholanthrene. Drug Metab Dispos 1974; 2: 583–588.

43 Tassaneeyakul W, Mohamed Z, Birkett DJ, McManus ME, Veronese ME,Tukey RH et al. Caffeine as a probe for human cytochromesP450: validation using cDNA-expression, immunoinhibition andmicrosomal kinetic and inhibitor techniques. Pharmacogenetics 1992;2: 173–183.

44 Distlerath LM, Reilly PE, Martin MV, Davis GG, Wilkinson GR,Guengerich FP. Purification and characterization of the human livercytochromes P-450 involved in debrisoquine 4-hydroxylation andphenacetin O-deethylation, two prototypes for genetic polymorphismin oxidative drug metabolism. J Biol Chem 1985; 260: 9057–9067.

45 Tost J, Gut IG. Analysis of gene-specific DNA methylation patternsby pyrosequencing technology. Methods Mol Biol 2007; 373:89–102.

46 von Bahr C, Groth CG, Jansson H, Lundgren G, Lind M, Glaumann H.Drug metabolism in human liver in vitro: establishment of a human liverbank. Clin Pharmacol Ther 1980; 27: 711–725.

47 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurementwith the Folin phenol reagent. J Biol Chem 1951; 193: 265–275.

48 Omura T, Sato R. The carbon monoxide-binding pigment of livermicrosomes, I. Evidence of its hemoprotein nature. J Biol Chem 1964;239: 2370–2378.

49 Karlgren M, Gomez A, Stark K, Svard J, Rodriguez-Antona C, Oliw E etal. Tumor-specific expression of the novel cytochrome P450 enzyme,CYP2W1. Biochem Biophys Res Commun 2006; 341: 451–458.


217


Allele-specific expression and gene methylation in the control of CYP1A2 mRNA level in human livers

Documents

Transcript of Allele-specific expression and gene methylation in the control of CYP1A2 mRNA level in human livers