Download Post-mortem SNP analysis of CYP2D6 gene reveals correlation

Forensic Science International 135 (2003) 9–15
Post-mortem SNP analysis of CYP2D6 gene reveals correlation
between genotype and opioid drug (tramadol)
metabolite ratios in blood
Antti Levoa, Anna Koskib, Ilkka Ojanperäb, Erkki Vuorib, Antti Sajantilaa,*
a
Laboratory of Forensic Biology, Department of Forensic Medicine, P.O. Box 40, 00014, University of Helsinki, Finland
b
Laboratory of Toxicology, Department of Forensic Medicine, P.O. Box 40, 00014, University of Helsinki, Finland
Received 20 May 2002; accepted 7 March 2003
Abstract
Tramadol is an opioid drug metabolised in phase I by cytochrome P450 (CYP) enzymes, of which CYP2D6 is mainly responsible
for the O-demethylation of tramadol, but is not involved in N-demethylation. Defects in the genes encoding drug metabolising
enzymes (DMEs) may lead to adverse drug effects, even to death. To aid interpretation of the forensic toxicology results, we studied
how the genetic variation of the CYP2D6 gene is reflected in tramadol metabolite ratios found in post-mortem samples.
In 33 Finnish autopsy cases where tramadol was found, we analysed both the CYP2D6 genotype and the concentrations of
tramadol and its metabolites O- and N-demethyltramadol. As expected, we found a correlation between the number of functional
CYP2D6 alleles and the ratio of tramadol to O-demethyltramadol. We also found a correlation between the number of functional
alleles and the ratio of tramadol to N-demethyltramadol. This can be explained by the complementary nature of the two main
tramadol demethylation pathways. No known CYP2D6 inhibitors were associated with exceptional metabolic ratios.
Furthermore, no accidental tramadol poisonings were associated with a defective CYP2D6 gene.
Our results on the tramadol are among the first to demonstrate that genetic variation in drug metabolising enzymes can be
analysed in post-mortem blood, and that it correlates well with the parent drug to metabolite ratios. The results also suggest that
genetic factors play, in general, a dominant role over other factors in the metabolism of individual drugs.
# 2003 Elsevier Ireland Ltd. All rights reserved.
Keywords: SNP analysis; CYP2D6 gene; Tramadol; Pharmacogenetics
1. Introduction
Drug metabolising enzymes (DMEs) show genetic variation, leading to inter-individual differences in response to
drugs and consequently adverse side effects [1–3]. One of
the best-characterised genetic factors affecting drug metabolism in humans is the variation in the cytochrome P450
CYP2D6 gene on the chromosome 22q13.1 (MIM124030).
CYP2D6 variation due to DNA-level point mutations and
large rearrangements determines at least four groups of
metabolic rate for drug elimination [1,4]. It has been estimated
that approximately 5–10% of the European population is
*
Corresponding author. Tel.: þ358-9-19127472;
fax: þ358-9-19127518.
E-mail address: [email protected] (A. Sajantila).
homozygous for a defective CYP2D6 gene and therefore
‘poor metabolisers’ [5]. Such individuals may exhibit, through
the accumulation of a parent drug or its metabolites, toxic side
effects or a lack of response during drug treatment. Another
risk group consists of individuals whose metabolism is
substantially accelerated, e.g. due to more than two functional
copies of the CYP2D6 gene. They require higher than usual
doses to achieve therapeutic parent drug levels in the blood.
Here we have studied how genetic variation of DME,
when analysed in post-mortem blood, correlates with parent
drug to metabolite ratios. In order to this phenomenon, we
have focused on the opioid drug tramadol, since it is commonly used, its metabolism is well-characterized and the
concentrations of the parent drug and the main metabolites
are readily analysed. Furthermore, we have developed a
dedicated single nucleotide polymorphism (SNP) typing
0379-0738/$ – see front matter # 2003 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/S0379-0738(03)00159-2
10
A. Levo et al. / Forensic Science International 135 (2003) 9–15
method to facilitate the genotyping of the CYP2D6 gene,
which is mainly responsible for the O-demethylation pathway of tramadol metabolism.
2. Materials and methods
2.1. Subjects
The subjects were selected from all of the autopsies
performed at the Department of Forensic Medicine, University of Helsinki, between June 1998 and June 2000. The
selection criterion was a positive result for tramadol in the
routine drug screening of post-mortem blood. The study
group consisted of 11 males and 22 females, who had died
unexpectedly. Subjects were 23–91 years old (mean 65 years,
median 70 years). Autopsy findings, hospital background
information and police department records were used to
establish the cause and manner of death as well as additional
relevant information on diseases or pathological conditions.
2.2. CYP2D6 genotyping
From each autopsy case DNA was extracted according to
manufacturer’s recommendations from a blood stain dried
on dedicated paper (FTA GeneCard, Life Technologies, Inc.,
cat no. 10786-036) and preserved in the archives as a part of
documentation in the quality assurance process of the
autopsy routine.
The CYP2D6 gene locus was screened for large deletions
or amplifications and 18 selected SNPs [5]. The basic
genotyping approach was modified from Sachse et al. [6],
Løvlie et al. [7], and Steen et al. [8]. Three parallel long PCR
amplifications were run for each sample—the first reaction
amplified a 4.7 kb fragment over the whole CYP2D6
sequence, while the second and third reactions were specific
for duplications and deletions, respectively. All the positive
results in either the duplication-specific (3.6 kb) or deletionspecific (3.5 kb) amplifications were confirmed by an additional long PCR amplification (fragments of 10.0 or 9.5 kb)
using an independent set of allele-specific primers [9,10].
The SNP typing along the CYP2D6 sequence was accomplished by restriction enzyme digestions subsequent to nested
CYP2D6-specific amplifications. For each sample, eight separate fragments of 186–471 bp were re-amplified from the
4.7 kb amplificate and digested by a panel of 14 restriction
enzymes (HphI, MspI, BspMI, PstI, BsmAI, BstNI, BsaAI,
MboII, HinPI, FokI, BanII/BstEI, HaeII, SacII, EagI). Distinction between duplicated/amplified alleles 1 N, 2 N,
and 4 N was as in Sachse et al. [6]. Since DNA may be
highly degraded or modified in post-mortem samples,
we verified the SNPs defining alleles CYP2D6 3, 4, and
6 by allele-specific PCR [11] in all mutation-positive
samples. These results were consistent with those obtained
Table 1
RFLP detection of CYP2D6 alleles, primer sequences used, expected PCR fragment length and restriction enzyme for detection of the allele
indicator mutation
CYP2D6 position
CYP2D6 allele
Sequence of the primer (50 to 30 )
Fragment length (bp)
Restriction enzyme
E1 100 C > T
E1 124 G > A
E1 138 insT
4, 10, 14
12
15
CAGTCAACACAGCAGGTTCA
CCTGTGGTTTCACCCACCAT
437
HphI
MspI
BspMI
I1 883 G > C
E2 974 C > A
E2 984 A > G
E2 997 C > G
E2 1023 C > T
11
4A, 4B
4A, 4B
4A, 4B
17
TGCTCACTCCTGGTAGCC
CCCGGGTCCCACGGAAATCT
323
PstI
HaeII
SacII
EagI
HphI
E3 1661 G > C
TAATGCCTTCATGGCCACGCG
GAGACTCCTCGGTCTCTCG
471
BsmAI
E3 1707 T > del
E3 1758 G > T/A
I3 1846 G > A
6
8, 14
4
CCTGGGCAAGAAGTCGCTGGACCAG
GAGACTCCTCGGTCTCTCG
352
BstNI
MspI
BstNI
E5 2549 A > del
3
GCTGGGGCCTGAGACTT
GGCTGGGTCCCAGGTCATAC
200
BsaAI
E5 2613–15 delAGA
E6 2850 C > T
9
2
AGGCCTTCCTGGCAGAGATGAAG
CCCCTGCACTGTTTCCCAGA
386
MboII
HinP1I
E6 2935 A > C
7
CCCGTTCTGTCCCGAGTATG
GGGCTCACGCTGCACATCAGGA
186
FokI
E9 4180 G > C
2
GCCACCATGGTGTCTTTGCTTTC
CTCAGCCTCAACGTACCCCT
263
BanII/BstEII
2
A. Levo et al. / Forensic Science International 135 (2003) 9–15
by PCR-RFLP. The SNP positions, alleles, PCR primers,
amplification product lengths and restriction enzyme data
are given in Table 1.
2.3. Determination of tramadol and its metabolites
O-demethyltramadol and nortramadol
(N-demethyltramadol)
The venous blood samples were obtained at autopsy
and stored in polyethylene test tubes containing 1% of NaF
11
at 4 8C. Tramadol was determined in 1 ml of blood submitted to routine drug screening. The method involved
extraction at pH 9 into butyl acetate and subsequent analysis
by dual-column gas chromatography with nitrogen-selective
detection [12]. The method was linear over a range of
0.05–10 mg/l, with a limit of quantitation of 0.1 mg/l.
The metabolites O-demethyltramadol and nortramadol were
determined by a dedicated method in 1 ml blood samples
stored frozen. The extraction was carried out at pH 9
with dichloromethane isopropyl alcohol, and the analysis
Table 2
Summary of CYP2D6 genotyping, tramadol analysis and autopsy findings in the 33 cases [17,27]
12
A. Levo et al. / Forensic Science International 135 (2003) 9–15
Table 2 (Continued )
involved a liquid-chromatographic separation on a C18
column and detection by tandem mass spectrometry using
multiple reaction monitoring. Linear calibration was used
from 0.0025 to 0.3 mg/l and quadratic calibration above
0.3 mg/l. The cut-off limit of quantitation was 0.01 mg/l.
2.4. Statistical analysis
The individuals were classified according to the number
of functional alleles, and for each individual the tramadol/
O-demethyltramadol and tramadol/nortramadol ratios were
determined. For each genotype group the median metabolite
ratios along with their 95% confidence intervals were
plotted, after a logarithmic transformation, as a function
of the number of alleles. The statistical differences between
the medians were calculated using the Mann–Whitney test
included in the MINITAB 13.30 software package.
3. Results and discussion
The commonly used opioid drug tramadol is eliminated
mainly in urine via O- and N-demethylation [13]. According
to current knowledge [14–16], the CYP2D6-mediated
O-demethylation of tramadol is the major metabolic route
with functional polymorphism, whereas N-demethylation
represents a not yet fully characterised pathway. We studied
post-mortem blood samples for the inter-individual variation
in the CYP2D6 gene and its effect on the metabolism of
tramadol. We determined two major genetic rearrangements
and 18 SNPs along the CYP2D6 gene [5,17] in 33 cases,
in which tramadol was found in forensic toxicology analysis (Table 2). Genotyping enabled us to classify these
individuals into four groups, according to the number of
functional alleles in their CYP2D6 gene. The study groups
included those with no functional alleles (Group 0, n ¼ 4),
A. Levo et al. / Forensic Science International 135 (2003) 9–15
13
Fig. 1. Median (a) tramadol/O-demethyltramadol (MR1) and (b) tramadol/nortramadol (MR2) concentration ratios with 95% confidence
intervals in post-mortem blood of four groups of CYP2D6 genotypes. Group 0: no functional alleles, Group 1: one functional allele, Group 2:
two functional alleles, Group 3: three or more functional alleles.
one functional allele (Group 1, n ¼ 9), two functional alleles
(Group 2, n ¼ 16), and three or more copies of functional
alleles (Group 3, n ¼ 4). Distribution of alleles corresponds
roughly to that found in the random Caucasian population
[6,18], although slight overpresentation of UM genotypes
(12% versus 1–5% in Caucasians) and consequent deviation
in other genotypes was observed, most probably due to a
limited sample size for a population genetic study.
We analysed the concentration of the parental form of
tramadol and its metabolite O-demethyltramadol in the blood
and calculated the tramadol/O-demethyltramadol metabolite
ratio (MR1). When the number of functional alleles increased,
the median MR1 decreased (Fig. 1a). Moreover, the median
MR1 of Group 0 was significantly different from those of
Groups 2 and 3 (P < 0:01 and P ¼ 0:03, respectively).
To further characterise the effect of CYP2D6 polymorphism
on tramadol metabolism, we analysed the concentration of
nortramadol in the blood and calculated the tramadol/nortramadol metabolite ratio (MR2). The median MR2 also correlated with the number of functional alleles, but in the reverse
direction (Fig. 1b), as can be expected based on the complementary nature of these two pathways. Again, the median MR2
in Group 0 was significantly different from the medians in
Groups 2 and 3 (P ¼ 0:01 and P ¼ 0:03, respectively).
When the frequency distributions of MR1 and MR2 in the
genotype groups was calculated, the distribution of MR1 in
particular supported the relationship between the metabolite
ratios and the CYP2D6 genotype groups (Fig. 2).
Drug pharmacokinetics may be affected by several other
factors besides genetic variation, especially by metabolic
drug interactions, age, and renal or liver malfunction [3,19].
Furthermore, post-mortem pharmacokinetic determinations
are inevitably limited to one-time sampling instead of
measuring the area under the curve, which may complicate
interpretation. In case of tramadol, the elimination half lives
of the parent compound and O-demethyltramadol are fairly
similar (6 h) [14] and consequently no large changes in
MR1 are expected in the course of antemortem time. The 33
cases in this study showed a fairly high median age of 70
years. Among the cases in Groups 1–3, in which enzyme
inhibition by other drugs may alter metabolism, we generally found no connection between the presence of known
CYP2D6 inhibitors or substrates [16] and exceptionally
high MR1 or low MR2 values. However, one case (99 0403)
in Group 1 involved possible metabolic inhibition by
diphenhyldramine (0.8 mg/l in the blood), levomepromazine (0.3 mg/l) and trimipramine (0.1 mg/l), with a MR1
value as high as 80.0, while the Group 1 median was 8.1.
Among the cases studied was one fatal drug poisoning due
to tramadol alone (9 mg/l), as judged by a forensic pathologist. In five cases, tramadol and its metabolites were the
only drugs present, while in the remaining cases 2–10 other
drugs were found. High tramadol concentrations, 2 mg/l
or more, were found in ten cases. Such high parent drug
concentrations could not be explained in terms of the
CYP2D6 polymorphisms studied or drug interactions.
14
A. Levo et al. / Forensic Science International 135 (2003) 9–15
Fig. 2. Frequency distribution of tramadol/O-demethyltramadol concentration ratios (MR1) in post-mortem blood of four groups of CYP2D6
genotypes (for group names, see Fig. 1).
Instead, advanced age coupled with multiple diseases
appeared to be a common factor in these cases.
Several studies have shown the significance of CYP2D6
polymorphism in clinical settings [20,21]. The corresponding post-mortem literature is scarce. There is one report on a
fluoxetine-related death of a child with a deficient CYP2D6
gene detected in the post-mortem investigation [22]. Two
other studies fail to correlate the genotype with the suggested
phenotype [23,24]. Obviously, when post-mortem material
is used instead of intact, high-quality clinical samples, the
methods of genetic analysis should meet higher standards to
provide useful information.
Postmortem samples are routinely taken for toxicological
analysis in cases where the autopsy findings or background
information indicate poisoning. These samples are intriguing both for studying the effect of DMEs in drug metabolism and for estimating the contribution of genetic factors
to poisonings in general. It has been suggested that behavioural risk factors and the availability of drugs have a major
impact in suicidal poisonings [25]. However, the effect of
genetic factors could explain those cases where intentional
overdose can reasonably be ruled out, but toxicological
analysis reveals an unexpectedly high concentration of the
parent drug, or an exceptional parent drug/metabolite ratio,
suggesting accidental poisoning.
To conclude, our results on tramadol are among the first to
demonstrate that analysis of genetic variation of DMEs
using post-mortem blood is both feasible and relevant for
the evaluation of adverse drug effects. The results indicate
that the dominant role of genetic factors in the metabolism of
individual drugs can still be seen post-mortem, since the
average metabolite ratios were not obscured by various
pathological conditions, by the presence of potentially interacting substances. Our data also suggests that the alternative
N-demethylation route can prevent the accumulation of the
parent drug in ‘poor metabolisers’ [15,26]. We anticipate
that the approach described in this study, when applied to
drugs of higher toxicity, such as antidepressants and antipsychotics, will aid in the interpretation of poisoning cases
in forensic and clinical settings
Acknowledgements
We are grateful to the University of Helsinki, Instrumentarium Scientific Fund, Ella och Georg Ehrnrooths Stiftelse,
Paulo Foundation and Research and Science Foundation
of Farmos. We acknowledge Grünenthal GmbH (Aachen,
Germany) for kindly providing us with the tramadol metabolites. We also thank Merja Gergov and Ilpo Rasanen for
A. Levo et al. / Forensic Science International 135 (2003) 9–15
performing the chemical analyses, Kirsti Höök for assistance
in the DNA analyses, and Johanna Sistonen for generous
help in the final steps of the project.
[13]
[14]
References
[15]
[1] U.A. Meyer, U.M. Zanger, Molecular mechanisms of genetic
polymorphisms of drug metabolism, Annu. Rev. Pharmacol.
Toxicol. 37 (1997) 269–296.
[2] W.E. Evans, M.V. Relling, Pharmacogenomics: translating
functional genomics into rational therapeutics, Science 286
(5439) (1999) 487–491.
[3] M. Ingelman-Sundberg, M. Oscarson, R.A. McLellan,
Polymorphic human cytochrome P450 enzymes: an opportunity for individualized drug treatment, Trends Pharmacol. Sci.
20 (8) (1999) 342–349.
[4] H.-G. Xie, R.B. Kim, A.J.J. Wood, C.M. Stein, Molecular
basis of ethnic differences in drug disposition and response,
Annu. Rev. Pharmacol. Toxicol. 41 (2001) 815–850.
[5] A.K. Daly, J. Brockmöller, F. Broly, M. Eichelbaum, W.E.
Evans, F.J. Gonzalez, J.D. Huang, J.R. Idle, M. IngelmanSundberg, T. Ishizaki, E. Jacqz-Aigrain, U.A. Meyer, D.W.
Nebert, V.M. Steen, C.R. Wolf, U.M. Zanger, Nomenclature
for human CYP2D6 alleles, Pharmacogenetics 6 (3) (1996)
193–210.
[6] C. Sachse, J. Brockmöller, S. Bauer, I. Roots, Cytochrome
P450 2D6 variants in a Caucasian population: allele
frequencies and phenotypic consequences, Am. J. Hum.
Genet. 60 (2) (1997) 284–295.
[7] R. Løvlie, A.K. Daly, A. Molven, J.R. Idle, V.M. Steen,
Ultrarapid metabolizers of debrisoquine: characterization and
PCR-based detection of alleles with duplication of the
CYP2D6 gene, FEBS Lett. 392 (1) (1996) 30–34.
[8] V.M. Steen, O.A. Anreassen, A.K. Daly, T. Tefre, A.-L.
Børresen, J.R. Idle, A.-K. Gulbrandsen, Detection of the
poor metabolizer-associated CYP2D6(D) gene deletion allele
by long-PCR technology, Pharmacogenetics 5 (4) (1995)
215–223.
[9] A.K. Daly, K.S. Fairbrother, O.A. Andreassen, S.J. London,
J.R. Idle, V.M. Steen, Characterization and PCR-based
detection of two different hybrid CYP2D7P/CYP2D6 alleles
associated with the poor metabolizer phenotype, Pharmacogenetics 6 (4) (1996) 319–328.
[10] I. Johansson, E. Lundqvist, M.-L. Dahl, M. IngelmanSundberg, PCR-based genotyping for duplicated and deleted
CYP2D6 genes, Pharmacogenetics 6 (4) (1996) 351–355.
[11] T. Stüven, E.-U. Griese, H.K. Kroemer, M. Eichelbaum, U.M.
Zanger, Rapid detection of CYP2D6 null alleles by long
distance and multiplex-polymerase chain reaction, Pharmacogenetics 6 (5) (1996) 417–421.
[12] I. Rasanen, I. Ojanperä, J. Vartiovaara, E. Vuori, P. Sunila,
The advantage of dual-column approach and retention indices
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
15
combined with refined reporting in gas chromatographic drug
screening, J. High Resolut. Chromatogr. 19 (6) (1996) 313–321.
L.J. Scott, C.M. Perry, Tramadol. A review of its use in
perioperative pain, Drugs 60 (1) (2000) 139–176.
W.D. Paar, S. Poche, J. Gerloff, H.J. Dengler, Polymorphic
CYP2D6 mediates O-demethylation of the opioid analgesic
tramadol, Eur. J. Clin. Pharmacol. 53 (3-4) (1997) 235–239.
V. Subrahmanyam, A.B. Renwick, D.G. Walters, P.J. Young,
R.J. Price, A.P. Tonelli, B.G. Lake, Identification of
cytochrome P-450 isoforms responsible for cis-tramadol
metabolism in human liver microsomes, Drug Metab. Dispos.
29 (8) (2001) 1146–1155.
S. Rendic, Human P450 Metabolism Database, 2001 (http://
www.gentest.com/human_p450_database/).
M. Oscarson, M. Ingelman-Sundberg, A.K. Daly, D.W.
Nebert, Human cytochrome P450 (CYP) Allele Nomenclature
Committee, 2001 (http://www.imm.ki.se/CYPalleles/).
D. Marez, M. Legrand, N. Sabbagh, J.M. Guidice, C. Spire,
J.J. Lafitte, U.A. Meyer, F. Broly, Polymorphism of the P450
CYP2D6 gene in a European population: characterization of
48 mutations and 53 alleles, their frequencies and evolution,
Pharmacogenetics 7 (3) (1997) 193–202.
C.R. Lee, D. McTavish, E.M. Sorkin, Tramadol A preliminary review of its pharmacodynamic and pharmacokinetic
properties, and therapeutic potential in acute and chronic pain
states, Drugs 46 (2) (1993) 313–340.
G. Alvan, L. Bertilsson, M.-L. Dahl, M. Ingelman-Sundberg,
F. Sjöqvist, Moving toward genetic profiling in patient care:
the scope and rationale of pharmacogenetic/ecogenetic
investigation, Drug Metab. Dispos. 29 (4 Pt 2) (2001) 580–585.
D.W. Nebert, Polymorphisms in drug-metabolizing enzymes:
what is their clinical relevance and why do they exist? Am. J.
Hum. Genet. 60 (2) (1997) 265–271.
F.R. Sallee, C.L. DeVane, R.E. Ferrell, Fluoxetine-related
death in a child with cytochrome P-450 2D6 genetic
deficiency, J. Child Adolesc. Psychopharmacol. 10 (1)
(2000) 27–34.
H. Druid, P. Holmgren, B. Carlsson, J. Ahlner, Cytochrome
P450 2D6 (CYP2D6) genotyping on post-mortem blood as a
supplementary tool for interpretation of forensic toxicologal
results, Forensic Sci. Int. 99 (1) (1999) 25–34.
B. Bailey, R. Daneman, N. Daneman, J.M. Mayer, G. Koren,
Discrepancy between CYP2D6 phenotype and genotype
derived from post-mortem dextromethorphan blood level,
Forensic Sci. Int. 110 (1) (2000) 61–70.
A. Öhberg, E. Vuori, I. Ojanperä, J. Lönnqvist, Alcohol and
drugs in suicides, Br. J. Psychiatr. 169 (1) (1996) 75–80.
P.R. Jackson, G.T. Tucker, M.S. Lennard, H.F. Woods,
Polymorphic drug oxidation: pharmacokinetic basis and
comparison of experimental indices, Br. J. Clin. Pharmacol.
22 (5) (1986) 541–550.
ICD-10. International Statistical Classification of Diseases
and Related Health Problems, 10th Revision, WHO, Geneva,
Switzerland, 1992.