| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Clinical Chemistry, Georg-August-University, Robert-Koch-Strasse 40, 37075 Göttingen, Germany.
a Author for correspondence. Fax 49-551-39-8551; e-mail eschuetz{at}med.uni-goettingen.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We used four LightCyclerTM capillaries to investigate all eight mutations. The three mutations on exon 10 were detected in one capillary with a single "shared" anchor labeled 5' with Cy5.5 and 3' with fluorescein. A wild-type-compatible 3'-fluorescein-labeled probe 5' adjacent to the anchor covered the TPMT*7 mutation, and a 5'-LC-RED640-labeled probe 3' adjacent to the anchor covered the TPMT*3C mutation. For TPMT*4, the forward amplification primer was internally labeled with a fluorescence quencher [6-carboxytetramethylrhodamine (TAMRA)], and a 3'-fluorescein-labeled antisense wild-type-compatible probe was placed at the mutation. For TPMT*2 and TPMT*3D, located on exon 5, a shared anchor approach was chosen. TPMT*3B and TPMT*6 were detected in multiplex technique and TPMT*5 in conventional manner. Anchors and probes were designed using a thermodynamic nearest-neighbor model.
Results: All mutations were detected using four capillaries with one amplification protocol in 40 min. The concentrations of the shared anchors had to be decreased to reduce their intrinsic fluorescence resonance energy transfer signals. The quenching approach using TAMRA produced a very reproducible upside-down-shaped melting curve in channel 1 of the LightCycler. Deviations from wild type were easily detected because the smallest melting point shift for any possible mutation under the core of the probes was 1.5 °C.
Conclusions: This total TPMT genotyping approach shows that it is possible to use double site-labeled anchor oligonucleotides, that channel 1 of the LightCycler can be used as detection channel for mutations using a quenching design, and that the designed probes enable detection of wild types with 100% likelihood.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TPMT is subject to a genetic polymorphism that leads to a heterozygous deficiency of this enzyme in 11% of the Caucasian population and a homozygous deficiency in 0.3% (2)(3)(4). If patients with a homozygous deficiency of TPMT are given thiopurine derivatives at a standard therapeutic oral dosage, 6-thioguanine nucleotides will accumulate, usually within 4–6 weeks, to toxic concentrations. The consequences include severe myelosuppression (5)(6), leading to life-threatening pancytopenia (7). Phenotyping of this enzyme is possible, and determination of catalytic TPMT activity can be performed in a cytosolic preparation of erythrocytes (8). However, the methods are extremely laborious and technically demanding; therefore, phenotyping is performed in only a few specialized laboratories. Genotyping of this defect is hampered by the fact that, to date, eight mutations that lead to a deficient phenotype are known (9)(10)(11)(12). These mutations have been detected mainly by PCR-restriction fragment length polymorphism techniques.
Approximately 90–95% of the phenotypic TPMT deficiencies can be
attributed to one of the known mutations, with 
75% of these known
mutations attributable to the TPMT3 subtypes A to D. The
most frequent of these is the TPMT3A variant allele, which
can be found in 55% of deficient phenotypes and which incorporates
two mutations. The first single-base mutation, G460A, located on exon
7, leads to an Ala154Thr amino acid exchange, whereas the
second mutation, A719G, which leads to a Thr240Cys amino acid exchange,
is found on exon 10. The TPMT3B allele is defined by the
presence of G460A (7%), whereas the TPMT3C allele is
defined by the A719G mutation (13%). In the context of
TPMT3A, one case has been described that showed an
additional mutation (G292T) leading to a premature stop codon, which
was assigned TPMT3D (11). However, according to
the latter, an additional problem in the genotyping of the
TPMT alleles is the possibility of a pseudo-heterozygosity
of TPMT3A/1, which cannot be discriminated from
TPMT3B/3C, the latter bearing the consequence
of complete phenotypic TPMT deficiency. Therefore, if the phenotype
(catalytic activity) is unknown, cloning of reverse
transcription-PCR products followed by re-testing for the
mutations is unavoidable in such cases for definitive diagnostic
genotyping.
The TPMT gene is located on chromosome 6p22.3 and consists
of 9 introns and 10 exons, with a cDNA of 3000 bp and an open
reading frame (ORF) of 735 bp that encodes for a 245-amino acid peptide
with a molecular mass of 35 kDa. Fig. 1
shows a schematic overview of the genomic organization of the
gene together with the known mutations that are thought to cause a
catalytic deficient phenotype. Because the existence of a processed
pseudogene on 18q21.1 with an ORF homology of 96% compared with
TPMT has been described (13), genotyping on the
basis of genomic DNA using intronic amplification primers is mandatory.
Despite the superiority of phenotyping for most clinical situations in
which a homozygous deficiency must be excluded, e.g., before an
intravenous loading dose with azathioprine, a therapy used in chronic
inflammatory diseases (14)(15), genotyping is
still relevant. Genotyping of TPMT deficiency is of scientific interest
because there are still unknown mutations accounting for 10% of
phenotypic deficiencies that cannot be explained by mutations
TPMT2 through TPMT7. Such mutations and their
frequencies need to be studied. In such cases, the presence of one of
the known genotypes must be initially excluded. In addition, patients
usually are investigated for TPMT deficiency after the occurrence of
myelosuppression during therapy with a thiopurine drug. Such patients
have often received blood transfusions because of severe anemia, and
phenotyping will no longer be reliable because the sample will contain
a combination of patient and donor erythrocytes. In such cases,
genotyping will be more reliable. We therefore aimed to develop a
strategy for rapid and reliable detection of all known single-point
mutations of the TPMT gene that are known to cause a loss of
enzyme activity. For this purpose, hybridization probe assays for the
LightCyclerTM were designed in such a way as to
minimize the costs by using multicolor, "shared" anchor, and
internal quenching approaches.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Capillaries were sealed, centrifuged in a microcentrifuge, and
placed into the LightCycler rotor.
|
mutation detection
After the amplification (45 s at 95 °C, followed by 45 cycles
of 0 s at 95 °C, 5 s at 55 °C, and 10 s at
72 °C), an analytical melting step from 40 °C to 75 °C with a
temperature transition rate of 0.1 °C/s was performed after an
initial denaturing step (30 s at 95 °C) and a hybridization step (45
s at 40 °C). During the melting step, fluorescence resonance energy
transmission (FRET) occurs from the excited fluorescein dye to the
detection dye (LC-RED640 or Cy5.5), and the emissions of the latter
dyes are recorded in dedicated channels of the instrument. The
photometer gains were set to 5 for channel 1 (fluorescein), 20 for
channel 2 (LC-RED640), and 40 for channel 3 (Cy5.5). If the stability
of the probe oligonucleotide-DNA duplex is reduced because of a
mutation, the temperature needed to break up the duplex is lower than
for the wild type. This melting phenomenon is accompanied by a
cessation of the FRET signal monitored by the LightCycler. The acquired
data are converted into melting profile curves by calculating the first
derivative of the respective fluorescence signal vs temperature for
each channel.
assay strategies
Because two mutations are described for exon 10 and a third for
the intron 9/exon 10 splice junction, which changes a nucleotide in the
conserved acceptor region downstream of the variable pyrimidine
stretch, a quenching approach was investigated to include all three
mutations in a single analysis. The amplicon covered the splice
junction mutation (TPMT4) because intronic amplification
primers were used. Therefore, 6-carboxytetramethylrhodamine (TAMRA), a
quenching dye often used for this purpose (e.g., in
TaqMan® assays), was linked into the forward
amplification primer via an amino-modifier. A probe covering the
TPMT4 mutation was synthesized in antisense orientation and
3' labeled with fluorescein. Because of the length of exon 10, the
backward primer had to be placed in the coding region. All other
amplification primers were placed at intronic sequences. Fig. 2
shows an overview of the hybridization assay strategy for exon
10. It is noteworthy that for the TPMT7 and
TPMT3C mutations, which show a 38-bp gap, only one shared
anchor oligonucleotide labeled at both ends was used. We observed a
high intrinsic fluorescence signal with this approach, which produced a
low signal-to-noise ratio during melting analyses. This problem could
be overcome by reducing the concentration of this double-labeled anchor
oligonucleotide, yielding reliable melting profiles. The same
shared anchor strategy was used for TPMT2 and
TPMT3D, which are separated by a 56-bp gap on exon 5 of the
gene. For the remaining mutations, conventional assays with one anchor
and one probe were used. TPMT3B and TPMT6 were
multiplexed in one capillary.
|
color compensation and alternative dyes
Because of the broad emission spectra of the fluorescence dyes and
the bandwidth of the photometer filters, cross-talk of fluorescence
emission into channels other than the color-specific detection channels
was observed. This was mostly eliminated by a mathematical procedure
incorporated into the LightCycler software.
The procedure and the Color-Compensation reagent set provided
by the manufacturer (Roche) were used as recommended. The compensation
for channel 3 worked equally well for LCRed705 and Cy5.5, an
alternative dye with essentially the same characteristics as LCRed705.
Alternatively, a specific color-compensation file for Cy5.5 can be
generated using an appropriate amount of Cy5.5 instead of LCRed705 for
color calibration. More detailed discussions of color compensation in
LightCycler assays have been published recently
(17)(18). The results presented in this report
may also serve as examples for the use of Cy5.5 in LightCycler assays.
As can be seen in Figs. 3B
, 4B
, 5A
, and 5B
, when coupled to
oligonucleotides by standard chemistry, this dye works well, which is
not surprising because the original LightCycler (Idaho Technology) has
often been used with similar dyes (e.g., Cy5). The advantage of such
dyes compared with those proprietary dyes recommended by Roche is to
substantially reduce assay costs.
|
|
|
amplification and hybridization oligonucleotides
A complete overview of the oligonucleotides used for all assays is
presented in Table 1
. The oligonucleotides used for PCR amplification
were essentially those published elsewhere (11). All
hybridization probes were designed using the
MeltCalc© software (19). This
software maximizes the difference between the melting point
(Tm) for the wild type and a given
mutation of interest based on thermodynamic nearest-neighbor
calculations (19)(20)(21)(22)(23)(24)(25) within an assigned temperature or
oligonucleotide length range. In addition, hybridization probe sets
that cross-anneal, form hairpins, or tend to self-anneal are
automatically excluded. Furthermore, for multiplexing assays, it is of
particular importance that none of the primers or probes used hybridize
with another oligonucleotide present in the reaction mixture. The
above-mentioned computer program is capable of calculating the risk of
such a pair for eight given oligonucleotides. Every oligonucleotide is
calculated to hybridize with the other oligonucleotides in a "walking
over" mode that uses thermodynamic data, and the highest possible
Tm for every combination is given as a
cross-tabulated result. We usually do not accept probe/anchor sets that
show a cross-hybridization Tm
>10 °C because, in our experience, it is possible to avoid this by
choosing other probes with the same discrimination characteristics.
When we used this approach, all probe sets that were synthesized worked
well and displayed the predicted good discrimination of the mutation.
An example of the beneficial effect of this approach is shown at
http://server1.medikc.med.uni-goettingen.de/Meltcalc/Example.htm.
PCR amplification primers, fluorescein-labeled oligonucleotides, and Cy5.5-labeled oligonucleotides were purchased from MWG-Biotech. LC-RED640-labeled oligonucleotides were synthesized by GenSet.
control samples
As controls for the hybridization probe assays, either DNA samples
of patients with homozygous mutations (TPMT3B and
TPMT3C) or cloned DNA from heterozygous patients
(TPMT2) were used. Control DNA sequences for the remaining
mutations were generated by site-directed mutagenesis and cloned into
TOPO-TA vector (Invitrogen) basically as described elsewhere
(26). The mutation was then verified by sequencing of the
vector insert with a Licor Model 4200 DNA-Sequencer (MWG-Biotech) using
infrared fluorescence-labeled primers. To demonstrate that the
mutations (TPMT4 to TPMT7) can also be
discriminated from the wild type (TPMT1) even in
heterozygous cases, plasmid DNA was mixed with confirmed wild-type DNA.
These mixtures were then subjected to enzymatic amplification with
consecutive analytical melting with the LightCycler as described above.
In addition, to verify the usefulness of this total genotyping approach, we genotyped 50 individuals with a phenotype [TPMT <12.5 nmol/h per mL of red blood cells (RBCs)] suspected of being attributable to a heterozygous genotype. These samples were taken from of a total of 170 routine samples analyzed during 10 consecutive weeks. This TPMT cutoff was high because these samples were from our routine screening, and therefore, we could not exclude that patients were taking a thiopurine derivative at the time of sampling. Because it is known that thiopurine therapy can increase the TPMT activity by up to 25%, we increased the cutoff from 10.0 nmol/h per mL of RBCs, which we use for individuals not on a thiopurine drug, to 12.5 nmol/h per mL of RBCs.
TPMT activity was measured in washed RBCs according to the method described by Weinshilboum et al. (8) with minor modifications (4). Briefly, the conversion of the 3H-labeled methyl group in S-adenosyl-methionine to 6-MP over 1 h at 37 °C was followed by ß-scintillation counting after extraction of the product (6-methyl-MP).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The most demanding challenge was the attempt to detect all three
mutations found on exon 10 in one assay. Fig. 3
shows the melting profiles for channel 1 (fluorescein;
TPMT4; Fig. 3A
), channel 2 (LC-RED640; TPMT7;
Fig. 3B
) and channel 3 (Cy5.5; TPMT3C; Fig. 3C
). It is
clear that, for all of these mutations, reliable discrimination between
wild type and mutation is possible. This was also true for the cases
with heterozygous mutations. Furthermore, the results for channel 1
showed that a quenching approach is also possible in the context of
conventional hybridization probe assays. The necessary 10-fold
reduction of the shared anchor oligonucleotide concentration for the
TPMT3C and TPMT7 hybridization probe assay did
not lead to detection problems, but decreased the signal-to-noise ratio
(data not shown). The appropriate concentration of such double-labeled
oligonucleotides must be determined by experimental investigations. It
can be speculated that the length of such a probe may have a major
influence on the strength of the inherent FRET signal, but other
factors such as GC content may also be of relevance. The second shared
anchor approach was used for the two mutations on exon 4:
TPMT2, for which LC-RED640 dye and channel 2 were used; and
TPMT3D, for which Cy5.5 dye and channel 3 were used. As
shown for exon 10, a clear discrimination of both mutations from their
respective wild types was possible, and heterozygous cases were clearly
identified. Typical results for each genotype are shown in Fig. 4
. This was also true for the remaining hybridization probe
assays, where TPMT5 (exon 4) and TPMT6 (exon 8)
were amplified and analyzed in one capillary in a multiplex assay (Fig. 5, A and B
). For the latter approach, the risk of
cross-hybridization of the eight oligonucleotides was calculated using
the MeltCalc program. It appears that for this combination of
oligonucleotides, no significant cross-hybridization is likely
because there were no hybridization
Tms >10 °C across the
oligonucleotides. The remaining known mutation (TPMT3B) was
genotyped in a fourth capillary in conventional manner; the resulting
melting curves are shown in Fig. 5C
. In the vicinity of the
TPMT3B mutation (G460A), a silent nucleotide transition
(T474C) occurred in 22% of a Caucasian population (11);
therefore, the probe/anchor pair was designed not to cover this
position.
The melting temperatures of all described hybridization assays are
given in Table 2
. It is apparent that there is good agreement between the
predicted (calculated) Tms for the
different hybridizations and the measured
Tms on the LightCycler. This is the
case for the perfectly matched probes, which are the wild types for all
assays described herein, as well as for the mismatched probes
(mutations). As a consequence, the predicted difference between each
wild-type Tm and mutation
Tm agreed very well with those
measured with the LightCycler. To show the usefulness of this approach,
DNA from 50 individuals suspected of being heterozygous or homozygous
deficient was used for genotyping. Of these, 19 were confirmed as
heterozygous (n = 16) or homozygous (n = 3) for
mutations that cause TPMT deficiency. The distribution of RBC TPMT
activity in the groups with and without mutations of shown in Fig. 6
. It is obvious that there is good separation of these groups
with a small overlap at 9 to 10 nmol/h per mL of RBCs. As was expected,
the allele frequency among the mutations was highest for
TPMT3A (n = 18 alleles; 14 heterozygous, 2
homozygous); this was followed by TPMT3C (n =
2), and 1 case each of TPMT2 and TPMT3B. The
latter mutation was found in a compound homozygous case who appeared to
be heterozygous for TPMT3A in genotyping but exhibited a
TPMT activity of 3.6 nmol/h per mL of RBCs. Because in our experience
heterozygous TPMT3A cases with catalytic activity <5
nmol/h per mL of RBCs are unlikely (4)(29), we
suspected a TPMT3C/3B genotype, which as
mentioned above cannot be discriminated by DNA-based methods. We
therefore cloned a reverse transcription-PCR TPMT product (1035 bp)
of this individual into a TOPO-TA vector (Invitrogen) and subjected 25
colonies to reamplification. The resulting amplicons were then mixed
with the anchor and probe oligonucleotides of TPMT3B
and TPMT3C (Table 1
), and a melting point analysis was
performed with the LightCycler as described above. The clones showed
either TPMT3C/1 or
TPMT3B/1. This result provided evidence for the
presence of a TPMT3B/TPMT3C genotype that
erroneously appears as a pseudo-heterozygous
TPMT3A/1 genotype when a regular genomic
DNA-based genotyping is performed.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, it
can be seen that every deviation from the wild type can be detected
with a Tm difference that exceeds this
limit. This can be interpreted as a diagnostic specificity of 100%. In
other words, each one of the eight known mutations can be categorically
excluded. The smallest calculated difference in
Tm under the core of the probe,
1.5 °C, which was found for TPMT3B, was still sufficient
for unambiguous detection of this deviation from wild type
(28). Considering that the differences in
Tm for the other known mutations are
4.5 °C (TPMT3C) or greater, these mutations are also
easily detectable.
The question can, however, be raised as to how definitive is the
detection of one of the known mutations solely on the basis of a
Tm lower than the
Tm for the wild type in the
LightCycler assay? To answer this question, we compared the
Tm shift caused by the mutation of
interest with the Tm shifts that would
be caused by other putative mutations under the probe. Such
calculations are possible only for the core region of the probes
because thermodynamic data are lacking for ultimate and penultimate
mutations. Taking this limitation into account, it can be seen (Table 2
) that for the common mutation TPMT3C, 35 of 69 possible
mismatches other than TPMT3C will produce a
Tm within ± 1 °C of that seen
for this mutation. For a difference of ± 1.5 °C, the number of
such possible mismatches increases to 40. Thus, the high reliability of
the mutation diagnosis depends on the much greater likelihood of the
presence of a known mutation compared with an unknown mutation under a
given probe. For this most critical probe set, the reliability is still
very high because even if all unknown mutations of the TPMT
gene (5%) are attributable to the same single-point mutation under
this probe and have a Tm difference of
<1.5 °C, the predictive value of the positive result would still be
>95%. This is a worst-case assumption. In reality, the unknown
mutations will most likely be distributed within the ORF of the gene
(735 bp) or may also be intronic (in the last 3 years, we have
experienced two cases with phenotypically defined TPMT deficiency
without a mutation in the ORF). Therefore, the predictive value of the
positive result increases to >99.9%, based on the small likelihood of
the existence of an unknown mutation under the probe.
Such critical considerations are essential when carrying out diagnostic genotyping with hybridization techniques because they enable the investigator to make a more reliable interpretation of the results. We therefore have included a feature in the MeltCalc program that calculates the Tm of all possible mismatches under the core of a given probe (19). Ultimate and penultimate mismatches are not calculated because there are no published data available. This software is freely available (http://www.meltcalc.de) for noncommercial use and should allow those working with hybridization techniques for diagnostic genotyping to gain a deeper insight into both the diagnostic power and the restrictions of these assays. Nevertheless, a final definitive result can be achieved only by use of a mutation-specific probe that shows the performance characteristics described above in addition to the wild-type probe. On the other hand, it is an advantage of this mutation detection system that mutations other than the mutation of interest can be detected, taking into consideration the above-mentioned limitations, if they are covered by the hybridization probe. This does not hold true for other mutation assays, such as restriction fragment length polymorphism assays, for mutations such as, e.g., TPMT3, where only the existence of the wild type or a specific mutation can be detected (10)(11). The recognition sites cover only 4 bp (MwoI used for TPMT3B) or 6 bp (AccI used for TPMT3C). In turn, only unknown mutations within this small proximity can be seen. The same holds true for allele-specific PCRs that can detect the wild type or a mutation but are designed to be specific for only one particular base in the gene. For other techniques, such as single-strand conformation polymorphism or denaturing gel electrophoresis, there are no published data regarding this issue. TaqMan assays, usually based on a wild-type- and a mutation-specific probe, may be less prone to the problems discussed, but from a theoretical standpoint, they cannot be excluded because small differences is Tm may not be detectable in a system that does not show the actual Tms of the probes but relies only on a fixed hybridization temperature, as shown recently (30).
Two principles of multiplexing can be realized by hybridization techniques; one is the so-called temperature multiplexing (31), and the other is color multiplexing (17)(18), which we describe here. The main advantage of color multiplexing is the possibility of designing the oligonucleotide probes solely on the basis of their discrimination capability because fluorescence cross-talk is largely eliminated by the software. In contrast, a lower specificity cannot be strictly avoided in temperature multiplexing because one probe must be designed to have a higher Tm, which ultimately leads to longer probes prone to the risk of undetectable mutations (28). In addition, the risk of "temperature cross-talk" must be considered, and a method for HFE genotyping that was published with several slight modifications may serve as an example for the limitations of a temperature multiplexing strategy (31)(32)(33). In the vicinity of the HFE-H63D mutation is another known mutation (HFE-S65C), which is found under the same probe. The presence of this second mutation will cause two mismatches with the HFE-H63D-compatible probe and, consequently, stronger destabilization of the strands (31)(33). The Tm of this mutation falls within the range of the multiplex probe with the lower Tm (31)(33) and is therefore undetectable, a fact that was already mentioned in the report by Bernard et al. (31).
On the basis of these critical considerations, we cannot completely share the enthusiastic conclusion of a recently published editorial (34). Even if it is theoretically possible to gain up to 24 results within one assay by a combination of temperature and color multiplexing, one has to consider the limitations of temperature multiplexing described above. When mutation-specific probes are used, "Tm cross-talk" may occur, whereas for wild-type-compatible probes, mutations may be present that are not detectable. Color multiplexing using thermodynamically designed probes seems to give more reliable results. If temperature multiplexing is used, thermodynamic calculations such as those shown here are helpful for designing probes that are specific and for understanding both the advantages and the limitations of existing assays.
final considerations for tpmt genotyping
Phenotyping of TPMT is still the gold standard, but it is
restricted to specialized laboratories. On the other hand, the
demand for TPMT determination has been growing during the last 5 years
as more clinicians have become aware of the pharmacogenetic
relationship of TPMT with thiopurine drugs and the consequences that
may occur if a deficiency is overlooked. Genotyping can be set up
easily in a laboratory that has the necessary technical prerequisites.
With the color multiplex hybridization assays described here, it is
possible to identify or exclude all seven known TPMT
mutations within a single LightCycler run, using four capillaries, with
a run time <1 h. The use of a quenching approach with dyes such as
TAMRA is also applicable to the LightCycler, as we showed for the
TPMT4 mutation. The resulting upside-down-shaped melting
curve (Fig. 3A
) in channel 1 (fluorescein) shows good separation of the
genotypes. The software does not automatically recognize the
Tm, which must be estimated manually.
The distance between TAMRA and fluorescein is 20 bp, which is within
the usual range used for such approaches. Recently, it was shown that a
shared anchor double end-labeled with fluorescein can be used for
genotyping of mutations in close proximity in one gene (35).
We demonstrate that it is also possible to use a pair of classical FRET
dyes, in our case, Cy5.5 and fluorescein, to label both ends of an
oligonucleotide. This was possible because anchors could be constructed
between the detection probes (distance between probes was 25
bp). Because of an intrinsic FRET, the anchor concentration
had to be decreased. Thus, high concentrations of anchors are not
necessary for a successful detection.
Double labeling of anchors with the same dyes (35) as well
as with different dyes enables a more flexible probe design for "hot
spot" regions. We could show in our data set that the lowest TPMT
activity for the homozygous wild type (TPMT1) was 9.0
nmol/h per mL of RBCs and the highest for a heterozygous genotype was
10.1 nmol/h per mL of RBCs. In this collective, the frequency of
heterozygous genotypes was 9.5%, which is similar to the value
(11%) that has been reported for Caucasian populations
(3)(11). However, the homozygous cases were
overrepresented (1.8%), probably because these were cases with
clinically manifest leukopenia after thiopurine challenge that were
sent to our laboratory for diagnosis. It is particularly noteworthy
that one case was found to be heterozygous for the defective alleles
TPMT3B and TPMT3C. This constellation cannot be
distinguished from TPMT3A/1 in routine
genotyping. As a consequence, such a patient has a high risk of
developing toxicity but would be falsely categorized as having moderate
or low risk. The existence of such genotypes, which has been proposed
but never demonstrated in humans, shows that phenotyping of TPMT still
has its place in the clinical laboratory despite the much greater
simplicity of genotyping assays.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1-antitrypsin genotyping by multiplex real-time fluorescence PCR with the LightCycler. Clin Chem 2000;46:156-162.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  N. von Ahsen and M. Oellerich Chip-Based Genotyping: Translation of Pharmacogenetic Research to Clinical Practice Clin. Chem., July 1, 2007; 53(7): 1186 - 1187. [Full Text] [PDF]
|
|
|
|
|
|
M. Reinshagen, E. Schutz, V. W. Armstrong, C. Behrens, C. von Tirpitz, A. Stallmach, H. Herfarth, J. Stein, P. Bias, G. Adler, et al. 6-Thioguanine Nucleotide-Adapted Azathioprine Therapy Does Not Lead to Higher Remission Rates Than Standard Therapy in Chronic Active Crohn Disease: Results from a Randomized, Controlled, Open Trial Clin. Chem., July 1, 2007; 53(7): 1306 - 1314. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Schutz, M. Scharfenstein, and B. Brenig Genotyping of Ovine Prion Protein Gene (PRNP) Variants by PCR with Melting Curve Analysis, Clin. Chem., July 1, 2006; 52(7): 1426 - 1429. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Moser, D. R. Christensen, D. Norwood, and J. R. Prudent Multiplexed Detection of Anthrax-Related Toxin Genes J. Mol. Diagn., February 1, 2006; 8(1): 89 - 96. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E M Frohman, E Havrdova, B Levinson, and O Slanar Azathioprine myelosuppression in multiple sclerosis: characterizing thiopurine methyltransferase polymorphisms Multiple Sclerosis, February 1, 2006; 12(1): 108 - 111. [Abstract] [PDF]
|
|
|
|
|
|
N. von Ahsen, V. W. Armstrong, C. Behrens, C. von Tirpitz, A. Stallmach, H. Herfarth, J. Stein, P. Bias, G. Adler, M. Shipkova, et al. Association of Inosine Triphosphatase 94C>A and Thiopurine S-Methyltransferase Deficiency with Adverse Events and Study Drop-Outs under Azathioprine Therapy in a Prospective Crohn Disease Study Clin. Chem., December 1, 2005; 51(12): 2282 - 2288. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Graham, M. Liew, C. Meadows, E. Lyon, and C. T. Wittwer Distinguishing Different DNA Heterozygotes by High-Resolution Melting Clin. Chem., July 1, 2005; 51(7): 1295 - 1298. [Full Text] [PDF]
|
|
|
|
 |
|
 K. Takatsu, T. Yokomaku, S. Kurata, and T. Kanagawa A FRET-based analysis of SNPs without fluorescent probes Nucleic Acids Res., November 8, 2004; 32(19): e156 - e156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. von Ahsen, V. W. Armstrong, and M. Oellerich Rapid, Long-Range Molecular Haplotyping of Thiopurine S-Methyltransferase (TPMT*) *3A, *3B, and *3C Clin. Chem., September 1, 2004; 50(9): 1528 - 1534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Takatsu, T. Yokomaku, S. Kurata, and T. Kanagawa A new approach to SNP genotyping with fluorescently labeled mononucleotides Nucleic Acids Res., April 15, 2004; 32(7): e60 - e60. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Haglund, M. Lindqvist, S. Almer, C. Peterson, and J. Taipalensuu Pyrosequencing of TPMT Alleles in a General Swedish Population and in Patients with Inflammatory Bowel Disease Clin. Chem., February 1, 2004; 50(2): 288 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Vrettou, J. Traeger-Synodinos, M. Tzetis, G. Malamis, and E. Kanavakis Rapid Screening of Multiple {beta}-Globin Gene Mutations by Real-Time PCR on the LightCycler: Application to Carrier Screening and Prenatal Diagnosis of Thalassemia Syndromes Clin. Chem., May 1, 2003; 49(5): 769 - 776. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. M. Mackay, K. E. Arden, and A. Nitsche Real-time PCR in virology Nucleic Acids Res., March 15, 2002; 30(6): 1292 - 1305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Burggraf, S. Kosel, S. Lohmann, R. Beck, and B. Olgemoller Unexplained DNA Melting Behavior in a Genotyping Assay Clin. Chem., January 1, 2002; 48(1): 199 - 201. [Full Text] [PDF]
|
|
|
|
|
|
N. von Ahsen, C. T. Wittwer, and E. Schutz Oligonucleotide Melting Temperatures under PCR Conditions: Nearest-Neighbor Corrections for Mg2+, Deoxynucleotide Triphosphate, and Dimethyl Sulfoxide Concentrations with Comparison to Alternative Empirical Formulas Clin. Chem., November 1, 2001; 47(11): 1956 - 1961. [Abstract] [Full Text] [PDF] |
