| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Clinical Chemistry, Georg-August-University, Robert-Koch-Strasse 40, 37075 Goettingen, Germany.
a Author for correspondence. Fax 49-551-39-8551; e-mail nahsen{at}gwdg.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
Methods: Hybridization probes were designed complementary to the wild type (TA)6 and to alleles with (TA)7 and (TA)8 repeats in the promoter region. Melting points were measured in samples representing all currently known alleles with (TA)5 to (TA)8 repeats. Probe melting points were predicted with a thermodynamic nearest-neighbor model for Watson-Crick paired probes. The dominant secondary structures resulting from probe hybridization were predicted by thermodynamic free energy calculations. Alternatively samples were genotyped based on amplicon size resolved by high-resolution polyacrylamide gel electrophoresis.
Results: Only short probes (22–24 bases) could be successfully used for genotyping this locus because of the very low stability of this TA repeat. Assays based on (TA)7 or (TA)8 genotype-compatible hybridization probes effectively discriminated five to eight TA repeats. The consecutive use of two different detection probes was necessary for better discrimination of some heterozygous genotypes. All results were in concordance with the alternative genotyping method. Of 100 investigated Caucasians (50 males, 50 females), 9 (9%) were homozygous for the (TA)7 allele.
Conclusions: The presented method for genotyping the (TA)n promoter polymorphism of the UGT1A1 gene with the LightCycler has the potential to genotype all currently known (TA)n repeats in a single assay and is sensitive toward possible new genotypes. Our findings also show that thermodynamic calculations are of practical value for the design of hybridization probe assays for the genotyping of insertion/deletion polymorphisms.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
Methods for genotyping this insertion polymorphism are laborious and time-consuming (3)(4)(6). Recently, two methods were described that allow for higher sample throughput. One method, based on denaturing HPLC (7), cannot discriminate different homozygous genotypes in one analysis. The other uses the melting temperature (Tm) of the PCR product as an indirect estimate of strand length and thus of the (TA)n promoter genotype (8).
In general, GC as well as TA tandem repeats in DNA are intrinsically unstable because of hairpin formation, replication slippage, and probably also unequal sister chromatid exchange. GC and TA tandem repeats are involved in the pathogenesis of various diseases, and the understanding of their molecular turnover is increasing (9). Therefore, the easy and reliable detection of such nucleotide repeat polymorphisms is of importance for present genotyping methods. The suitability of the LightCyclerTM (Roche Biochemica) for hybridization probe genotyping to detect insertions or deletions has been demonstrated (10)(11)(12). In these situations and in contrast to the detection of single or multiple mismatches, other secondary structures such as bulges, loops, or unmatched ends often result from mismatched probe hybridization. Unpaired bases, so-called bulges, are formed in nucleic acid helices when one or more consecutive bases do not pair to bases on the other strand. If two or more bases remain unpaired, the structure is called a bulge loop. These secondary structures influence DNA structure and stability (13) in ways different from that of single mismatches.
Because of its high TA content, the UGT1A1 locus requires an
optimized probe design to achieve safe discrimination of genotypes in
this DNA region of low stability. We have shown that the
Tm observed from genotyping of single
nucleotide polymorphisms can be predicted by a thermodynamic
nearest-neighbor model (14). The thermodynamic parameters
(enthalpy and entropy) that describe the destabilization of a
nearest-neighbor DNA doublet by bulge or loop formation have not been
published, whereas the free energies (
G0) for
most of the different secondary structures are known. This is useful
for prediction of stability trends resulting from mismatched probe
hybridization and for assessment of the resulting secondary structures.
We present a PCR method for genotyping the UGT1A1 (TA)n polymorphism using hybridization probes on the LightCycler. Patterns resulting from probe hybridization on insertion/deletion polymorphisms are discussed based on thermodynamic considerations. The method was compared with the results from genotyping by PCR amplicon size resolution using polyacrylamide gel electrophoresis (PAGE).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
. The choice of a (TA)7 probe should allow
detection of all known genotypes, i.e., five to eight TA repeats. To
avoid probe slipping, a tight hybridization was enforced by the GCC
sequence at the 5' terminus (Fig. 1a
). For discrimination between
(TA)7 and (TA)8 genotypes,
the probe contained a 3' G. The anchor probe, UGT1A1-anchor (6 bp
downstream; 5'-Cy5.5-AGG GCG AAC CTC TGG CAG GA-3'-PHO) was 5'-labeled
with the Cy5.5 dye (Amersham Pharmacia Biotech), which exhibits an
emission maximum at 694 nm, comparable to the LC-Red705 dye (Roche
Biochemicals). The 3' end of the UGT1A1-anchor was phosphorylated.
Alternatively, the following probes were used as detection probes
together with the UGT1A1-anchor probe:
UGT1A1(TA)6 probe (5'-TGC CAT ATA TAT ATA TAT AAG
T-3'fluorescein); or UGT1A1(TA)8 probe (5'-GCC
ATA TAT ATA TAT ATA TAT AAG-3'fluorescein). All primers and probes were
from MWG-Biotech.
|
genomic dna extraction
Genomic DNA was extracted as described previously (15).
The protocol yielded a genomic DNA concentration of 
0.75 ng/µL; 1
µL of this solution was used in the PCR assay.
site-directed mutagenesis
The (TA)5 and (TA)8
alleles have not been detected in the samples investigated to date.
They were therefore constructed by site-directed mutagenesis using a
megaprimer approach. A sample with the (TA)6
genotype was cloned into a TOPO TA vector (Invitrogen). Plasmid DNA was
then amplified with a proofreading enzyme mixture
(ExpandTM High Fidelity PCR system; Roche
Biochemica) using mutagenic primers that introduced the
(TA)5 or (TA)8 repeats into
the sequence. The products were cloned into a TOPO TA vector, and the
expected mutation was confirmed by sequencing (sequenase cycle
sequencing method; Amersham) on an automated DNA sequencer
(Licor 4200; Licor). Melting curves of heterozygous samples with the
(TA)5 and (TA)8 alleles
were acquired after PCR amplification of homozygous genomic DNA to
which diluted plasmid DNA had been added.
pcr protocol
PCR reactions were carried out in a final volume of 10 µL in
LightCycler glass capillaries. The reaction mixture consisted of 1 µL
of DNA solution, 1 U of Taq DNA polymerase (Roche Biochemica), 1 µL
of 10x PCR buffer (Roche Biochemica), 0.2 mmol/L each dATP, dCTP,
dGTP, and dTTP (Roche Biochemica), 2.5 mmol/L
MgCl2, 500 mg/L bovine serum albumin (New England
BioLabs), and 50 mL/L dimethyl sulfoxide (Sigma). Amplification primers
and hybridization probes concentrations were as follows: 0.5 µmol/L
UGT1A1-for, 0.5 µmol/L UGT1A1-rev, 0.1 µmol/L
UGT1A1(TA)7 probe, and 0.3 µmol/L
UGT1A1-anchor. PCR-grade water was added to 10 µL. The fluorometer
gain setting was 30 in channel 3. The cycling program consisted of
30 s of initial denaturation at 95 °C and 45 cycles of 95 °C
for 0 s, 50 °C for 5 s, and 72 °C for 5 s, with
the maximum ramp rate. The analytical melting program was 95 °C for
30 s and 29 °C for 30 s, increasing to 55 °C at a ramp
rate of 0.1 °C/s, with continuous fluorescence acquisition. Use of
the minimal possible ramp rate minimized the difference between the
experimental Tm and the true
equilibrium Tm as calculated with the
nearest-neighbor model. Amplification and detection were complete
within 30 min.
Each assay included a DNA control with the (TA)6/(TA)7 genotype and a contamination control. Genotypes of control DNA were confirmed by cycle sequencing of samples initially genotyped with the LightCycler method. In all cases, sequencing confirmed the genotype expected from the hybridization probe assay.
genotyping of the UGT1A1 (TA)n
polymorphism by high-resolution page
Genomic DNA was amplified for 40 PCR cycles using the same
protocol as described above but without addition of detection probe and
anchor. After PCR, the dNTPs (0.02 mmol/L each), Taq DNA polymerase
(0.1 U), and the UGT1A1-rev primer (0.05 µmol/L) 5' labeled with
IRD-800 dye were added to each tube. The samples were centrifuged and
amplified for an additional 10 PCR cycles. The PCR product was mixed
with 3 µL of gel-loading solution (Sigma), and 1 µL was loaded on a
sequencing gel (Licor 4200) together with size calibrators
comprising five to eight TA repeats. Genotypes were assigned to samples
by comparison to size calibrators.
thermodynamic calculations
Thermodynamic predictions for Watson-Crick paired
oligonucleotides were calculated using the MeltCalc software add-in for
Excel. This software is free for noncommercial use and can be
downloaded from http://www. meltcalc.de
(14)(16). Alternatively, the HyTher software
(http://jsl1.chem.wayne.edu) (17)(18)(19) may be used. For our
PCR, we assumed 220 mmol/L Na+ equivalents and
0.1 µmol/L hybridizing strand concentration for duplex stability
calculations. G037 for
secondary structure stability assessment was calculated under standard
conditions (1 mol/L NaCl). See the Appendix for additional
details.
Several DNA motifs frequently occur when probes hybridize to targets. These include so-called dangling ends, which occur on both ends of a completely hybridized probe; penultimate and terminal mismatches may also occur. Recently, the thermodynamic parameters for DNA sequences with dangling ends were investigated (20). Accordingly, certain dangling ends can contribute as much to duplex stability as an additional Watson-Crick base pair. The MeltCalc 2.0 software and the HyTher program both calculate double strand stability, including dangling end effects. Currently, only the HyTher program can account for penultimate and terminal mismatches, but it uses as yet unpublished nearest-neighbor parameters for the calculation of these situations. Thermodynamic calculations were carried out using both software programs. Calculations with MeltCalc included a 5' dangling end and no contribution of the 3' unmatched end. Calculations with HyTher included a 5' dangling end and a 3' terminal/penultimate mismatch. For completely Watson-Crick paired probes, 5' and 3' dangling ends were considered.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
secondary structure considerations
A homozygous (TA)7 genotype with no mismatch
under the UGT1A1(TA)7 probe displayed a single
melting peak at 46.2 °C (Fig. 1a
and Table 1
). G037 was calculated
as described in the Appendix. Hybridization with the
wild-type DNA leads to formation of a 4-bp unmatched end
(G037 = -14.6 kcal/mol; Fig. 1b
-I) or of a 2-bp bulge loop
(G037 = -15.05 kcal/mol;
Fig. 1b
-II), and a lower Tm of
39.9 °C. A (TA)8 genotype produces a 2-bp
unmatched end (G037 = -16.0
kcal/mol; Fig. 1c
-I) or a 2-bp bulge loop
(G037 = -16.5 kcal/mol; Fig. 1c
-II), and a Tm of 42.2 °C.
Hybridization with a (TA)5 genotype leads to a
tandem and a single mismatch where
G037 cannot be calculated
because of missing thermodynamic data. When we considered only the
16-bp Watson-Crick paired duplex without hydrogen bonding in the 6-bp
unmatched end, the calculated
G037 was -13.1 kcal/mol
(Fig. 1d
), which is within the prediction error of the model for the
alternative formation of a 4-bp bulge loop
(G037 = -12.41 kcal/mol).
There was good agreement between observed melting points and those
predicted with the thermodynamic nearest-neighbor model (Table 1
). The
consideration of terminal mismatches did not improve the prediction
compared with the observed Tms using
the LightCycler in this specific situation. The mean prediction error
for the 12 cases (3 probes and 4 alleles) was 0.4 ± 1.12 °C
(mean ± SD) without and 1.5 ± 1.20 °C with terminal
mismatch calculations. For the reasons discussed above,
Tm calculations are not possible at
present for probe hybridizations that are destabilized by bulge or loop
formation.
|
genotype assessment
The Tms for the different probes
are summarized in Table 1
. The CV was <1.5% for all run-to-run
data shown in Table 1
. Examples for the genotyping of homozygous
samples are shown in Fig. 2
. The Tm difference
between matched and mismatched probes must be sufficient to allow
discrimination of heterozygous mutations by two melting peaks. Initial
genotyping should be performed with the
UGT1A1(TA)7 probe. This resolves samples with the
common (TA)6/(TA)7 genotype
and several other rare heterozygous genotypes (Fig. 3
) as well as those homozygous for (TA)6 or
(TA)7 alleles. Some genotypes are not well
resolved with the UGT1A1(TA)7 probe, e.g., the
(TA)7/(TA)8 genotype (Fig. 3
). If the Tm difference is too small,
a single broad melting peak is produced, as illustrated by the
(TA)6/(TA)8 genotype (Fig. 3
). However, resulting melting curves are clearly differentiated from
samples with other genotypes by their shapes and
Tms. Such samples must be reanalyzed
using the UGT1A1(TA)8 probe. The
(TA)6 probe is useful only for confirmation of a
(TA)6 genotype, but it is not needed for clinical
genotyping. A general strategy for the unambiguous genotyping of all
known (TA)n repeat genotypes using the
UGT1A1(TA)7 and the
UGT1A1(TA)8 probe is outlined in Fig. 4
. The resolution of melting curves improves with a lower ramp
rate for melting curve acquisition and with a smaller number of samples
processed in the assay. This should be considered when insufficient
resolution of heterozygotes is suspected.
|
|
|
One hundred DNA samples (50 male and 50 female) were anonymously provided by our blood bank from local blood donors. Genotyping was performed using the UGT1A1(TA)7 probe in the LightCycler assay. Allelic frequencies were comparable to those reported for a Caucasian population (3). The homozygous (TA)7 genotype that confers a susceptibility to Gilbert syndrome was found in 6% of males and 12% of females, whereas 54% of males and 34% of females carried the homozygous wild-type allele.
method comparison with high-resolution page
Enough DNA material was left from 71 of these samples so that they
could also be genotyped by high-resolution PAGE. Results were 100%
concordant to the hybridization probe assay.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
, b–d). Tandem mismatch refers to two immediately
adjacent opposing unpaired nucleotides in each strand (an internal
2 x 2 loop). Multiple mismatches refers to several single
mismatches (an internal 1 x 1 loop) interrupted by Watson-Crick
paired nucleotides. The base pairing of a probe hybridizing over an
insertion/deletion polymorphism will be such that a minimum free energy
results. Therefore, in contrast to what is shown in Fig. 1
of the
report by Nauck et al. (10), a single-base bulge will form
rather than a loop because the binding of four G bases is energetically
much more favorable than loop formation. TA repeats have a low
stability, and the destabilizing effect of the polymorphism on the
probe is weak. We were unable to achieve sufficient
Tm discrimination with longer
detection probes (>32 bp, Tm
>55 °C) at this locus. In our method, the shortest probe possible
was chosen, which just covered the investigated region. Thus, probe
hybridization is maximally destabilized by the insertion/deletion
polymorphism.
In accordance with previous observations (14), we found good
agreement between observed and predicted melting temperatures (Table 1
). The Tm calculated on the
assumption that an unmatched end is not significantly contributing to
the probe stability is a good estimate of the experimental melting
temperature (Table 1
). In the case of a (TA)5
genotype hybridized with the (TA)7 probe, base
pairing may be possible in the 6-bp unmatched end (Fig. 1d
-I); however,
this can produce unfavorable steric interactions of internal mismatches
(22). Similar considerations are possible for the
(TA)6 and (TA)8 probe.
The genotyping of samples with five to eight TA repeats is
possible with a single properly designed probe. Some rare allelic
combinations require the use of a second probe, as outlined in Fig. 4
.
In Caucasian populations, the (TA)6 and
(TA)7 alleles are most prevalent; consequently,
the (TA)7 probe is the best choice for genotyping
because (TA)6 and (TA)7
alleles are well resolved. Repeated elements in the genome are unstable
and prone to insertions and deletions because of unequal crossing over
in meiosis (3)(9). It is possible that
insertions/deletions other than those reported to date will be found
once larger samples from different ethnic groups are screened with
sensitive methods. LightCycler genotyping has the potential of
detecting previously unknown or unexpected base substitutions
(11)(14)(23). Hybridization
probe-based methods could identify the presence of putative new alleles
such as (TA)4 or (TA)9. We
anticipate that nonspecific methods using an indirect measure of the
(TA)n repeat number by PCR product
Tm (8) will give wrong
genotyping results in the presence of certain allelic combinations.
This is a point of concern because the presence of a heterozygous
(TA)6/(TA)8 genotype was
recently reported in an Italian patient with Gilbert syndrome
(24). For example, a
(TA)6/(TA)8 heterozygous
individual might be misclassified as homozygous
(TA)7 genotype by these indirect methods.
In conclusion, the presented method has the potential to genotype all currently known (TA)n repeats in a single assay and is sensitive toward possible new genotypes. Our findings also show that thermodynamic calculations are of practical value for the design of hybridization probe assays for the genotyping of insertion/deletion polymorphisms.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
G037(i) is
the free energy at 37 °C for the 10 possible nearest neighbors and
ni is the number of occurrences of each nearest neighbor
(19):
 |
Because DNA base bulges in a duplex have properties similar to RNA
molecules (13), we used the known free energy penalties for
the latter: a G037 bulge of
+3.8 kcal/mol for a 1-bp bulge, +2.8 kcal/mol for a 2-bp bulge, +3.2
kcal/mol for a 3-bp bulge, and +3.6 kcal/mol for a 4-bp bulge
(25).
The penalty for bulge formation adds to the nearest-neighbor stacking pair free energy. The more negative the stacking free energy of a nearest-neighbor pair, the less likely it is that a base bulge will be inserted at that site (13). The formation with the lowest free energy is considered the dominant secondary structure. Resulting free energies for different secondary structures within ± 0.5 kcal/mol of each other are within the experimental error. In these situations, the dominant secondary structure cannot be predicted.
example
Hybridization of the 22mer (TA)7 probe on
the (TA)6 wild-type strand can lead to different
secondary structures (Fig. 1
, b-I and b-II). The Watson-Crick pairing
of the first 18 bp (Fig. 1b
-I) produces a
G037 total of -14.6
kcal/mol. Alternatively (Fig. 1b
-II), a 2-bp bulge loop can form
between a TA/AT nearest-neighbor pair. The resulting 20-bp Watson-Crick
paired duplex has a G037
total of -17.85 kcal/mol, to which the loop penalty of
+2.8 kcal/mol must be added. The G037
total of -15.05 kcal/mol for this secondary structure is
not significantly different from the G037
total assuming unmatched end formation. Therefore, for this
oligonucleotide the dominant secondary structure cannot be predicted.
Similar considerations led us to the postulated secondary structures
depicted in Fig. 1c
(unmatched end vs 2-bp bulge loop formation) and
Fig. 1d
(unmatched end vs 4-bp bulge loop formation).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
1-antitrypsin genotyping by multiplex real time fluorescence PCR with the LightCycler. Clin Chem 2000;46:156-161.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  G. Pont-Kingdon, R. L. Margraf, K. Sumner, A. Millson, E. Lyon, and E. Schutz Design and Application of Noncontinuously Binding Probes Used for Haplotyping and Genotyping Clin. Chem., June 1, 2008; 54(6): 990 - 999. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Diaz, T. Boekhout, B. Theelen, M. Bovers, F. J. Cabanes, and J. W. Fell Microcoding and flow cytometry as a high-throughput fungal identification system for Malassezia species. J. Med. Microbiol., September 1, 2006; 55(Pt 9): 1197 - 1209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. von Ahsen Two for Typing: Homogeneous Combined Single-Nucleotide Polymorphism Scanning and Genotyping Clin. Chem., October 1, 2005; 51(10): 1761 - 1762. [Full Text] [PDF]
|
|
|
|
 |
|
 G. Pont-Kingdon and E. Lyon Direct molecular haplotyping by melting curve analysis of hybridization probes: beta 2-adrenergic receptor haplotypes as an example Nucleic Acids Res., June 3, 2005; 33(10): e89 - e89. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Jiang, T. Ellis, and A. R. Greenlee Genotyping Parkinson Disease-Associated Mitochondrial Polymorphisms Clin. Med. Res., May 1, 2004; 2(2): 99 - 106. [Abstract] [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]
|
|
|
|
|
|
N. von Ahsen, M. Oellerich, and E. Schutz Limitations of Genotyping Based on Amplicon Melting Temperature Clin. Chem., July 1, 2001; 47(7): 1331 - 1332. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
