| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Department of Clinical Chemistry, University Hospital, 79106 Freiburg i. Br., Germany
a address correspondence to this author at: Department of Clinical Chemistry, University Hospital, Hugstetter Strasse 55, 79106 Freiburg i. Br., Germany
Apolipoprotein (apo) E is a Mr
34 000 plasma protein involved in lipid metabolism (1) and
is genetically polymorphic. There are three common codominant alleles,
designated 
2, 3, and
4 (2). The genetic basis of the common
alleles lies within codons 112 and 158 of the gene that codes for the
299-amino acid protein (3). At these two sites
(112/158), 2, 3, and 4
alleles contain TGC/TGC (Cys/Cys), TGC/CGC (Cys/Arg), and CGC/CGC
(Arg/Arg), respectively. The apo 3 allele is the
predominant isoform in all populations studied to date. The apo
4 allele is associated with increased total serum
cholesterol and greater risk for coronary heart disease (4);
it also constitutes a major risk factor for Alzheimer disease
(5). The apo 2 allele seems to have a
protective effect against Alzheimer disease and is associated with
longevity (6). However, most patients with familial
hyperlipoproteinemia type III are homozygous for the 2
allele (7). Thus, interest in apo E genotyping is high, both
on the basis of epidemiological research and for the purpose of
diagnosing dyslipidemia or dementia.
The three common apo E alleles lead to six common phenotypes originally disclosed by isoelectric focusing and immunoblotting (7). However, with this methodology, different degrees of protein posttranslational modification may give rise to misleading results, particularly in pathological states (8). These ambiguities can be circumvented by the use of genotyping methods such as allele-specific oligonucleotide hybridization (9), restriction enzyme analysis (ASRA) (10)(11), the amplification refractory mutation system (12), oligonucleotide ligation assay (13), heteroduplex analysis (14), or single-strand conformation polymorphism (15). Such systems generally are time-consuming and difficult to automate because they require postamplification procedures such as restriction enzyme digestion and/or electrophoresis. Of these methods, the ASRA procedure has been most widely adopted, although it has some drawbacks for large-scale DNA diagnosis, particularly related to the complex electrophoretic pattern that results from partial enzymatic digestion (16).
In this report, we describe the clinical evaluation of a commercially available DNA assay for apo E genotyping that uses rapid-cycle PCR and fluorescence resonance energy transfer (FRET) with the LightCycler (17). To allow simultaneous analysis of the two polymorphic codons in a single reaction, two reporter dyes with different excitation and emission spectra, LightCycler-Red 640 (LC-Red 640) and LC-Red 705, were used, with color compensation software to correct for the temperature-dependent crossover (crosstalk) between the emission spectra of the dyes.
In the assay, a 265-bp fragment harboring exon 4 of the apo E gene is amplified from human genomic DNA. The detection probes covering codons 112 and 158 are 5' labeled with LC-Red 640 and LC-Red 705, respectively. The corresponding anchor probes are fluorescein-labeled at their 3' ends. When a pair of hybridization probes hybridizes to the same strand internal to the unlabeled PCR primers, the probes come in close proximity, producing FRET. During FRET, the acceptor fluorophores LC-Red 640 and LC-Red 705 emit fluorescence, which is measured in exact temporal coincidence in channels 2 and 3, respectively, of the optical system of the LightCycler, using a linear arrangement of dichroic bandpass filters (17).
Because the intensity of the FRET signal depends on the amount of specific PCR product generated, this detection strategy allows monitoring of the amplification process on a per-cycle basis. More importantly, homogeneous genotyping is achieved by analysis of the melting behavior of detection probes covering the polymorphic codons. When fluorescence is monitored as the temperature increases through the melting point (Tm) of the probe/single-stranded PCR product duplex, a characteristic melting profile for each genotype is obtained, depending on the presence and type of base pair mismatch in the heteroduplex. Consequently, the presence of a particular mismatch is reflected in a specifically lower Tm for the hybrid in the melting curve graph generated by the LightCycler software.
However, when the two fluorophores are measured simultaneously, the fluorescence overlap between channels must be corrected to obtain signals that can be directly correlated with the amount of hybridized probes. This "crosstalk compensation" is achieved by use of the color compensation module integrated into the LightCycler software package.
In the current study, genomic DNA from 120 consecutive patients attending the outpatient clinic for lipid disorders at the University Hospital Freiburg was isolated from whole blood or buffy coats with the High-Pure PCR Template Preparation Kit (Roche Diagnostics, Mannheim, Germany).
All samples had been genotyped previously for the apo E polymorphism by conventional ASRA (10).
The LightCycler Apo E Mutation Detection Kit is commercially available from Roche Diagnostics. The test kit consists of three reagent vials. Vial 1 contains the mutation detection mixture with the sequence-specific amplification primers, the fluorophore-labeled hybridization probes, and an optimized concentration of Mg2+ as a 10-fold stock solution. Vial 2 contains a 10-fold concentrated ready-to-use reaction mixture containing Taq polymerase, buffer, and dNTPs. Vial 3 contains control templates heterozygous for both polymorphic codons.
After preparation of the master mixture, 18 µL of the reaction mixture and 2 µL of the isolated genomic DNA template or the control template were loaded into precooled LightCycler capillaries. For the negative control, sterile PCR-grade H2O was added instead of template. Sealed capillaries were centrifuged briefly in a microcentrifuge and then placed in the LightCycler rotor. The cycling program consisted of a 60-s initial denaturation at 94 °C, and 45 cycles of 95 °C for 0 s, 60 °C for 10 s, and 72 °C for 10 s, with a maximum ramp rate of 20 °C/s. Fluorescence was measured at the end of the annealing period of each cycle to monitor amplification. After amplification was complete, a final melting curve was recorded by cooling to 42 °C for 4 min to achieve maximum hybridization and then heating slowly at 0.1 °C/s to 80 °C. The fluorescence emitted by LC-Red 640 and LC-Red 705 was measured continuously in channels 2 and 3 during the slow temperature ramp to monitor the dissociation of the fluorophore-labeled detection probes from the complementary single-stranded DNA. The fluorescence signals recorded in the respective channels were then converted to melting peaks by plotting the negative derivative of the fluorescence with respect to temperature vs temperature (-dF/dT vs T). The peaks represent the sequence-specific melting points (Tm) at which 50% of the probes has melted off the target DNA. The entire process took ~50 min and did not require any postamplification steps.
During the amplification cycles, the fluorescence signals in both
channels increased as product accumulated. The process of hybridization
and melting of the detection probes to the target was monitored by
melting curve analysis. The detection probes for both codons matched
the alleles coding for arginine (sequence CGC). Accordingly, when we
examined codon 112 with a DNA homozygous for the sequence CGC, the
Tm was 61.5 °C, whereas a DNA
coding for cysteine (sequence TGC) produced a markedly lower
Tm of 55.5 °C. Heterozygous samples
contained both types of targets and, thus, generated both peaks (Fig. 1
A). The fluorescence signal acquired in channel 3 was used to
genotype codon 158. However, because of the interference of the
fluorescence produced by LC-Red 640, which was also recorded in channel
3, the channel 3 signals needed to be corrected for the contribution of
LC-Red 640 by the crosstalk compensation module of the LightCycler
system. With this crosstalk compensation, the alleles coding for
arginine and cysteine at codon 158 were distinguishable, with
Tms at 66 and 58 °C, respectively
(Fig. 1B
). With different samples showing different amplification
efficiencies, the derivative melting curves were reproducible, with
both fluorescence wavelengths displaying melting peaks differing by
<1.0 °C for the same allele. This allowed easy and unambiguous
assignment of genotypes to the respective melting curves. The results
of the fluorescence genotyping in 120 samples of patients were in 100%
concordance to the genotypes determined with the allele-specific
restriction enzyme analysis used as the comparison method. In
this group of patients, we found a genotype distribution typical for a
Caucasian population (Table 1
).
|
|
Routine determinations of disease-causing mutations require accurate, rapid, reliable, and low-cost methods. The potential benefits of homogeneous detection systems for the identification of mutations have long been recognized, and recently, several fluorescence-based methods for typing biallelic systems such as the factor V Leiden mutation, the prothrombin variant, and the hemochromatosis-associated HFE mutations have been described that facilitate sample processing (18)(19)(20). Aslanidis and Schmitz (21) recently reported a homogeneous assay for genotyping apo E with two pairs of hybridization probes, each of which used LC-Red 640 as acceptor dye. However, when all of the hybridization probes were used simultaneously in one PCR reaction, the insufficient discrimination between the individual Tms did not allow reliable genotyping. Therefore, the authors recommended running two separate PCR reactions with the respective hybridization probes for each sample, with negative consequences on sample throughput and costs.
We found the dual-color detection method presented here to be reliable and accurate. The set-up of the method is convenient because it uses complex stock solutions, and a major feature of the procedure is the speed with which the results are delivered. The protocol allows fluorescence genotyping of both codons in 32 samples in <50 min without any need for enzyme digestion or electrophoresis. Because this method is performed in a closed system with no postamplification processing, potential problems with sample tracking errors and end-product contamination are reduced. Furthermore, because the "hands-on" time for setting up the assay is short and no manual intervention is required after the LightCycler is loaded, genotyping with this method is cost-effective. In summary, because of its robustness, speed, and accuracy, this assay is well suited for determination of apo E genotypes in both small and large numbers of samples.
Acknowledgments
The kit was developed by Roche Molecular Diagnostics. We thank Roche Molecular Diagnostics for making the kit available for this study before the kit was launched. We also thank Ulrike Stein, Sabine von Karger, and Andreas Albers for excellent technical assistance.
Footnotes
fax 49-761-270-3444, e-mail msnauck{at}med1.ukl.uni-freiburg.de
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  G. I. Tagarakis, F. Tsolaki-Tagaraki, M. Tsolaki, A. Diegeler, N. B. Tsilimingas, and A. Papassotiropoulos The Role of Apolipoprotein E in Cognitive Decline and Delirium after Bypass Heart Operations American Journal of Alzheimer's Disease and Other Dementias, July 1, 2007; 22(3): 223 - 228. [Abstract] [PDF]
|
|
|
|
 |
|
 W. Marz, H. Scharnagl, M. M. Hoffmann, B. O. Boehm, and B. R. Winkelmann The apolipoprotein E polymorphism is associated with circulating C-reactive protein (the Ludwigshafen risk and cardiovascular health study) Eur. Heart J., December 1, 2004; 25(23): 2109 - 2119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Chen, M. A. Wollmer, F. Hoerndli, G. Munch, B. Kuhla, E. I. Rogaev, M. Tsolaki, A. Papassotiropoulos, and J. Gotz Role for glyoxalase I in Alzheimer's disease PNAS, May 18, 2004; 101(20): 7687 - 7692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J.-F. de Quervain, R. Poirier, M. A. Wollmer, L. M.E. Grimaldi, M. Tsolaki, J. R. Streffer, C. Hock, R. M. Nitsch, M. H. Mohajeri, and A. Papassotiropoulos Glucocorticoid-related genetic susceptibility for Alzheimer's disease Hum. Mol. Genet., January 1, 2004; 13(1): 47 - 52. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Maddalena, A. Papassotiropoulos, B. Muller-Tillmanns, H. H. Jung, T. Hegi, R. M. Nitsch, and C. Hock Biochemical Diagnosis of Alzheimer Disease by Measuring the Cerebrospinal Fluid Ratio of Phosphorylated tau Protein to {beta}-Amyloid Peptide42 Arch Neurol, September 1, 2003; 60(9): 1202 - 1206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. S. Bernard and C. T. Wittwer Real-Time PCR Technology for Cancer Diagnostics Clin. Chem., August 1, 2002; 48(8): 1178 - 1185. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
