| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Center for Molecular Medicine and Genetics and
2
Department of Surgery, Wayne State University School of Medicine, Detroit, MI 48201-1928
aaddress correspondence to this author at: Center for Molecular Medicine and Genetics, 3116 Gordon H. Scott Hall of Basic Medical Sciences, 540 East Canfield Ave., Detroit, MI 48201-1928; fax 313-577-5218, e-mail tromp{at}sanger.med.wayne.edu
A common constraint faced by genetic studies is the limited amount of DNA available from study participants. It is often not practical or possible to either collect a large amount of tissue such as liver or blood from individuals or to ask for a second sample, particularly if the critical individuals died since the first sample was taken. Immortalization of white blood cells by use of Epstein–Barr virus in theory provides an unlimited source of material. The immortalization protocol itself, however, must be carried out within 48 h of blood collection and is time-consuming.
DNA linkage studies in which microsatellite or other polymorphic DNA markers are used for a genome scan require appreciable amounts of DNA. As multiple PCRs are carried out with each sample, valuable genomic DNA is easily depleted. To prevent loss of irreplaceable samples and to maximize the available genetic materials for DNA linkage and genetic association studies as well as for future studies of screening for mutations in candidate genes, we used the primer-extension preamplification (PEP) protocol (1) to carry out whole-genome amplification.
The advantages of using the PEP protocol are that only a small amount of genomic DNA is needed for the entire genome scan and the remaining genomic DNA can be stored for future analyses. One possible caveat is that PEP may not amplify both parental copies of an individual’s DNA or may not amplify them equally at all loci. In turn, genotyping inaccuracies may potentially obscure allele sharing. To investigate the reliability of genotyping results when PEP was used to produce PCR templates, we compared the results for PCR products from PEP with those obtained with genomic DNA templates. All experiments were carried out according to protocols approved by the Institutional Review Board of Wayne State University School of Medicine.
Before PEP, 30–250 ng of the genomic DNA was denatured at 95 °C for 5 min and cooled on ice. The PEP was 45 cycles of denaturation at 92 °C for 1 min, annealing at 37 °C for 2 min (programmed increase in time of 4 s/cycle and increase in temperature of 0.4 °C/cycle so that in the final, 45th, cycle the annealing was carried out at 55 °C), and a polymerase extension step of 4 min at 55 °C (PEC9600; PE Biosystems). The PEP products were diluted 100-fold in two steps. In the first step, the 50-µL PCR was added to 750 µL of 10 mmol/L Tris–1 mmol/L EDTA (pH 8.0). In the second step, 16 µL of the first dilution was added to 84 µL of 10 mmol/L Tris–1 mmol/L EDTA (pH 8.0). The second diluation was used for genotyping.
For genotyping, we used previously published primer sets and PCR conditions for the matrix metalloproteinase-1 (MMP1) (2), matrix metalloproteinase-3 (MMP3) (3), and plasminogen activator inhibitor-1 (PAI-1) (3) genes. A primer set was designed for MMP13 (GenBank accession no. X75308) PCR (sense, 5'-GATACGTTCTTACAGAAGGC-3'; antisense, 5'-ACCCATCTGGCAAAATAAAC-3') and purchased from Integrated DNA Technologies. Primers for dinucleotide repeat markers (MapPairs) were purchased from Research Genetics. One of the genotyping primers was radioactively labeled using T4 polynucleotide kinase (New England BioLabs) and [
-32P]ATP (Amersham). PCRs were carried out in 15-µL reaction volumes that included 5 µL of 100-fold diluted template from PEP (1.7–7 ng of DNA based on absorbance reading; see below) or 30–100 ng of genomic DNA as template, 5 pmol of unlabeled primer, 5 pmol of radiolabeled primer, 200 µM deoxynucleotide triphosphates (dNTPs; final concentration), 0.3 U of Taq DNA polymerase [AmpliTaq Gold (PE Biosystems) for MMP1, MMP13, PAI-1, and all microsatellite markers, and Taq DNA polymerase (Qiagen) for MMP3]. Several reactions containing all the reagents except template DNA, which was replaced by water, were carried out as negative controls in the PCRs along with the experimental samples, and they were consistently negative. PCR products were analyzed on 7% polyacrylamide gels containing formamide or 6% sequencing gels (Sequagel-6; National Diagnostics) to resolve the 1- to 4-base differences between different alleles. After fixing and drying, the gels were exposed to x-ray film (Eastman Kodak Company) at -80 °C for autoradiography.
The reliability of the PEP protocol (1) was tested by (a) sequencing the PCR products (ThermoSequenase Cycle Sequencing Kit; USB) generated with and without PEP from 16 individuals for a 306-bp PCR product from the MMP-8 promoter and (b) by comparing the genotyping results obtained from PEP DNA templates with those obtained with genomic DNA templates for a total of 881 genotypes from five different polymorphic markers typed with 332 different templates (Table 1
). DNA sequencing did not reveal any differences in the PCR products with or without PEP (not shown). Furthermore, there was 100% concordance between the genotypes from PEP DNA and genotypes from the corresponding genomic DNA samples where both produced a result (Table 1
). The PEP-amplified DNA successfully yielded PCR products for polymorphic regions scattered throughout the chromosomes. Additionally, 6000 genotypes generated with 54 microsatellite markers and PEP DNA templates yielded no inconsistent inheritance in pedigrees (not shown). For highly variable microsatellite markers, the number of heterozygotes observed was consistent with expectation. Thus, the PEP protocol did not cause selective amplification or loss of one allele as determined by concordance, heterozygosity, and consistent inheritance in pedigrees (loss of one allele could also yield inconsistent inheritance in pedigree). Not only did the PEP protocol yield high-fidelity copies, the PEP DNA appeared to be a more suitable template for the PCR because a greater proportion of the PCRs were successful (Table 1
). In addition, the intensity of genotyping products was more uniform from sample to sample when PEP DNA templates were used than when genomic DNA templates were used, making interpretation of results easier.
|
The amount of DNA generated in the PEP was determined after the PEP products were purified from the remaining primers and nucleotides (Microcon YM-30 Centifugal Filter Devices; Millipore Corporation) and quantified by spectrophotometry at 260 nm. The PEP produced up to a 110-fold increase in the amount of DNA (Fig. 1
), a nearly linear increase in DNA. In our hands, the PEP protocol yielded sufficient amounts of product to perform >400 genotyping PCRs when as little as 30 ng of was used as the genomic DNA template. Typically, we carried out the PEP with 100 ng of genomic DNA and aliquoted the PEP for future use for thousands of genotyping reactions. In negative-control experiments, genomic DNA that was processed through the PEP procedure without the enzyme, primer, and dNTPs yielded PCR products only for exceedingly robust PCRs because 30 ng of genomic DNA was diluted to 
11 haploid genome templates in the final PCR.
|
The majority of our data were for products of 250 bp or fewer. We have, however, successfully amplified a 665-bp polymorphic region in the elastin gene (4) from PEP products, but we failed to amplify a 1200-bp region when we used a PEP template from the same gene (not shown). Because the PEP generates fragments that are shorter than the template, the proportion of intact copies of template for long PCRs is expected to decrease exponentially. PEP DNA may, therefore, be unsuitable for amplifying long products. When the PEP products were diluted 1:100 in molecular biology-grade water (Eppendorf), degradation occurred in products stored at 4 °C after 2 months. When the diluted products were stored at -20 °C or Tris-EDTA buffer was used as the diluent, degradation was prevented, suggesting that the low pH (5.5) of the water led to slow chemical degradation of the PEP products.
Modifications of the PEP, including fewer cycles and lower primer concentrations, may be suitable for near-linear amplification of DNA from paraffin-embedded tissues, an important consideration for diseases with spontaneous lethal outcomes, where samples are difficult to obtain and frequently the only available samples are paraffin-embedded tissue blocks or microscope slides prepared from tissue sections.
Microsatellite markers have been used to detect aneuploidy in small amounts of fetal material amplified by quantitative PCR (5)(6) or after whole-genome amplification using comparative genome hybridization (7). Although our genotyping experiments were not quantitative, they were encouraging in the sense that both alleles seemed to be amplified. Detailed quantitative analyses are necessary to establish the usefulness of PEP-amplified templates in quantitative PCRs. Several investigators have noted the problem of poor fidelity when PEP or other whole-genome amplification methods, such as the degenerate oligonucleotide-primed PCR, tagged PCR, or Alu-PCR, were used with single cells (8)(9)(10). We would, therefore, not recommend using these methods with such limited amounts of genetic material. Rather, our study was meant to show that PEP would reduce the amount of template DNA required by 300-fold from the amounts currently requested by large genotyping laboratories and would improve the uniformity of intensity and reproducibility of the genotypes.
In conclusion, we have found that the PEP protocol is simple, robust, and reliable. With PEP, the amount of genomic DNA used for one conventional genotyping PCR (30 ng) is more than sufficient for an entire genome scan at 10-centimorgan spacing [400 markers (11)].
Acknowledgments
This work was supported in part by grants from the NIH (Grants NS34395 and HL64310), and a grant from the American Heart Association-Michigan Affiliate.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. Weinsheimer, K. A.B. Goddard, A. R. Parrado, Q. Lu, M. Sinha, E. R. Lebedeva, A. Ronkainen, M. Niemela, E. K. Khusnutdinova, R. I. Khusainova, et al. Association of Kallikrein Gene Polymorphisms With Intracranial Aneurysms Stroke, October 1, 2007; 38(10): 2670 - 2676. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Shibamura, J. M. Olson, C. van Vlijmen-van Keulen, S. G. Buxbaum, D. M. Dudek, G. Tromp, T. Ogata, M. Skunca, N. Sakalihasan, G. Pals, et al. Genome Scan for Familial Abdominal Aortic Aneurysm Using Sex and Family History as Covariates Suggests Genetic Heterogeneity and Identifies Linkage to Chromosome 19q13 Circulation, May 4, 2004; 109(17): 2103 - 2108. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), average (
), and maximum (
) amounts. The thick solid line is the observed data (x). The experimental data agree well with the prediction that amplification reaches a limit either because of the depletion of nucleotides or because the concentration of DNA generated by amplification becomes limiting (