| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Istituto di Tecnologie Biomediche, Consiglio Nazionale delle Ricerche (ITB-CNR), via Fratelli Cervi 93, 20090 Segrate (MI), Italy
2 Istituto di Biologia e Biotecnologia Agraria (IBBA-CNR), via Bassini 15, Milan, 20133 Italy
3 CISI and Dipartimento di Scienze e Tecnologie Biomediche, Università di Milano, via Fratelli Cervi 93, 20090 Segrate (MI), Italy
aauthor for correspondence: 39-02-26422770, e-mail gianluca.debellis{at}itb.cnr.it
Genetic studies in Alzheimer disease (AD) have indicated that its etiology is multifactorial. The apolipoprotein E locus (APOE) is a known major susceptibility factor, and additional genetic loci have been associated with disease development (1)(2). Among many others, the transferrin gene (TF) has been suggested (3) as a candidate locus for AD because it is the major transport protein for iron, which itself is an important factor in free-radical generation. Oxidative stress and free-radical damage occur in AD, which justifies the interest in this protein. Previous studies have shown contrasting results regarding the influence of combinations of TF and APOE alleles (4)(5). We therefore designed a large-scale study of AD patients and controls to ascertain the relevance of TF as a risk factor for AD in conjunction with APOE. Here, we present the method that we have established for the simultaneous typing of the APOE and TF genes based on the ligation detection reaction (LDR)/universal array approach proposed by Gerry et al. (6). This method (Supplemental Fig. 1
, A and B, accompanying the online version of this Technical Brief at http://www.clinchem.org/content/vol49/issue9/) is based on the PCR amplification of the regions including the polymorphic loci for TF and APOE genes. The resulting products are subjected to a multiplexed cycled ligation reaction that uses oligonucleotides designed to differentiate all possible alleles and that includes positive controls useful for normalizing the signals. This approach requires the design of a common LDR probe and two differentiating oligonucleotides for each polymorphic site. The common probe is phosphorylated on the 5' end and contains a zip-code complement on its 3'-terminal position. This tag is used to direct LDR products to specific zip-code addresses attached covalently to the glass support. Zip-codes are sequences that have comparable behavior in terms of thermodynamics and kinetics of hybridization while simultaneously maintaining distinct chemical identities that prevent cross-hybridization. These sequences are completely unrelated to those used for polymorphism detection. Cy3 or Cy5 fluorophores on the 5' end of each differentiating oligonucleotide are used to distinguish a single polymorphism (7).
|
Following such an approach, we designed a set of LDR oligonucleotides to type-three polymorphic sites on the corresponding genomic regions of APOE (codon 112C/T and 158C/T) and TF (codon 570C/T). We included a positive control for each amplicon and a negative (zip-code 15) for determining the background. Our results on a first set of 39 samples of variable genotype (for a total of 117 determinations of single nucleotide determinations) demonstrated the reliability and sensitivity of the procedure.
Universal arrays were prepared by spotting 5' amino-modified zip-code oligonucleotides randomly selected from those proposed by Gerry et al. (6) with a poly(A) tail appended to their 5' end. Eight arrays (6 zip-codes x 10 replicated spots on a 3 x 4 mm area) were generated on Code-Link slides (Amersham) with a Biorobotics MicroGrid II system.
To perform the LDR procedure, we designed a total of 15 oligonucleotides (presented in Table 1
). By PCR, we amplified APOE and TF polymorphic regions from 100 ng of human genomic DNA, as detailed elsewhere (3)(8). The amplicons were pooled, purified by GFX columns (Amersham Pharmacia Biotech), and were quantified on an Agilent 2100 Bioanalyzer. Ligation reaction was performed in a final volume of 20 µL, containing 20 mmol/L Tris–HCl, pH 7.5, 20 mmol/L KCl, 10 mmol/L MgCl2, 0.1 g/L NP-40, 0.01 mmol/L ATP, 1 mmol/L dithiothreitol, 2 pmol of each oligonucleotide, and 50–100 fmol of purified PCR products. The reaction mixture underwent a denaturation step for 2 min at 94 °C, and then 1 µL of 4 U/µL Pfu DNA ligase (Stratagene) was added. The LDR was cycled for 30 cycles at 94 °C for 30 s and at 68 °C for 4 min in a GeneAmp PCR System 9700 thermal cycler (Applied Biosystem). The hybridization solution (65 µL), consisting of ligation reaction mixture, 5x standard saline citrate (SSC; 0.15 mol/L NaCl and 0.015 mol/L sodium citrate), and 0.1 g/L salmon sperm DNA, was heated at 94 °C for 2 min and applied on the slide with a Press-to-seal Silicon isolator (Schleicher&Schuell) system (a gasket with eight incubation chambers). Hybridization reactions were carried out in the dark at 65 °C for 2 h, in a temperature-controlled oven. Arrays were washed in prewarmed 1x SSC–0.1 g/L sodium dodecyl sulfate for 15 min at 65 °C.
|
Fluorescent signals were acquired at 10-µm resolution with a ScanArray 4000 laser scanning system and were quantified with the QuantArray Quantitative Microarray Analysis software (Perkin-Elmer). Signals from replicate spots were background-subtracted and averaged. Normalization between Cy3/Cy5 channels was performed with signals from positive controls, yielding the corrected averaged intensities that were used to calculate allele fractions [allele I/(allele I + allele II)], ranging from 1 (homozygous for allele I) to 0 (homozygous for allele II). To validate our approach, we have typed, by this method, 39 samples [117 single-nucleotide polymorphism (SNP) genotypes] of known genotype [determined by traditional restriction fragment length polymorphism (RFLP) analysis] in the APOE and TF genes. We assayed a blank (no template) LDR reaction to assess potential template-independent ligation and spurious hybridization on the universal array. No signals were present above background, which demonstrated the selectivity of the LDR oligonucleotide mixture. Detection limits were established with a repeated dilution of known heterozygous amplicons. PCR amplicons as small as 10 fmol (
2 ng for each template) were detected by LDR/universal array (data not shown). This detection limit enabled the detection of very faint samples. In Fig. 1A
, some examples of microarray analysis are illustrated. Two of the 117 genotypes determined by this method did not match the corresponding results obtained by RFLP analysis. These samples were sequenced, confirming results obtained by microarray analysis.
The overall performance of our system is summarized in Supplemental Table 1
(accompanying the online version of this Technical Brief at http://www.clinchem.org/content/vol49/issue9/) and is depicted in Fig. 1B
, which details the clustering of data that come from different samples of the identical genotype for each polymorphic site. It should be emphasized that the data unequivocally cluster, and, therefore, unambiguous genotyping can be derived for all 117 tests performed. The use of 8 subarrays (which could be extended up to 12) allows for the parallel analysis of several samples, which increases the throughput and lowers the cost. Furthermore, the multiplexing of LDR in combination with PCR has already been demonstrated. Our experience on the HLA region suggests the feasibility of multiplexing up to 27 polymorphisms (Consolandi et al., manuscript in preparation). This study suggests the feasibility of the LDR/universal array approach as a sensitive tool for allele discrimination. With respect to other approaches for SNP typing (PCR/RFLP, allele-specific PCR, or TaqMan-type assays), our method is more sensitive but requires two steps [two rounds of amplification (PCR + LDR) are needed]. The costs for setup are larger (PCR primers and three labeled oligonucleotides are needed per SNP) with respect to PCR/RFLP and allele-specific PCR but are similar for the TaqMan assay. The main advantage consists in the inherent multiplexing capability that can increase throughput, which decreases cost and time per scored SNP. The capability of performing multiple assays on the same slide further enhances this feature.
To improve the molecular characterization of AD, several other polymorphisms potentially involved in the disease could be included within this analytical system.
Acknowledgments
We thank MIUR and the Consiglio Nazionale delle Ricerche (CNR) for partial financial support to G.D.B and C.B. (FIRB project, COFIN project, 5% Nanotecnologie). Additional data (including a Supplemental Table and a Supplemental Figure) are available as a Data Supplement accompanying the online version of this Technical Brief at http://www.clinchem.org/content/vol49/issue9/.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
