Practical approach on microarray based allelespecific primer extension

There are multiple ways to successfully accomplish the multiplexed allele-specific primer extension assays using microarray format. In the following section we describe the protocols used in our laboratory, which have been proven to be robust and suitable for the multiplexed SNP genotyping assays for medium throughput projects. The different steps and estimated time span in the multiplexed allele-specific primer extension genotyping assays are schematically presented in Figure 8.2.

Manufacturing microarrays for allele-specific primer extension

In genotyping microarrays, a probe is hybridized to a single sample (or to a pooled sample mixture), unlike in gene expression two-color arrays, in which two samples compete in the hybridization reaction. The genotyping microarrays can be used as regular microarrays, where the whole array surface is being used to monitor the hybridization of one sample. Alternatively, the array surface can be divided into subarrays, where each subarray is hybridized with a different sample (see Figure 8.1A for the array-of-arrays layout). In the latter case the same set of allele-specific oligos are printed on each subarray. We routinely use duplicate or even triplicate spots on each subarray, to guarantee the reliability of genotyping. Due to the

pOOOOl oooooo oooooo oooooo oooooo oooooo oooooo oooooo

Two ASOs printed on the glass with different 3' nucleotides


Two ASOs printed on the glass with different 3' nucleotides

Allele specific nucleotide

Glass support

Allele specific nucleotide


Glass support

Hybridization of the RNA



Extension of the RNA, only the ASO with 3' G will be extended with fluorescent nucleotides





standardized automation, it is conventional to use the same well-to-well spacing as in the regular 384-well plates.

The design of arrays starts with the design of oligonucleotides for each bi-allelic SNP to be genotyped. PCR primers are designed for each ampli-con as described in Figure 8.1B. For each SNP two ASOs are designed which define the alleles in the SNP locus. Each ASO comprises three structural elements (see Figure 8.1C). The first element of an ASO is an amine group at the 5' terminus, which covalently binds the oligonucleotide to the chemically activated slide surface. The amine group is followed by a spacer sequence (for example TTT TTT TTT), which provides physical distance from the slide surface. The third element is a locus-specific sequence followed by the allele-specific sequence, which detects the variant alleles in the sample. The locus-specific sequence of an ASO is 16 to 22 nucleotides in length resulting in a homogenous melting temperature for all ASOs present in the hybridization assay. In the design of each ASO stringent physical parameters are followed, such as avoiding long homonucleotide repeats, secondary structures and self-dimerization. These parameters can be estimated by computer software, for example Oligonucleotide properties calculator ( The two ASOs required for each SNP differ only in their 3' nucleotides, which define the two alleles to be detected.

Microarray slides, carrying the ASOs on their surface, can be manufactured in a variety of ways. Contact printing is one of the most commonly used methods. In contact printing or 'spotting' a robotic

Figure 8.1.

Array-of-array layout, ASO oligonucleotides and PCR primers, allele-specific primer extension reaction. (A) Array-of-array layout. Each genotyped sample is applied to a subarray. The subarrays are spaced according to the well-to-well distances of 384-well plates. The subarrays are formed with tape-gridding or with a histological wax pen. All the subarrays are identical, containing the ASOs, two for each SNP locus and typically duplicated in order to increase the assay reliability.

(B) Localization of the different oligonucleotides around the SNP locus. The PCR primers (priF and priR) are designed to amplify a region of 100-200 bps around the SNP to be genotyped. In the forward primer (priF), a T7 promoter sequence is added to the 5' end. The two ASOs, targeting the SNP (here Y, i.e. C or T alleles in the forward strand), correspond to a sequence in the upper strand.

(C) Composition of the ASOs. The ASO contains an amino-group at its 5' end, required for the covalent attachment to the array surface. A linker region, typically T9, is required to provide a physical distance from the locus-specific sequence to the slide surface and to enhance the flexibility of the oligonucleotide. The two ASOs are identical in their sequence, except for their 3' terminal nucleotide (either G or A, complementary to the SNP alleles). (D) Hybridization of the amplicon to the ASOs. The PCR-amplified region, containing the SNP to be genotyped, is transcribed to RNA and hybridized with the ASOs on the array. (E) Allele-specific primer extension. Depending on the alleles, either ASO1, ASO2 (homozygote) or both (heterozygote) is/are extended in the reverse-transcription reaction. Fluorescent nucleotides are incorporated to the extension product, and are required for the subsequent detection.

Oligonucleotide design

Computer assisted design

Manufacturing slides

Setup 1/2 hr, automated printing overnight

Multiplexed PCR amplification

Automated setup V2 hr, thermocycling 2 hrs

Reduction of sample complexity: IVT & DNAse I

Automated setup V2 hr, incubate 2V2 hrs

Hybridization on microarray

Automated setup 15 min, incubate 20 min

Allele-specific primer extension reaction

Automated setup 15 min, incubate 20 min

Image collection & signal quantitation

Analyze 15-30 min / slide

Data analysis, allele calling and genotype assignment

Analyze 15 min / slide

Flow chart of the allele-specific primer extension genotyping and data analysis. Most of the steps are carried out in multi-well plates using pipetting robotics. (i) Once the SNPs to be genotyped are identified, the primers are designed for the locus, as described in Figure 8.1B and Table 8.2. (ii) The slides are manufactured with a robotic arrayer and the printing is done overnight, depending on the amount of arrays to be produced. (iii) The loci containing SNPs to be genotyped are multiplex PCR-amplified from the samples. PCR reactions are performed with touchdown annealing in about 2 h. (iv) In order to prevent self-pairing of the PCR products in the subsequent hybridization step, the PCR products are transcribed to RNA by T7 polymerase, followed by DNA template degradation using DNasel. These reactions take about 2,5 h. (v) Hybridization of the RNAs to the ASOs is carried out on the microarray in a humid chamber and takes around 0.5 h. (vi) The allele-specific primer extension is carried out using reverse-transcriptase, for example MMLV-RT. The allele-specific extension incorporates the fluorescent nucleotides into the extension product. (vii) The fluorescent emission is detected using a standard microarray scanning instrument, producing an image, which is then quantitated. (viii) The quantitated image data is background-subtracted and normalized. The allele calling is done by clustering methods, followed by genotype assignments.

arrayer is used to transfer small volumes of ASOs from a microtiter plate onto a microscopic slide. These slides are aminosilane-coated for cova-lent binding with ASOs (36). The arrayer dips a quill pin into ASO solution containing 20 ^M of oligonucleotide in a 1 x Micro Spotting Solution (Arraylt Microarray Technology) and subsequently moves the pin over the slide. When all spots are printed for a given ASO, the pin is washed in an ultrasonic water bath washing station and vacuum-dried to prevent carry-over contamination between ASO spots. By repeating the cycle of dipping, printing and washing, the arrayer builds an array-of-arrays layout. With contact printing hundreds of spots can be replicated from a single dip. The produced spots are 100-500 ^m in diameter depending on the size of the pin and surface chemistry used. The temperature and humidity of the printing unit also affect the printing process and extra care should be paid to control for this.

Sample preparation: PCR of the DNA samples

Selected genomic regions containing the SNPs to be monitored are amplified by PCR to yield a sufficient amount of DNA molecules for microarray-based detection of the SNP genotypes. PCR amplification of the sample is performed in a multiplexed fashion, that is all primer pairs are amplified simultaneously in a single-tube reaction, each primer pair producing a 100- to 200-nucleotide-long amplicon for the SNP locus. A feasible level of PCR multiplexing is up to 20 SNP loci in a single reaction. This can be extended by pooling different multiplex PCR products. The growing complexity of the oligonucleotide mixture in the multiplex PCR reactions typically results in around 80% of successful genotyping assays giving distinct genotype clusters in data analysis. Success rate is expected to decrease as the level of multiplexing increases.

Successful multiplexed PCR assay requires careful primer design, which takes into account uniform melting temperature and amplicon length for all primers. Parameters for oligonucleotide selection are shown in Table 8.2. In order to prevent mishybridization of sample DNA and oligonucleotides during the polymerase chain reaction, two different actions are taken. Firstly, a mispriming library containing known repetitive elements of the human genome, such as Alu repeats, is utilized, preventing the primers from targeting any known repetitive sequences. Secondly, cross binding to other targets in the multiplex PCR is prevented by including all the other multiplexed loci sequences and primers in the primer design process. This second step is then iterated for all the loci in the same multiplex PCR design. The multiplex PCR primer design system is accessible on our website at, where the actual underlying primer design algorithm is the Primer3 program (37). Each PCR product contains a T7 RNA polymerase promoter sequence (TAA TAC GAC TCA CTA TAG GGA GA) introduced by a T7-tagged forward primer, needed later for in vitro transcription, which is introduced by a tailing of the 5' end of the PCR primer on the opposite strand of the ASOs (see Figure 8.1B).

Multiplexed PCR reactions are carried out in a microtiter plate format in a reaction volume of 5-20 ^.l using 1-20 ng of DNA as a template. Thermocycling is performed in a touchdown manner where the annealing

Table 8.2. PCR primer design parameters with Primer3







PCR Primer size

18 nt

20 nt

23 nt

PCR Primer Tm value3




Primer maximum GC content


ASO oligonucleotide length



ASO oligonucleotide Tm




a Tm values calculated using the Nearest Neighbor method a Tm values calculated using the Nearest Neighbor method temperature is decreased by 0.5-1°C during the first few cycles, which produces few specific copies of amplicons at the optimal annealing temperature. After touchdown cycling a final amplification is performed at the lowest annealing temperature of the primers in the assay.

Improving the specificity of the hybridization: reduction of template complexity

In order to avoid self-pairing of the PCR products, they are not directly used for hybridization with the ASOs. Rather both the specificity of the hybridization as well as number of target molecules is increased by transcription of the PCR products to single-stranded RNA molecules, using the T7 promoter sequence tailed on the forward PCR primer (see above). In vitro transcription is performed in a 4-^l reaction volume containing 2.0 ^l of PCR template, 0.85 x T7 reaction buffer, 6.17 mM of each deoxyribonu-cleotides, 8.65 mM of DTT and 0.35 ^l of T7 RNA polymerase solution (modified from Ampliscribe T7 High Yield Transcription Kit, Epicentre Biotechnologies).

The transcription reaction is followed by the degradation of the PCR products by the addition of 1.0 ^l of DNaseI solution containing 0.1 U of DNAseI in 1 x T7 reaction buffer. All the enzymatic steps are easy to automate and can be carried out in a microtiter plate format in a reaction volume as low as 5 ^l. This results in RNA target molecules that act as a template for the extension of the spotted ASOs.

Hybridization of samples and allele-specific primer extension

In the hybridization step each sub-array on the slide is covered with a droplet of transcribed ssRNA sample and incubated in a humid chamber at 42°C for 20 min. To prevent contamination of adjacent sub-arrays, sample wells are formed using special tape grids or a histological wax pen (Pap Pen, Daido Sangyo Co., Ltd, Japan). In the hybridization the complementary ssRNA molecules anneal to ASOs on the microarray surface. After incubation the slide is washed in buffer containing 0.5 x TE, 0.3 M NaCl and 0.1% Triton X-100, rinsed in distilled water and dried by pressurized air.

For the allele-specific primer extension each sub-array is covered with 2.0 ^l of primer extension cocktail containing 2 U Moloney Murine Leukemia Virus reverse transcriptase (MMLV-RT), 10 mM of DTT, 1 ^.M of

Cy5-labeled dCTP and dUTP nucleotides, 1.0 |M of dATP and dGTP, 0.46 M trehalose and 8% glycerol in 1 x MMLV-RT reaction buffer. The slide is subsequently incubated at 52°C for 20 min. High incubation temperature enhances allelic discrimination in the allele-specific primer extension reaction. Trehalose and glycerol are used to stabilize the polymerase in the extension reaction performed above the optimal temperature for MMLV-RT.

During the primer extension reaction MMLV-RT polymerizes the ASOs having complete hybridization with the ssRNA, including the crucial 3' nucleotide with deoxynucleotides in the cocktail, simultaneously introducing fluorescent nucleotides. If the ssRNA sample has a mismatch with the 3' nucleotide of the ASO, the primer extension is suppressed. However, often the allelic discrimination of the primer extension reaction is not complete and some residual extension can take place, which needs to be compensated for by the data analysis. In a sample that is homozygous for a given SNP only one of the two ASOs is fluorescently labeled, whilst in a heterozygous sample both ASOs are labeled. After primer extension, the microarray slide is washed in buffer containing 0.5 x TE, 0.3 M NaCl and 0.1% Triton X-100, rinsed in distilled water and dried with pressurized air.

Image collection of the microarray slide and signal quantification

A digital image of the microarray slide is obtained by a microarray scanner with CCD detector (Scan Array 4000 laser scanner, GSI Lumonics/Packard Bioscience). The scanner measures the emitted fluorescence of the excited ASO spots on the microarray surface and produces a corresponding digital image. Usually a 16-bit TIFF image is used to store a high dynamic range of values per pixel.

The signal intensities of the microarray spots are quantified by image analysis software. The basis for signal quantification is to identify the spot location in the image, define its borders and morphology and quantify the signal and background intensities as well as other parameters. The simplest form of the quantitative spot analysis consists of defining the center of the spot and measurement of the signal within a given radius. This approach is hampered by the fact that contact-printed spots seldom are perfect circles and there might be differences in size and morphology between different ASO spots. The reliability of the allelic discrimination can be increased by utilizing an internal hybridization control oligonucleotide printed within each ASO spot. This can be accomplished by printing an equal amount of a control oligonucleotide to each ASO spot and respectively adding 5'-phos-phorylated Cy3-labeled oligonucleotide, complementary to the control oligonucleotide on the array, to the primer extension cocktail. The emitted signal from the internal control is acquired using a different wavelength to the ASOs and is used to normalize the ASO signal.

0 0

Post a comment