Targeted genome-wide enrichment of functional regions

Abstract

Only a small fraction of large genomes such as that of the human contains the functional regions such as the exons, promoters, and polyA sites. A platform technique for selective enrichment of functional genomic regions will enable several next-generation sequencing applications that include the discovery of causal mutations for disease and drug response. Here, we describe a powerful platform technique, termed "functional genomic fingerprinting" (FGF), for the multiplexed genomewide isolation and analysis of targeted regions such as the exome, promoterome, or exon splice enhancers. The technique employs a fixed part of a uniquely designed Fixed-Randomized primer, while the randomized part contains all the possible sequence permutations. The Fixed-Randomized primers bind with full sequence complementarity at multiple sites where the fixed sequence (such as the splice signals) occurs within the genome, and multiplex amplify many regions bounded by the fixed sequences (e.g., exons). Notably, validation of this technique using cardiac myosin binding protein-C (MYBPC3) gene as an example strongly supports the application and efficacy of this method. Further, assisted by genomewide computational analyses of such sequences, the FGF technique may provide a unique platform for high-throughput sample production and analysis of targeted genomic regions by the next-generation sequencing techniques, with powerful applications in discovering disease and drug response genes.

Figure 1. The consensus sequences of genetic regulatory elements.

The consensus sequences of the gene regulatory regions are usually 6–12 bases long. Since the donor (5') splice signal reads into the intron, its complementary sequence is used to design a primer that will read into and amplify the exon. In this example, the fixed sequence is on the 5' end, and the randomized sequence is on the 3' end of the FR primer.

Figure 2. Design of an FR primer.

To a core fixed sequence (ATCTG), a series of Ns (A, T, G and C) are added in equimolar concentrations at each step of the oligo-nucleotide synthesis. This process generates all possible sequences of length n, when n bases are randomized, such that each variable sequence is attached to the end of the fixed sequence.

Figure 3. Multiplexed amplification of exons by splice signal FR primers.

The FR primers for the donor (5′ splice signal) and the acceptor (3′ splice signal) splice sequences bind the exons with complementary base pairing over the entire length of the primers, by virtue of the presence of all the possible sequences within the randomized sequence portion of the FR primer. Only the specific primer molecule from the FR primer population is expected to bind selectively with full complementarity at the fixed sequence target site at a high Tm condition. The FR primers amplify multiple exons since they are capable of binding many exons within the human genome with complete sequence complementarity, wherever the fixed sequence binds.

Figure 4. Multiplex genome-wide exon amplification based on MYBPC3 gene exons 7 and 8.

The human genomic DNA was PCR-amplified under standard conditions at 58°C with primers designed from the donor (5′) splice signal sequence (from exon 7) and the acceptor (3′) splice signal sequence (from exon 8) of the MYBPC3 gene. It was also amplified by the same FR primer pairs with decreasing number of fixed bases and increasing number of random bases (Ns) as shown in Table 1. The expected fragment (438 bases for the combined exon 7, intron 7 and exon 8) is present in all the lanes, and the number of fragments amplified increased with increasing Ns in the FR primers. M1 & M2 are marker lanes. Lane CS shows the computer simulated exon fingerprint obtained with the same primers used for lane 5, with four bases removed from the 5′ end of the forward primer and three bases removed from the 5′ end of the reverse primer (see text).