Up until this point we have considered how nucleic acid hybridization is commonly used as an assay. That is, a well-known nucleic acid or oligonucleotide probe population is used to interrogate some other, less understood nucleic acid population, and the object is simply to get some information about the latter population (we might want to know the sizes of the hybridizing fragments, or their copy number, or where they are expressed in cells and tissues, and so on). But we can also use nucleic acid hybridization for a different purpose: to selectively purify a desired type of nucleic acid sequence. That is made possible by covalently attaching to the probe molecules some other molecule that can bind with very high specificity to “capture molecules” immobilized on a surface.
The most popular approach is to covalently attach biotin to the probe. Biotin, a naturally occurring vitamin (known as vitamin B7 or vitamin H), just happens to have an extraordinarily high affinity for streptavidin, a protein that originates from a Streptomyces bacterium. (A homotetramer of streptavidin can bind biotin with a dissociation constant of about 10−14 mol/L, one of the strongest noncovalent interactions between natural bio logical molecules; the strength of the interaction depends on the formation of numerous hydrogen bonds and van der Waals interactions between biotin and streptavidin.)
After biotin-linked probe molecules have been allowed to hybridize to complementary sequences in a test sample, the probe–test-sample heteroduplexes can be captured using magnetized beads coated with streptavidin. The magnetized beads, with attached probe–test-sample heteroduplexes, can then be selectively removed using a magnet and the desired nucleic acids can be eluted (Figure 1).
Fig1. Selective purification of desired nucleic acid sequences using nucleic acid hybridization. (A) Structure of streptavidin-coated magnetic beads. (B) Example of purification. Here we imagine a situation where we wish to purify sequences from a family of many genes that work in some common cell signaling pathway, X, and are considered as possible candidates for having a disease-causing mutation in a group of patients. The idea is to take genomic DNA from individual patients, fragment it, and then mix the fragments with a combination of cloned DNA sequences representing all the normal genes that work in pathway X. The DNA sequences in the mixture are then denatured and allowed to re-anneal, allowing heteroduplexes to form between biotinylated pathway X sequences and complementary sequences from the patient. After streptavidin-coated magnetic beads are added, heteroduplexes with an end-biotin group are bound via the streptavidin to the beads, and can be selectively removed using a magnet. The desired X pathway gene sequences in the patient can be eluted by heating the sample so as to break the hydrogen bonds of the heteroduplex. The resulting purified sequences from each patient can then be investigated by DNA sequencing.
