Full Text (PDF)
RNA molecules, which can be simply prepared and amplified by transcription using DNA templates, display versatile functionalities depending on their sequences and higher-order structures. This characteristic allows us to generate novel species of RNA that bind target molecules (aptamers) and catalysts (ribozymes) by in vitro selection methods using large populations of random RNA sequences. Over the last 15 years, researchers have created many aptamers and ribozymes for their purposes. Consequently, a modified RNA aptamer that binds to vascular endothelial growth factor was recently approved as a new treatment for age-related macular degeneration (1). In addition, efficient ribozymes for charging amino acid analogs onto a transfer RNA (tRNA) are available for incorporating amino acid analogs into proteins (2). As a result, RNA molecules have become useful biopolymers for biomedical research, diagnostics, and therapeutics.
Despite these prospects, RNA-based biotechnology is still restricted. One of the problems is that RNA molecules comprise only four different, but similar, components (nucleotides). In contrast, proteins consist of 20 different, unique amino acids, including acidic, basic, and hydrophobic residues. In fact, in the evolution of life, the limited variety of nucleotides caused protein enzymes to displace most of the catalytic RNA functions, although various ribozymes might have appeared in the early stage of the beginning of life. This fact paradoxically suggests that even simple RNA molecules are capable of working as functional molecules and that their potential could be increased further by introducing extra components into RNA.
Nucleotide analogs can be introduced into RNA by several methods, such as chemical synthesis, posttranscriptional modification, and enzymatic incorporation by transcription. Although chemical synthesis is useful for the site-specific introduction of chemically stable analogs (3), it is difficult to synthesize RNA molecules longer than 100 nucleotides (100-mer). Other methods, such as chemical modifications of RNA molecules and enzymatic incorporation during transcription, were also developed. For example, chemical biotinylation of RNA at its 3′ terminus was achieved by using biotin-conjugated hydrazide for the immobilization of RNA molecules, in which the 2′,3′-dihydroxy groups of the 3′-terminal nucleoside are oxidized to be functionalized (2,4,5). In addition, RNA can be biotinylated at the 5′ end by a one-step transcription procedure, using biotinyl-guanosine 5′-monophosphate under the control of the conventional T7 promoter (6) or N6-biotin derivatives of AMP as transcription initiators under the T7 Φ2.5 promoter (7). However, these methods are restricted to the terminal modification of RNA molecules. Alternative methods involve random incorporation of modified nucleotides by transcription (8), but it is obviously impossible to control the modification positions.
A more attractive method is the expansion of the genetic alphabet by an unnatural base pair system (9), allowing for the site-specific, enzymatic incorporation of extra components by RNA polymerases, mediated by the unnatural base pair (Figure 1 A). For this purpose, an unnatural base pair that selectively and efficiently functions together with the natural A-T(U) and G-C base pairs in transcription is required. In this essay, we will discuss the unnatural base pair systems, focusing on our research, for the site-specific incorporation of nucleotide analogs into RNA molecules by transcription using T7 RNA polymerase, the most useful enzyme for RNA preparation.
Creation of Unnatural Base Pairs
The first unnatural base pair system was developed by Benner and colleagues. They designed and synthesized a series of unnatural base pairs, such as isoguanine-isocytosine (isoG-isoC) and xanthosine-diaminopyrimidine (X-K) (Figure 1B), with different hydrogen-bonding patterns from those of the natural base pairs (10,11). The A-T(U) and G-C base pairs have two and three hydrogen bonds, respectively. Each hydrogen bond is composed of the hydrogen-donor and hydrogen-acceptor residues or atoms, and the different combinations of donor-acceptor patterns in the unnatural base pairs from those in the natural base pairs confer their selectivity in DNA and RNA biosyntheses. The isoG-isoC pair was used for the introduction, by T7 RNA polymerase (12) of a modified isoG, N6-(6-aminohexyl)-isoG, into RNA fragments, using DNA templates containing isoC. The extra amino group of the modified isoG in the RNA fragments could be useful as a platform for post-transcriptional modification. Unfortunately, this unnatural base pair system has some shortcomings, including mispairing with the natural bases because of the tautomerization of isoG and the poor recognition of the 2-aminopyrimidines, such as isoC and K, by some RNA polymerases.
