Naturally occurring bases can be divided into two classes according to their structure: •
Pyrimidines are six-membered heterocyclic with nitrogen atoms in position 1 and 3. •
Purines are bicyclic, consisting of a pyrimidine fused to an imidazole ring. Artificial nucleotides (
Unnatural Base Pairs (UBPs) named
d5SICS UBP and
dNaM UBP) have been inserted into bacterial DNA but these genes did not template mRNA or induce protein synthesis. The artificial nucleotides featured two fused aromatic rings which formed a (d5SICS–dNaM) complex mimicking the natural (dG–dC) base pair.
Mutagens One of the most common base analogs is
5-bromouracil (5BU), the abnormal base found in the mutagenic nucleotide analog BrdU. When a nucleotide containing 5-bromouracil is incorporated into the DNA, it is most likely to pair with adenine; however, it can spontaneously shift into another
isomer which pairs with a different
nucleobase,
guanine. If this happens during DNA replication, a guanine will be inserted as the opposite base analog, and in the next DNA replication, that guanine will pair with a cytosine. This results in a change in one base pair of DNA, specifically a
transition mutation. Additionally, nitrous acid (HNO2) is a potent mutagen that acts on replicating and non-replicating DNA. It can cause deamination of the amino groups of adenine, guanine and cytosine. Adenine is deaminated to
hypoxanthine, which base pairs to cytosine instead of thymine. Cytosine is deaminated to uracil, which base pairs with adenine instead of guanine. Deamination of guanine is not mutagenic. Nitrous acid-induced mutations also are induced to mutate back to wild-type.
Fluorophores Commonly
fluorophores (such as
rhodamine or
fluorescein) are linked to the ring linked to the sugar (in para) via a flexible arm, presumably extruding from the major groove of the helix. Due to low processivity of the nucleotides linked to bulky adducts such as fluorophores by
Taq polymerases, the sequence is typically copied using a nucleotide with an arm and later coupled with a reactive fluorophore (indirect labelling): • Amine reactive:
aminoallyl nucleotides contain a primary amine group on a linker that reacts with the amino-reactive dye such as
cyanine or
Alexa Fluor dyes, which contain a reactive leaving group like succinimidyl ester (NHS). Base-pairing amino groups are not affected. • Thiol reactive: thiol-containing nucleotides react with the fluorophore linked to a reactive leaving group like maleimide. •
Biotin-linked nucleotides rely on the same indirect labelling principle (and fluorescent streptavidin) and are used in
Affymetrix DNAchips. Fluorophores find a
variety of uses in medicine and biochemistry. The most commonly used and commercially available fluorescent base analogue, 2-aminopurine (2-AP), has a high-fluorescence quantum yield free in solution (0.68) that is considerably reduced (appr. 100 times but highly dependent on base sequence) when incorporated into nucleic acids. The emission sensitivity of 2-AP to immediate surroundings is shared by other promising and useful fluorescent base analogues like 3-MI, 6-MI, 6-MAP, pyrrolo-dC (also commercially available), modified and improved derivatives of pyrrolo-dC, furan-modified bases and many other ones (see recent reviews). This sensitivity to the microenvironment has been utilized in studies of e.g. structure and dynamics within both DNA and RNA, dynamics and kinetics of DNA-protein interaction and electron transfer within DNA. A newly developed and very interesting group of fluorescent base analogues that has a fluorescence quantum yield that is nearly insensitive to their immediate surroundings is the tricyclic cytosine family. 1,3-Diaza-2-oxophenothiazine, tC, has a fluorescence quantum yield of approximately 0.2 both in single- and in double-strands irrespective of surrounding bases. Also the oxo-homologue of tC called tCO (both commercially available), 1,3-diaza-2-oxophenoxazine, has a quantum yield of 0.2 in double-stranded systems. However, it is somewhat sensitive to surrounding bases in single-strands (quantum yields of 0.14–0.41). The high and stable quantum yields of these base analogues make them very bright, and, in combination with their good base analogue properties (leaves DNA structure and stability next to unperturbed), they are especially useful in fluorescence anisotropy and FRET measurements, areas where other fluorescent base analogues are less accurate. Also, in the same family of cytosine analogues, a FRET-acceptor base analogue, tCnitro, has been developed. Together with tCO as a FRET-donor this constitutes the first nucleic acid base analogue FRET-pair ever developed. The tC-family has, for example, been used in studies related to polymerase DNA-binding and DNA-polymerization mechanisms.
Natural non-canonical bases In a cell, there are several non-canonical bases present: CpG islands in DNA (often methylated), all eukaryotic mRNA (capped with a methyl-7-guanosine), and several bases of rRNAs (methylated). Often, tRNAs are heavily modified postranscriptionally in order to improve their conformation or base pairing, in particular in or near the anticodon:
inosine can base pair with C, U, and even with A, whereas thiouridine (with A) is more specific than uracil (with a purine). Other common tRNA base modifications are pseudouridine (which gives its name to the
TΨC loop), dihydrouridine (which does not stack as it is not aromatic), queuosine, wyosine, and so forth. Nevertheless, these are all modifications to normal bases and are not placed by a polymerase. Diaminopurine basepairs perfectly with thymine as it is identical to adenine but has an amine group at position 2 forming 3 intramolecular hydrogen bonds, eliminating the major difference between the two types of basepairs (weak A-T vs strong C-G). This improved stability affects protein-binding interactions that rely on those differences. Other combination include: • Isoguanine and isocytosine, which have their amine and ketone inverted compared to standard guanine and cytosine. They are not used probably as tautomers are problematic for base pairing, but isoC and isoG can be amplified correctly with PCR even in the presence of the 4 canonical bases. • Diaminopyrimidine and xanthine, which bind like 2-aminoadenine and thymine but with inverted structures. This pair is not used as xanthine is a deamination product. However, correct DNA structure can form even when the bases are not paired via hydrogen bonding; that is, the bases pair thanks to hydrophobicity, as studies have shown with DNA
isosteres (analogues with same number of atoms) such as the thymine analogue 2,4-difluorotoluene (F) or the adenine analogue 4-methylbenzimidazole (Z). An alternative hydrophobic pair could be isoquinoline and pyrrolo[2,3-b]pyridine Other noteworthy basepairs: • Several fluorescent bases have also been made, such as the 2-amino-6-(2-thienyl)purine and pyrrole-2-carbaldehyde base pair. • Metal-coordinated bases, such as pairing between a pyridine-2,6-dicarboxylate (tridentate ligand) and a pyridine (monodentate ligand) through square planar coordination to a central copper ion. • Universal bases may pair indiscriminately with any other base, but, in general, lower the melting temperature of the sequence considerably; examples include 2'-deoxyinosine (hypoxanthine deoxynucleotide) derivatives, nitroazole analogues, and hydrophobic aromatic non-hydrogen-bonding bases (strong stacking effects). These are used as proof of concept and, in general, are not utilized in degenerate primers (which are a mixture of primers). • The numbers of possible base pairs is doubled when
xDNA is considered. xDNA contains expanded bases, in which a benzene ring has been added, which may pair with canonical bases, resulting in four additional possible base-pairs (xA-T, xT-A, xC-G, xG-C) with eight bases (or 16 bases if the unused arrangements are used). Another form of benzene added bases is yDNA, in which the base is widened by the benzene.
Metal base-pairs In metal base-pairing, the Watson-Crick hydrogen bonds are replaced by the interaction between a metal ion with nucleosides acting as ligands. The possible geometries of the metal that would allow for duplex formation with two
bidentate nucleosides around a central metal atom are
tetrahedral,
dodecahedral, and
square planar. Metal-complexing with DNA can occur by the formation of non-canonical base pairs from natural nucleobases with participation by metal ions and also by the exchanging the hydrogen atoms that are part of the Watson-Crick base pairing by metal ions. Introduction of metal ions into a DNA duplex has shown to have potential magnetic or conducting properties, as well as increased stability. Metal complexing has been shown to occur between natural
nucleobases. A well-documented example is the formation of T-Hg-T, which involves two deprotonated
thymine nucleobases that are brought together by Hg2+ and forms a connected metal-base pair. This motif does not accommodate stacked Hg2+ in a duplex due to an intrastrand hairpin formation process that is favored over duplex formation. Two thymines across from each other do not form a Watson-Crick base pair in a duplex; this is an example where a Watson-Crick basepair mismatch is stabilized by the formation of the metal-base pair. Another example of a metal complexing to natural nucleobases is the formation of A-Zn-T and G-Zn-C at high pH; Co2+ and Ni2+ also form these complexes. These are Watson-Crick base pairs where the divalent cation in coordinated to the nucleobases. The exact binding is debated. A large variety of artificial nucleobases have been developed for use as metal base pairs. These modified nucleobases exhibit tunable electronic properties, sizes, and binding affinities that can be optimized for a specific metal. For example, a nucleoside modified with a pyridine-2,6-dicarboxylate has shown to bind tightly to Cu2+, whereas other divalent ions are only loosely bound. The tridentate character contributes to this selectivity. The fourth coordination site on the copper is saturated by an oppositely arranged pyridine nucleobase. The asymmetric metal base pairing system is orthogonal to the Watson-Crick base pairs. Another example of an artificial nucleobase is that with hydroxypyridone nucleobases, which are able to bind Cu2+ inside the DNA duplex. Five consecutive copper-hydroxypyridone base pairs were incorporated into a double strand, which were flanked by only one natural nucleobase on both ends. EPR data showed that the distance between copper centers was estimated to be 3.7 ± 0.1 Å, while a natural B-type DNA duplex is only slightly larger (3.4 Å). The appeal for stacking metal ions inside a DNA duplex is the hope to obtain nanoscopic self-assembling metal wires, though this has not been realized yet.
Unnatural base pair (UBP) An unnatural base pair (UBP) is a designed subunit (or
nucleobase) of
DNA that is created in a laboratory and does not occur in nature. In 2012, a group of American scientists led by Floyd Romesberg, a chemical biologist at the
Scripps Research Institute in San Diego, California, published that his team had designed two unnatural base pairs named
d5SICS and
dNaM. More technically, these artificial
nucleotides bearing hydrophobic
nucleobases feature two fused
aromatic rings that form a d5SICS–dNaM complex or base pair in DNA. In 2014, the same team reported that they had synthesized a
plasmid containing natural T-A and C-G base pairs along with the best-performing UBP Romesberg's laboratory had designed and inserted it into cells of the common bacterium
E. coli, which successfully replicated the unnatural base pairs through multiple generations. This is the first known example of a living organism passing along an expanded genetic code to subsequent generations. This was in part achieved by the addition of a supportive algal gene that expresses a
nucleotide triphosphate transporter which efficiently imports the triphosphates of both d5SICSTP and dNaMTP into
E. coli bacteria. Transcription of DNA containing unnatural base pairs and translation of corresponding mRNA were actually achieved recently. In November 2017, the same team at the
Scripps Research Institute that first introduced two extra nucleobases into bacterial DNA reported having constructed a semi-synthetic
E. coli bacteria able to make proteins using such DNA. Its DNA contained six different
nucleobases: four canonical and two artificially added, dNaM and dTPT3 (these two form a pair). The bacteria had two corresponding RNA bases included in two new codons, additional tRNAs recognizing these new codons (these tRNAs also contained two new RNA bases within their anticodons) and additional amino acids, enabling the bacteria to synthesize "unnatural" proteins. Another demonstration of UBPs were achieved by Ichiro Hirao's group at
RIKEN institute in Japan. In 2002, they developed an unnatural base pair between 2-amino-8-(2-thienyl)purine (s) and pyridine-2-one (y) that functions
in vitro in transcription and translation, for the site-specific incorporation of non-standard amino acids into proteins. In 2006, they created 7-(2-thienyl)imidazo[4,5-b]pyridine (Ds) and pyrrole-2-carbaldehyde (Pa) as a third base pair for replication and transcription. Afterward, Ds and 4-[3-(6-aminohexanamido)-1-propynyl]-2-nitropyrrole (Px) was discovered as a high fidelity pair in PCR amplification. In 2013, they applied the Ds-Px pair to DNA aptamer generation by
in vitro selection (SELEX) and demonstrated the genetic alphabet expansion significantly augment DNA aptamer affinities to target proteins. == Applications ==