Xeno Nucleic Acids (XNAs) Revisited

  • X (‘Xeno’) in XNAs Are Structurally Strange Congeners of DNA or RNA
  • Liu et al. Report X = 3→2Phosphonomethylthreosyl
  • Xenobiology Is an Emerging and Intriguing New Frontier

In scientific nomenclature, ‘xeno‘ is used to indicate something strange or different relative to what exists in nature, and is derived from the Greek word xenos for strange. Xeno nucleic acids (XNAs), reviewed here, are intriguingly “strange” synthetic polymers. They have natural nucleobases (A/G/C/T or U) for encoding genetic information like DNA or RNA, but their backbones are different from the naturally occurring ribose and phosphodiester moieties.  A/G/C/T- or U-bearing XNAs can therefore be thought of as building blocks for non-natural nucleic acid polymers.

Several years ago, I blogged about XNAs in the context of the evolution of DNA/RNA-based life and genetics within the prebiotic universe, which is generally assumed to have existed ~4 billion years ago. All studies of XNAs have previously been limited to model systems in vitro, which begs the question of whether any type of XNA molecule can genetically function in vivo. By reading on, you’ll learn that the answer is yes, as recently reported by Liu et al., and I think you’ll be very surprised by the “strangeness” of the particular XNA found to perform this remarkable feat.


Adapted from Anosova et al. Nucleic Acids Res.44, 1007–1021 (2016)

To put the fascinating findings of Liu et al. into perspective, here’s a brief synopsis of the backstory involving the earlier work of others on (L)-α-threofuranosyl (aka threose)-based nucleic acid (TNA), which is shown below and represents the most advanced example of an XNA. The seminal work on TNAs containing opposingly connected (3’→2’) phosphodiester bridges was published in Sciencemagazine in 2000 by a team led by Albert Eschenmoser, who was already famous for his part in the prodigious total synthesis of vitamin B12, which took 12 years!

These researchers reported that TNAs undergo “informational base pairing in antiparallel strand orientation” (i.e. Watson-Crick hybridization), and are capable of cross-pairing with RNA and DNA. Derived from a sugar containing only four carbons vs. five carbons for ribose, TNA was posited as the structurally simplest of all naturally derived XNAs.

Follow-on investigations by Chaput and coworkers led to a publication in 2013 that detailed the surprising ability of TNA to be transcribed from DNA by an available polymerase and then reverse transcribed back into DNA using another available polymerase. The transcription step required chemical synthesis of the four TNA triphosphates (tNTPs) for incorporation by a DNA primer-template complex (black), as depicted here.

Adapted with permission from Chaput and coworkers J. Am. Chem. Soc. 135, 3583−3591 (2013)

More specifically, the DNA primer was annealed to the DNA template in buffer by heating it at 95 °C for 5 min and then cooling it on ice. Primer extension reactions to produce TNA (red) were performed for 1 h at 55 °C using 100 μM tNTPs, 500 nM primer−template complex, 1.25 mM MnCl2, and 0.1 U/μL Therminator™ DNA polymerase. Product analysis used denaturing polyacrylamide gel electrophoresis (PAGE).

Repetition of the above primer extension reaction was followed by strand separation using gel electrophoresis, as depicted below. Extraction of the gel afforded single-stranded DNA (black)-TNA (red) product to which a reverse complementary primer was annealed. SuperScript II™ (SSII)-mediated TNA reverse transcription (RT) was performed for 24 h at 42 °C and analyzed by denaturing PAGE. Mn2+is required to convert TNA into full-length cDNA.

Adapted with permission from Chaput and coworkers J. Am. Chem. Soc. 135, 3583−3591 (2013)

Fidelity of TNA Replication

Chaput and coworkers measured the fidelity of TNA replication by sequencing the cDNA product of the RT reaction after amplification by PCR. This fidelity assay measures the aggregate fidelity of a complete replication cycle (DNA → TNA → DNA), which is operationally different than the accuracy of a single-nucleotide incorporation event. The fidelity determined by this assay is the actual accuracy with which full-length TNA is synthesized and reverse transcribed. Therefore, it reflects the combined effects of nucleotide misincorporation, insertions and deletions (indel), and any mutations that occur during PCR amplification and cloning.

The researchers began by measuring the fidelity of TNA replication for a template (4NT.3G) used in the RT assay with SSII. While TNA replication on 4NT.3G resulted in an overall fidelity that was comparable with other XNA replication systems (~96%), detailed analysis of the mutation profile indicated that G → C transversions account for 90% of the genetic changes. While the precise molecular details of this transversion remain unknown, other results were said to suggest that base stacking plays an important role in the misincorporation of tGTP opposite dG in the template.

Functional XNA In Vivo

As mentioned in the introduction, Liu et al. have recently reported results that demonstrate the first-ever XNA to functionin vivo, thus paving the way for future developments in this intriguing unexplored area. The other novel aspect of their work is the use of a backboned-modified XNA, namely, 3′→2′phosphonomethylthreosyl nucleic acid (tPhoNA). This XNA can be seen below, and is quite different from DNA and TNA. I will now discuss both the in vitroand in vivostudies in this landmark publication.

Adapted with permission from Liu et al. J. Amer. Chem. Soc.140, 6690−6699 (2018)

The initial part of this work deals with the synthesis of the four “triphosphate”-like monomers [PMTApp etc. (PMTNpp) shown here] required for polymerase-mediated transcription reactions analogous to those described above for TNA. Preliminary experiments with Therminator™ led to inadequate transcription and thus to extensive site-directed engineering in order to find an alternative mutant polymerase with acceptable kinetics and processivity. Starting with TgoT, which is a Thermococcus gorgonarius DNA polymerase variant, researchers obtained a TgoT mutant (TgoT EPFLH), an efficient tPhoNA synthetase capable of synthesizing a 57-mer from a DNA template in less than 5 minutes.

Next, they screened the available reverse transcriptases (RTs) that could synthesize DNA against TNA templates and found two (mutants of TgoT and KOD DNA polymerase) that were efficient tPhoNA RTs capable of DNA synthesis from both DNA and RNA primers. Together, the TgoT EPFLH and mutant RTs tested enabled Liu et al. to transfer genetic information from DNA into tPhoNA and recover that information back to DNA with an approximate aggregate error rate of ~20 × 10−3per incorporation, which is a degree of fidelity compatible with the development of aptamers and aptazymes based on this new chemistry.

To evaluate the ability of tPhoNAs as templates for DNA synthesis in E. coli, six 18-mer 5′-phosphorylated DNA−PhoNA−DNA chimera oligonucleotides were chemically synthesized and tested using the established gapped-vector assay, based on the restoration of the active site of thymidylate synthase (thyA) depicted here.

Adapted with permission from Liu et al. J. Amer. Chem. Soc.140, 6690−6699 (2018)

This enzyme catalyzes the conversion of deoxyuridine monophosphate to thymidine monophosphate and is essential for E. coli growth in a medium lacking either thymine or thymidine. The six 18-mer 5′-phosphorylated DNA−PhoNA−DNA chimera oligonucleotides were ligated into a gapped heteroduplex plasmid, where 6 codons surrounding the catalytic Cys146 of the thyA gene had been deleted. The resulting ligation products were then transformed into an E. colistrain lacking thyA. Transformants are able to survive in thymidine-free media only when a PhoNA oligonucleotide chimera is recognized by the bacterial replication machinery and utilized as a template for DNA synthesis, restoring the thymidylate synthase active site. The ratio between bacterial colony numbers in media with or without thymidine indicates the extent of successful templating.

The replacement of a DNA unit by a tPhoNA building block yielded a 2.5-fold decrease in the number of prototrophic transformants compared to the positive controls. Further 2- and 6.5-fold drops in the yield of prototrophic transformants resulted from an extension of the tPhoNA stretch from one to two and three oligonucleotides, respectively. Thus, the successive addition of tPhoNA nucleotides significantly diminished DNA propagation in vivo.

Conclusions by Liu et al.

For in vivo applications, it is desirable that XNAs are both chemically and biologically orthogonal: neither the polymers nor the building blocks interact with natural nucleic acids or proteins, and XNA-synthesizing and -binding proteins do not synthesize or bind natural nucleic acids. The tPhoNAs showed significant levels of orthogonality at both chemical (oligonucleotide properties) and biological (recognition by natural protein) levels. The melting analyses (see Liu et al.) demonstrated that heavily modified tPhoNA did not exhibit detectable hybridization to complementary DNA or RNA, but did retain some potential to form homoduplexes, at least for AT-rich sequences.

tPhoNA also showed signs of biological orthogonality. The four PMTNpp’s were demonstrably poor substrates for natural polymerases, yet tPhoNA could be efficiently synthesized by the engineered polymerase TgoT_EPFLH. Importantly, as Liu et al. engineered TgoT for better tPhoNA synthesis, they observed a noticeable drop in its DNA synthetase function. Given the orthogonality demonstrated both chemically (oligonucleotide annealing) and in vivo(transliteration), Liu et al. suggested that it is likely that TgoT_EPFLH’s broadened substrate specificity can be further engineered to develop an orthogonal polymerase.

Liu et al. concluded that, collectively, the data further suggest that a fully orthogonal genetic system based on tPhoNA and specialist tPhoNA polymerases with minimal interaction with natural dNTPs, nucleic acids, or polymerases, is very much achievable.

Xenobiology: Quo Vadis?

Xenobiology (XB) has been defined as a subfield of synthetic biology involving the study of synthesizing and manipulating biological devices and systems using XNAs. In my opinion, XB is as an emerging branch of non-natural biology in search of utility. Consequently, the application of XB as the “ultimate biosafety tool,” as proposed by iGEM students, is quite intriguing. I should note here that iGEM, which stands for International Genetically Engineered Machine Competition, is intriguing in its own right, and well worth reading about later at this link.

The stated iGEM students’ premise for this ultimate biosafety tool is that the wide use of genetically modified organisms (GMOs) has caused serious concerns on how GMOs will interact with the natural environment. In particular, could a genetically modified microbe escape its constraints, and outcompete organisms found in the natural ecosystem? Since the early days of genetic engineering, biosafety strategies have been employed in order to control risks, and the advent of high-throughput synthetic biology is bringing these concerns to another level: the more we tinker with biology, the more our biosafety needs to be bullet-proof.

The stated aim of the iGEM students is to create a synthetic “man-made” version of biology, that respects the definition of life, but is based on entirely different mechanisms to function. The biochemistry of a xeno-organism uses new XNAs, genetic codes, and cofactors different from those explored in biology, and is therefore incompatible with other forms of life. This allows a much higher level of control: a xeno-organism will not be able to find the xenocompounds in the natural environment nor will it be able to use bacterial communication systems.

Only time will tell whether this technically challenging but valuable goal is achieved. I for one hope so, and, after writing and reflecting on this blog, will add that I’m amazed by the increasing breadth and depth of modified nucleic acid chemistry reflected in XNAs. It seems that the diversity of XNA chemical structures, which has been recently reviewedhere, can still be expanded further. Indeed, the findings of Liu et al. for tPhoNAs point the way to orthogonal genetic systems, and perhaps to the iGEM application mentioned here.

As usual, your comments are welcomed.