Chemistry & Biology

Review

The Emerging World of Synthetic Genetics

John C. Chaput,1,* Hanyang Yu,1 and Su Zhang1

1Center for Evolutionary Medicine and Informatics in the Biodesign Institute, and Department of Chemistry and Biochemistry,Arizona State University, Tempe, AZ 85287-5301, USA*Correspondence: john.chaput@asu.eduhttp://dx.doi.org/10.1016/j.chembiol.2012.10.011

For over 20 years, laboratories around the world have been applying the principles of Darwinian evolution toisolate DNA and RNA molecules with specific ligand-binding or catalytic activities. This area of syntheticbiology, commonly referred to as in vitro genetics, is made possible by the availability of natural polymerasesthat can replicate genetic information in the laboratory. Moving beyond natural nucleic acids requires organicchemistry to synthesize unnatural analogues and polymerase engineering to create enzymes that recognizeartificial substrates. Progress in both of these areas has led to the emerging field of synthetic genetics, whichexplores the structural and functional properties of synthetic genetic polymers by in vitro evolution. Thisreview examines recent advances in the Darwinian evolution of artificial genetic polymers and their potentialdownstream applications in exobiology, molecular medicine, and synthetic biology.

Merging Chemistry and BiologyIn a recent article marking the 40th anniversary of the first in vitro

evolution experiment, Gerald Joyce compared Sol Spiegelman’s

directed evolution of bacteriophage Qb RNA to Friedrich Woh-

ler’s synthesis of urea (Joyce, 2007). Joyce noted that, just as

Wohler’s synthesis of urea broke the false distinction between

biological chemistry and organic synthesis (Wohler, 1828),

Spiegelman’s experiments on Qb RNA broke a similar false

distinction between Darwinian evolution as a biological process

and Darwinian evolution as a chemical process (Mills et al.,

1967). This timely comparison reminds us of the seminal role

that both individuals played in the history of science and how

their pioneering work gave rise to the prodigious fields of organic

chemistry and in vitro genetics (Appella, 2010; Joyce, 2012).

Now, after many decades, these two disciplines are uniting to

form a new field of science called synthetic genetics that aims

to explore the structural and functional properties of synthetic

genetic polymers. Researchers working in this area are devel-

oping in vitro selection systems that allow unnatural genetic

polymers to evolve in response to imposed selection constraints

(Pinheiro et al., 2012; Yu et al., 2012). Previously, the only genetic

polymers capable of undergoing Darwinian evolution were

DNA and RNA, because these were the only molecules with

polymerases available that could replicate, transcribe, and

reverse-transcribe genetic information in the laboratory (Joyce,

2004;Wilson and Szostak, 1999). This paradigm is now changing

as a new generation of polymerases becomes available that can

copy sequence information back and forth between DNA and

XNA. Here, we use the term XNA, sometimes referred to as xen-

onucleic acids, to describe a general class of nucleic acid mole-

cules in which the natural ribose and 20-deoxyribose sugars

found in RNA and DNA have been replaced by an unnatural

moiety (X). While X usually refers to a genetic polymer with an

alternative sugar, nonsugar moieties are also possible, albeit

less common.

Herdewijn and Marliere (2009) introduced the term XNA in

a theoretical paper on the development of genetically altered

organisms that use alternative nucleic acid polymers with

backbones that do not interfere with normal DNA and RNA

1360 Chemistry & Biology 19, November 21, 2012 ª2012 Elsevier Ltd

biosynthesis. In this review article, we adhere to the traditional

view of XNA as synthetic genetic polymers with unnatural nucleic

acid backbones. We do, however, recognize that the last 2

decades have witnessed tremendous growth in all areas of

nucleic acid chemistry. To date, more than 100 different chemi-

cal modifications have been made to DNA and RNA, and

together, these modifications have improved our understanding

of nucleic acid structure and function. Notable accomplishments

include the following: (1) expansion of the genetic alphabet

beyond the four natural bases of A, C, T, and G (Henry and Ro-

mesberg, 2003; Kool, 2002b; Piccirilli et al., 1990); (2) formation

of Watson-Crick base pairs in the absence of complementary

hydrogen bond donor and acceptor groups (Krueger and Kool,

2007; Lavergne et al., 2012; Mitsui et al., 2003; Moran et al.,

1997); (3) formation of size and strand expanded DNA helices

(Chaput and Switzer, 1999; Kang et al., 2012; Liu et al., 2003);

(4) replication of modified bases using the polymerase chain

reaction (PCR) (Malyshev et al., 2009; Yang et al., 2010); (5)

development of in vitro selection strategies that support modi-

fied bases (Hollenstein et al., 2009a, 2009b; Vaish et al., 2000;

Vaught et al., 2010); and (6) replication and translation of

unnatural codons in living bacteria cells (Delaney et al., 2009;

Krueger et al., 2011). These and many other accomplishments

have profoundly impacted our knowledge of the chemical and

physical properties of DNA and RNA and led to the development

of new sensors, diagnostics, and therapeutic agents for

biotechnology and molecular medicine. Since a thorough

discussion of nucleic acid chemistry is beyond the scope of

this article, we direct readers interested in learning more

about modified nucleic acids and their recognition by DNA and

RNA polymerases to several excellent reviews on this topic

(Appella, 2009; Brudno and Liu, 2009; Kool, 2002a; Zhang

et al., 2010).

In this article, we examine recent advances in the in vitro

evolution of synthetic genetic polymers with novel functional

properties. These developments represent promising new strat-

egies for creating nuclease-resistant molecules that can be

made to function in a variety of natural and unnatural environ-

ments. We begin with a review of molecular evolution and the

All rights reserved

Chemistry & Biology

Review

transition from natural genetic systems composed of DNA and

RNA to synthetic genetic polymers with unnatural nucleic acid

backbones. Next, we discuss examples where in vitro evolution

has been used to isolate XNA molecules with discrete ligand-

binding properties. This discussion covers the complete spec-

trum of XNA polymers from early studies on molecules that

closely resemble natural genetic systems to more diverse back-

bone structures that have begun to explore new regions of

chemical space far beyond the local structural neighborhood

of DNA and RNA. Finally, we conclude with a discussion of

synthetic genetics and the potential impact that this new field

will have on the future of exobiology, molecular medicine, and

synthetic biology. It is our hope that this article will stimulate

others to think about new ways to explore the functional proper-

ties of synthetic genetic polymers and push the field of synthetic

genetics into mainstream molecular biology.

In Vitro EvolutionDarwinian evolution describes how a population changes

overtime to optimize the fitness of individual members for

a particular environment (Darwin, 1859). Evolution is driven by

natural selection, which allows certain inherited traits to become

more or less common as a result of differential survival. Individ-

ualswho adapt to a particular environment pass their genes on to

future generations, while members who are unable to do so go

extinct. Although evolution and natural selection are most often

discussed in terms of living organisms, these concepts can

also be applied to individual molecules (Joyce, 1994; Szostak,

1992). In this regard, genetic polymers are ideally suited for

this purpose, because they can fold into shapes with defined

functional properties (phenotypes) and their sequences (geno-

types) can be replicated in vitro to produce progeny molecules.

The ability to amplify individual molecules with desired pheno-

types and optimize their functions by directed evolution distin-

guishes genetic polymers, like DNA and RNA, from all other

organic molecules, which are incapable of replication because

they lack a genotype-phenotype connection.

To apply the principles of Darwinian evolution to nucleic acids,

a large population of nonidentical sequences is created and this

pool is assayed en masse for individual molecules that can fold

into shapes that can bind to a particular target or catalyze

a specific chemical reaction (Ellington and Szostak, 1990;

Robertson and Joyce, 1990; Tuerk and Gold, 1990). Molecules

with desired phenotypes are separated from the nonfunctional

pool, and their genotypes are amplified to generate a new

population of progeny molecules that has become enriched in

a particular trait. Amplification is usually done at the DNA level

using PCR (Saiki et al., 1985, 1988), but this process can also

be performed at the RNA level by isothermal RNA amplification

(Guatelli et al., 1990). Random mutations that occur during the

amplification process are critical to the selection outcome,

because these genetic changes introduce new diversity into

the sequence population. In general, mutations made to highly

conserved positions tend to have a negative impact on bio-

polymer function, while mutations made to permissive sites

either accumulate as a series of neutral mutations or improve

biopolymer function by correcting suboptimal contacts in the

tertiary structure or by forming new interactions to their substrate

or target ligand (Carothers et al., 2004, 2006). Of course, the real

Chemistry & Biology 1

power of in vitro evolution comes through iterative cycles of

selection and amplification, which makes it possible to search

vast regions of ‘‘sequence space’’ for individual molecules that

arose from one of 1015 different sequences present in the start-

ing population. This level of combinatorial power far surpasses

anything that can be accomplished using even the most sophis-

ticated high throughput screening methods, which is one reason

why in vitro evolution has emerged as a powerful tool in the

biotechnology arsenal.

While in vitro evolution has been used for more than 20 years

to isolate functional DNA and RNAmolecules from large pools of

random sequences (Ellington and Szostak, 1990; Robertson and

Joyce, 1990; Tuerk and Gold, 1990), applying the principles of

Darwinian evolution to artificial genetic polymers remains a chal-

lenging problem. Two major obstacles exist: (1) how to obtain

unnatural nucleic acid substrates that are not commercially

available; and (2) how to identify polymerases that can convert

genetic information back and forth between DNA and XNA.

The first problem can be overcome with organic chemistry;

however, this requires expertise in chemical synthesis or access

to collaborators with knowledge of organic chemistry. Indeed,

the chemical synthesis of modified nucleic acid triphosphates

remains a difficult and labor-intensive process. Nucleoside

triphosphates are not trivial to synthesize or purify, and many

analogs require 10 or more synthetic steps. While these chal-

lenges are not insurmountable, they do represent a significant

bottleneck in the discovery process and one that could slow

the pace at which new genetic polymers are examined. Presum-

ably, this problemwill become less severe over time as the prac-

tical applications of XNA technology motivate chemical and

biotechnology companies to construct these substrates as

commercial products. Until this happens, the focus will remain

on chemistry and the use of chemical synthesis to generate

new types of artificial genetic systems with a diverse repertoire

of chemical functionality.

The second problem is potentially more serious and involves

identifying polymerases that can either copy libraries back and

forth between DNA and XNA or copy XNA directly into XNA

without using DNA as an intermediate. Discovering enzymes

that are suitable for a given XNA system typically involves

screening commercial enzymes, but going forward polymerase

engineering likely will be more important. Unfortunately, natural

polymerases like T7 RNA polymerase and Taq DNA polymerase

rarely accept modified substrates with unnatural backbones.

Even nucleic acid substrates bearing subtle changes to the

ribose or deoxyribose sugar can inhibit these enzymes. This

stringent level of substrate specificity makes it difficult to identify

polymerases that will recognize modified substrates, either as

nucleoside triphosphates or as templates. Fortunately, many

variants of wild-type polymerases have been developed for

DNA sequencing and other biotechnology applications, and

these enzymes are often more permissive of unnatural sub-

strates. Determining which enzymes are best suited for a given

XNA polymer requires screening commercial enzymes in a

primer-extension assay where each polymerase is challenged

to extend a DNA primer annealed to a DNA template with XNA.

Since XNA tends to disrupt critical contacts in the enzyme-

substrate complex, most polymerases terminate primer exten-

sion after one or two nucleotide incorporations. However,

9, November 21, 2012 ª2012 Elsevier Ltd All rights reserved 1361

010200020991

First RNA

aptamers2’-Amino

Spiegelmers

2’-Methoxy

PNAmock

selection

MNA

2’-Fluoro &Phosphorothioate

Borono-phosphate

Macugen®

approvedFirst DNA

aptamers

First XNA aptamers(TNA & HNA)

Figure 1. Darwinian Evolution of Naturaland Artificial Genetic SystemsSince the discovery of nucleic acid aptamers in theearly 1990s, in vitro selection has been applied toincreasing numbers of RNA and DNA analogs. In2012, advances in nucleic acid chemistry andenzyme evolution made it possible to evolvefunctional XNA molecules in the laboratory. XNAmolecules are distinct genetic systems with alter-native (nonribose) nucleic acid backbones.

Chemistry & Biology

Review

some polymerases will synthesize longer XNA products and

these enzymes then become candidates for further optimization.

Optimization can proceed by identifying conditions that allow

the polymerase to function with enhanced efficiency or by

engineering the enzyme itself for improved activity. Both of

these approaches have been used successfully to identify

polymerases that will transcribe and reverse-transcribe XNA

polymers with high efficiency and fidelity (Betz et al., 2012;

Loakes et al., 2009; Pinheiro et al., 2012; Yu et al., 2012).

Certain engineered polymerases are even able to copy limited

stretches of XNA into XNA, suggesting that further polymerase

engineering could lead to DNA independent pathways for XNA

evolution.

Darwinian Evolution of Artificial Genetic SystemsIn vitro selection has allowed for the isolation of nucleic acid

sequences that can bind to a wide range of targets from small

molecules to whole cells. In common nucleic acid parlance,

these molecules are referred to as aptamers, which are single-

stranded sequences identified by in vitro evolution that mimic

antibodies by folding into shapes that can bind to a desired

target with high affinity and high specificity (Ellington and

Szostak, 1990). Aptamers represent a valuable class of affinity

reagents as ligand binding can alter or inhibit the biological prop-

erties of their target proteins (Famulok et al., 2007; Keefe et al.,

2010; Mayer, 2009; Nimjee et al., 2005). Unfortunately, genetic

polymers composed of natural DNA or RNA are poor candidates

for diagnostic and therapeutic applications as these molecules

are readily degraded by nucleases—enzymes present in biolog-

ical samples that cleave phosphodiester bonds. Introducing

chemical modifications into the backbone structure that stabilize

the molecule against nuclease degradation can enhance the

potential utility of aptamers as diagnostic and therapeutic

agents. In particular, substitution of the 20 hydroxyl position of

ribonucleotides with amino (NH2), flouro (F), and methoxy

(OCH3) groups confers resistance to nucleases that utilize the

20 hydroxyl group for cleavage of the phosphodiester bond.

Consequently, the first in vitro selection experiments focused

on subtle chemical modifications that were tolerated by natural

polymerases and rendered the aptamer more resistant to

nuclease degradation (Keefe and Cload, 2008). As increasing

numbers of mutant polymerases became available, the field

gradually shifted toward nucleic acid polymers withmore diverse

chemical structures (Figure 1). In the following sections, we

divide our examples into two sections: XNA polymers with back-

bone structures related to RNA and XNA polymers with novel

backbone structures. This distinction is designed to highlight

the progress of different artificial genetic systems and their

contribution to the field of synthetic genetics.

1362 Chemistry & Biology 19, November 21, 2012 ª2012 Elsevier Ltd

XNA Polymers Related to RNA20 Modified RNA

In 1994, Jayasena and coworkers reported the first example

where Darwinian evolution was applied to pools of modified

genetic polymers (Lin et al., 1994). Using recombinant T7

RNA polymerase, a library of RNA molecules bearing 20-amino-

modified pyrimidines in place of the natural 20 OH group and all

natural purine nucleotides was constructed by in vitro transcrip-

tion (Figure 2). Cell-free transcription gave themodified RNAwith

reduced yield (�50%) relative to natural RNA. This library was

used to isolate modified aptamers that bound to human neutro-

phil elastase (HNE) by performing iterative rounds of in vitro

selection and amplification against HNE molecules immobilized

on nitrocellulose filters. HNE is implicated in a number of respira-

tory diseases, including pulmonary emphysema, chronic bron-

chitis, and cystic fibrosis, and as such represents an interesting

therapeutic target. After 15 rounds of in vitro selection and ampli-

fication, amino-modified aptamers were identified that bound to

HNEwith low nanomolar affinity and showed only weak affinity to

porcine pancreatic elastase—a closely related protease. Com-

parative binding assays performed using DNA and RNA mole-

cules with the same nucleotide sequence showed that the

aptamer required the amino modification to achieve high affinity

and high specificity binding. The authors discovered that the

amino-modified aptamer has a half-life of �20 hr in human

serum, which is considerably longer than the serum half-life of

RNA (seconds to minutes, depending on sequence and struc-

ture). While the success of the HNE aptamer provided a clear

demonstration that modified RNA aptamers could be created

by direct selection methods (as opposed to postselection modi-

fication), the amino modification was quickly discarded due to

problems associated with the solid-phase synthesis of amino-

modified RNA and the unpredictable nature of the 20 amino

group (pKa 6.2) at physiological pH.

Recognizing the limitations of 20 amino-modified RNA, other

functional groups were explored as substrates for T7 RNA poly-

merase. Of the possible substitutions, 20 fluoro RNA emerged as

a popular RNA analog (Figure 2). Relative to the 20 amino group,

20 fluoro leads to increased coupling efficiency during solid-

phase synthesis and avoids the need for an additional protection

step when constructed as a phosphoramidite monomer. In addi-

tion, model studies on sequence-defined synthetic oligonucleo-

tides indicate that the 20 fluoro modification enhances the

thermal stability of a duplex (Lesnik et al., 1993), while the 20

amino modification destabilizes the helical structure (Aurup

et al., 1994). The best early example of a 20 fluoro aptamer is

Macugen (pegaptinib sodium), the first therapeutic aptamer

clinically approved by the Food and Drug Administration. Macu-

gen, which is used to treat age-related macular degeneration,

All rights reserved

OH

OB

O

OPO

O

O

OB

O

O P O

O

H

R

OB

O

OPO

O

R = F, NH2, OCH3

OB

O

OPO

SO

B

O

OPO

BH3

D-RNA L-RNA

Phosphorothioate Boranophosphate2'-modified RNA

(natural) (unnatural)

OB

O

OPO

O

DNA

Figure 2. Examples of Nucleic Acid Analogsthat Closely Resemble DNA and RNANucleic acids with subtle modifications to thephosphodiester backbone are tolerated to varyingdegrees by natural DNA and RNA polymerases.Each of these examples has been used success-fully as an unnatural genetic polymer for in vitroselection.

Chemistry & Biology

Review

functions as a drug by inhibiting the binding of vascular endothe-

lial growth factor (VEGF)-165 to its target receptor. In clinical

trials, 80% of the patients treated with this aptamer show

stable or improved vision 3 months after treatment (Eyetech

Study Group, 2003). Macugen was isolated from a random

sequence RNA library bearing 20-fluoro modified pyrimidines

and natural purine nucleotides (20 OH) (Ruckman et al., 1998).

After the selection, the 28-nucleotide RNA aptamer was

further modified by replacing the 20 OH group at most of

the purine positions with 20-methoxy groups (OCH3) and

appending polyethylene glycol moieties onto the structure

(Ruckman et al., 1998).

The simplicity of the 20 fluoro modification coupled with its

recognition by T7 RNA polymerase have helped make 20 fluoroRNA a widely used derivative for in vitro selection. More recent

examples of 20-fluoro substituted RNA aptamers include work

by the Rossi and Sullenger labs. In an exciting new development,

Rossi and coworkers evolved 20 fluoro aptamers that bind to HIV

glycoprotein 120 (gp120) with nanomolar affinity (Zhou et al.,

2009). In this study, a commercial T7 RNA polymerase mutant

called DuraScribe T7 RNA polymerase was used due to its

improved ability to incorporate 20-F CTP and 20-F UTP substrates

as well as ATP and GTP. Remarkably, the aptamer is internalized

into HIV-infected cells expressing the HIV envelope protein, sug-

gesting that this aptamer could be used as a drug delivery

system. To test this possibility, the anti-gp120 aptamer was

conjugated with anti-HIV siRNA to form a chimeric aptamer-

siRNA molecule that targets HIV-infected cells and leads to

inhibition of HIV replication and infectivity. In another exciting

development (Mi et al., 2010), the Sullenger lab designed an

in vivo selection method to target tumor-specific moieties. In

this system, an in vivo selection was applied to an animal model

of intrahepatic colorectal cancer metastases by injecting a

nuclease-resistant library of 20 fluoro pyrimidine-modified RNA

into mice bearing previously implanted hepatic tumors. Fol-

lowing an incubation period, the livers were isolated and RNA

molecules from the library were recovered and amplified by

RT-PCR. After multiple cycles of selection and amplification,

an RNA motif emerged that bound to p68 RNA helicase with

Chemistry & Biology 19, November 21, 2012 ª

high affinity. This aptamer colocalizes

with p68 and inhibits p68 induced

ATPase activity in tumor cells.

Another promising 20 RNA substitution

often used for the production of thera-

peutic aptamers is 20 methoxy RNA

(OCH3). The 20 methoxy substitution

(Figure 2) is a naturally occurring RNA

analog found in ribosomal RNA that arises

via the posttranscriptional machinery

inside the cell. While nature has found a way to make this modi-

fication, the incorporation of 20 methoxy residues into growing

RNA strands by in vitro transcription requires mutant versions

of T7 RNA polymerase. Fortunately, several variants of this

enzyme have now been identified that recognize 20-O-methyl

NTPs (Chelliserrykattil and Ellington, 2004; Fa et al., 2004). Using

a modified T7 RNA polymerase bearing the mutations K378R,

Y639F, and H784A, Burmeister et al., reported the first example

of a fully 20-O-methyl-substituted aptamer isolated by in vitro

selection (Burmeister et al., 2005). This unnatural nucleic acid

aptamer binds to VEFG165 with low nanomolar affinity and shows

superior stability in blood with a clearance half-life of 23 hr in

mice models. The same authors used a similar strategy to select

20-O-methyl substituted pyrimidine aptamers that bind to human

thrombin and human interleukin (IL)-23 (Burmeister et al., 2006).

These aptamers also showed high nuclease resistance in model

assays.

Modified Phosphodiester Linkages

Phosphorothioate (PS) oligonucleotides differ from natural oligo-

nucleotides (DNA and RNA) by substitution of a single nonbridg-

ing oxygen atom in the phosphodiester linkage for a sulfur atom

(Figure 2). The markedly lower electronegativity and greater

polarizability of the sulfur atom provides a useful probe for

mechanistic studies involving reactions at the phosphodiester

linkage (Li et al., 2011). This substitution also renders PS oligonu-

cleotides less sensitive to nuclease degradation (Xu and Kool,

1998); hence, this substitution is commonly found in antisense

technology (Dias and Stein, 2002). Nucleoside and deoxynucleo-

side triphosphates bearing sulfur at the a-phosphate position are

recognized by a number of DNA andRNApolymerases, although

their incorporation efficiency varies depending on the enzyme

and the nature of nucleic acid polymer. In a very early example,

Ellington and colleagues showed that fully substituted PS RNA

can be generated by in vitro transcription using standard T7

RNA polymerase and all four (aS) NTPs (Jhaveri et al., 1998).

Enzymatic incorporation of (aS) dNTPs by DNA polymerases

proceeds less efficiently and only oligonucleotides with partial

substitution are possible using standard DNA polymerases.

Recognizing this problem, Holliger and colleagues developed

2012 Elsevier Ltd All rights reserved 1363

Chemistry & Biology

Review

mutant taq DNA polymerases that can support the PCR amplifi-

cation of fully substituted phosphorothioate DNA polymers up to

2 kb in length (Ghadessy et al., 2004). While thioaptamers have

received broad interest, this backbone substitution is less

common than substitutions made to the 20 ribo-position. One

reason for this could be that PS oligonucleotides are stickymole-

cules that can recognize off-target proteins with nonspecific

binding interactions. Nevertheless, in vitro evolution has pro-

duced thioaptmaers with affinity to a number of proteins,

including molecules that bind to human basic fibroblast growth

factor (Jhaveri et al., 1998), nuclear factor for human IL6 (King

et al., 1998), the RNase H domain of HIV RT (Somasunderam

et al., 2005), the Venezuelan equine encephalitis virus capsid

protein (Kang et al., 2007), and transforming growth factor-b1

(Kang et al., 2008). The thioaptamer selected to bind the RNase

H domain of HIV RT is interesting in that it inhibited RNase H

activity in a dose dependent manner across a broad range of

virus inoculum. At an m.o.i. of < 0.005, viral suppression was

comparable to the HIV drug AZT (Somasunderam et al., 2005).

Another interesting substitution that has been examined by

in vitro evolution is boronophosphate RNA, or bRNA (Figure 2).

Similar to PS RNA, bRNA is constructed using synthetic chem-

istry that replaces the nonbridging oxygen atom in the phospho-

diester linkage with BH3. This substitution, which is isosteric and

isoelectric with respect to the natural phosphodiester linkage

leads to improved nuclease stability (Hall et al., 2004, 2006). In

the context of a DNA backbone, borano-DNA supports RNase

H activity, suggesting a possible role for this modification in

antisense technology (Li et al., 2007). Nucleotide triphosphates

bearing the (aB) NTP substitution are accepted as substrates

by T7 RNA polymerase. bRNA represents an interesting class

of therapeutic aptamers that could find practical application in

an experimental therapy called boron neutron capture therapy

(BNCT) (Hawthorne, 1993). In BNCT, a patient is first injected

with a boron compound that localizes to a cancerous tumor.

The patient is then exposed to neutron irradiation, which causes

the boron compound to emit alpha particles that destroy nearby

cancer cells. In theory, borano-aptamers could function as smart

therapeutics by delivering the boron atom to the cancer cell,

while avoiding healthy cells. In a proof-of-principle demonstra-

tion, Burke and coworkers evolved a series of borano-aptamers

that bound to the small molecule target, adenosine 50-triphos-phate (ATP) (Lato et al., 2002). ATP is a common target for new

in vitro selection technologies as the ATP-binding RNA aptamer

has a conserved motif that is easy to rediscover by in vitro

selection (Sassanfar and Szostak, 1993). In this case, partially

substituted bRNAwas constructed by in vitro transcription using

T7 RNA polymerase and NTP mixtures containing either (aB)

UTP or (aB) GTP. In addition to identifying the conserved

ATP-binding RNA motif, sometimes referred to as the Sassanfar

aptamer, named after its discoverer Mandana Sassanfar, the

authors found several aptamers whose ligand-binding activity

required the borano modification. This example suggests that

it should be possible to evolve borano-aptamers to other targets

of biological interest.

Mirror Image RNA

Mirror image design provides a clever strategy for generating

stable aptamers with high affinity and specificity for a given

biological molecule. This approach is based on the principle of

1364 Chemistry & Biology 19, November 21, 2012 ª2012 Elsevier Ltd

reciprocal chiral substrate specificity, which states that polymers

of one stereoconfiguration will exhibit reciprocal chiral substrate

specificity when constructed in the opposite stereoconfigura-

tion. In simple terms, this means that when the L-RNA version

of an RNA aptamer is synthesized, the new L-RNA aptamer will

bind the mirror image of the target (Figure 2). Thus, using mirror

image symmetry, L-RNA aptamers can be generated by carrying

out in vitro selection experiments against the mirror image of

natural biological targets. Furste and colleagues first demon-

strated this principle by identifying RNA sequences that could

bind to L-adenosine and D-arginine (the enantiomers of the

natural D-adenosine and L-arginine metabolites) (Klussmann

et al., 1996; Nolte et al., 1996). Nucleic acid sequences present

in the selection output were constructed by solid-phase

synthesis as both L- and D-RNA. The sequences constructed

of L-RNAwere given the name ‘‘spiegelmers’’—from theGerman

word ‘‘spiegel,’’ meaning mirror—to distinguish them from the

traditional D-RNA aptamers. Functional analysis studies re-

vealed that spiegelmers and aptamers with the same sequence

have identical binding affinity but reciprocal binding specificity.

In this case, spiegelmers recognized the natural metabolites to

the near exclusion of the unnatural enantiomers, while the

aptamers exhibited the opposite binding pattern and bound

the unnatural D-adenosine and L-arginine metabolites. Because

L-RNA is an unnatural nucleic acid backbone, the evolved

spiegelmers remain stable in human serum after 60 hr, while

the natural RNA aptamers rapidly degrade.

The ability for spiegelmers to withstand nuclease degradation

suggests that these molecules could have potential therapeutic

value. The first biological assay involving a spiegelmer was re-

ported in 1997 by Kim and Bartel (Williams et al., 1997). Here,

a spiegelmer to the peptide hormone L-vasopressin was pro-

duced using the unnatural D-peptide analog as bait. After several

rounds of in vitro selection and amplification followed by directed

evolution, a spiegelmer was generated that binds to vaso-

pressin. This molecule was shown to antagonize the vasopressin

response in cultured kidney cells. Similar studies have since

been performed on other pharmacological targets, including

the peptide hormone gonadotropin-releasing hormone I

(GnRH) (Leva et al., 2002). Spiegelmers generated to GnRH

were shown to inhibit GnRH binding to its cognate receptor in

Chinese hamster ovary cells with half maximal inhibitory concen-

tration (IC50) values of 50–200 nM. While the concept of mirror

image design provides a convenient strategy for generating

nuclease-resistant ligands, the requirement for mirror image

targets limits this approach to small molecules and short

proteins that are accessible by chemical synthesis. Recognizing

this problem, Klussmann and colleagues developed a domain-

based approach in which the peptide segment of a surface-

accessible domain was used as a target in place of the whole

protein (Purschke et al., 2003). In this case, the spiegelmer,

which was selected to bind a 25-amino-acid segment of

staphylococcal Enterotoxin B, exhibited nanomolar affinity to

the full-length protein.

Joyce and colleagues recently extended the concept of mirror

image symmetry to include catalytic RNA (Olea et al., 2012). In

this case, a self-replicating all L-RNA enzyme and its all L-RNA

substrate were prepared by solid-phase synthesis from L-nucle-

oside phosphoramidites. The whole system was based on

All rights reserved

O

OBOPO

OO BOPO

O

O

NO

O

BHN

Hexitol Nucleic Acid(HNA)

Threose Nucleic Acid(TNA)

Peptide Nucleic Acid(PNA)

Figure 3. XNA Polymers with Diverse Nucleic Acid BackboneStructuresHNA and TNA are XNA systems in which the natural ribose sugar found in RNAhas been replaced with a nonribose sugar moiety. HNA and TNA requireengineered polymerases for their replication. PNA is a nucleic acid systembased on repeating amide bonds that replicates by nonenzymatic template-directed polymerization.

Chemistry & Biology

Review

a previously discovered RNA ribozyme that had been developed

for ligand sensing but was limited in its application due to its

susceptibility to ribonuclease degradation. The L-ribozyme

was shown to undergo isothermal, theophylline-dependent

exponential amplification, in a manner analogous to the natural

RNA ribozyme system. However, unlike the RNA ribozyme, the

L-RNA system was impervious to nuclease degradation. This

demonstration suggests that it should be possible to create

other nuclease-resistant sensors with aptamer recognition

domains that function under a wide range of natural conditions.

Mosaic Nucleic Acids

Motivated by a desire to understand chemical steps that led to

the emergence of early genetic systems, Szostak and coworkers

have demonstrated that nucleic acid molecules with heteroge-

neous backbones can evolve ligand-binding functions despite

the repeated shuffling of backbone chemistry between iterative

cycles of selection (Trevino et al., 2011). In a simplified model

system, a mutant T7 RNA polymerase (Y639F) capable of

transcribing nucleic acids that contain both deoxy- and ribonu-

cleotides was used to construct a library of oligonucleotides

that contained equal amounts of DNA and RNA residues. These

DNA-RNA chimeric polymers were given the name mosaic

nucleic acids (MNA). The library was used to isolate nucleo-

tide-binding MNA aptamers by in vitro selection. After each

round of selection, functional sequences were reverse-

transcribed into DNA and amplified by RT-PCR. The molecules

were then forward-transcribed back into MNA, which retained

the nucleotide sequence but randomized the sugar-phosphate

backbone. As a consequence, sequences that depended on

specific sugar arrangements were selected against, while

sequences that retained function despite variation at the sugar

positions were enriched. Following multiple rounds of selection

and amplification, MNA aptamers were identified from two

parallel selections that targeted ATP and GTP, respectively.

The selected aptamers bound specifically to their cognate

ligands, demonstrating that nonheritable backbones may not

have posed an insurmountable obstacle to the emergence of

early functional molecules. Whether similar selections could

yield MNA molecules with catalytic activity is an interesting

question that has yet to be determined.

XNA Polymers with Novel Backbone StructuresPeptide Nucleic Acid

One of the most interesting molecules developed as a nucleic

acid analog is peptide nucleic acid (PNA; Figure 3). Unlike all

DNA and RNA analogs described thus far, which have a sugar-

phosphate backbone, PNA has a backbone composed of

repeating N-(2-aminoethyl)-glycine units that are linked by

peptide bonds (Nielsen et al., 1991a). Since PNA has an

uncharged backbone, hybridization with DNA is stronger than

DNA/DNA hybridization due to the absence of electrostatic

repulsion. PNA is also more specific for a given target sequence

than DNA or RNA, as base pair mismatches are more destabiliz-

ing for PNA than a natural oligonucleotide of the same sequence

(Nielsen, 1995). These properties, coupled with enhanced

nuclease stability, make PNA a useful molecule for antisense

therapy (Demidov et al., 1994). Although PNA cannot easily cross

the cellularmembrane, conjugation of PNA antisense oligonucle-

otides to cell-penetrating peptides has been used successfully

Chemistry & Biology 1

to improve cellular delivery (Shiraishi and Nielsen, 2011; Wittung

et al., 1995).

Recognizing the potential that synthetic polymers could have

in expanding the functional utility of genetic polymers, Liu and

colleagues recently described an in vitro translation, selection,

and amplification system for PNA (Brudno et al., 2010). Since

PNA is not a substrate for any known polymerase, conditions

were developed that allowed for the nonenzymatic sequence-

specific oligomerization of tetramer and pentamer PNA building

blocks on DNA templates. These building blocks collectively

represent a ‘‘genetic code’’’ for PNA that can be faithfully ‘‘trans-

lated’’ on a DNA template using reductive amination chemistry to

link multiple PNA units together. For example, 10 consecutive

coupling reactions of pentamer codons lead to the synthesis of

a 50-mer PNA molecule containing secondary amine linkages

between every fifth nucleotide. Translation on a library of self-

priming DNA templates produced a pool of PNA-DNA hairpins

in which one strand was composed of PNA and the other strand

was composed of DNA. As illustrated in Figure 4, the DNA

portion of these molecules was then made double-stranded by

extending a DNA primer annealed to the stem-loop region of

the hairpin with dNTPs. This difficult step required Herculase II,

a strong thermophilic DNA polymerase to copy the DNA strand

by displacing the PNA strand. The resulting PNA-dsDNA mole-

cules contained the genotype-phenotype linkage necessary to

perform in vitro selection since the PNA molecules could be

selected on the basis of their chemical function and their encod-

ing DNA information could be amplified by their attached DNA

sequence. The authors validated this approach in a proof-of-

principle demonstration by performing six iterative cycles of

translation, selection, and amplification for PNA molecules that

could incorporate a biotinylated codon into their sequence.

The selection showed an overall enrichment of >106-fold of

PNA-encoding DNA templates on streptavidin-coated beads.

This example suggests that it should be possible to isolate

functional PNA aptamers by in vitro selection.

Threose Nucleic Acid

Among the many different alternative nucleic acid molecules

developed by Eschenmoser and colleagues, a-L-threose nucleic

acid (TNA) has emerged as the first genetic polymer with rele-

vance in synthetic genetics (Figure 3) (Schoning et al., 2000).

TNA is an artificial nucleic acid polymer in which the natural

five-carbon ribose sugar found in RNA has been replaced with

an unnatural four-carbon threose sugar and phosphodiester

9, November 21, 2012 ª2012 Elsevier Ltd All rights reserved 1365

XNA Synthesis

Display and Selection

PCR Amplification

Library Recovery

A

XNA Synthesis

Selection

Reverse transcription

Library Recovery

B

Figure 4. In Vitro Selection Strategies for XNA Evolution(A) DNA display provides a general strategy for evolving XNA polymers when an RT is not available to convert XNA sequences back into DNA for amplification byPCR. In this technique, a genotype-phenotype link is established by extending a self-priming DNA library with XNA to link each XNAmolecule to its encoding DNAsequence.(B) When an XNA transcriptase and RT are available, in vitro selection can be performed using a traditional two-enzyme replication strategy.

Chemistry & Biology

Review

linkages are connected at the 20 and 30 vicinal positions. Thissubstitution shortens the backbone repeat unit by one bond as

compared to natural DNA and RNA. Eschenmoser found that

TNA undergoes informational Watson-Crick base pairing in an

antiparallel strand orientation and also cross-pairs opposite

complementary strands of DNA and RNA (Schoning et al.,

2000). This latter feature is remarkable given the backbone

difference between TNA and the natural genetic polymers of

DNA and RNA.

The ability for TNA to exchange genetic information with RNA,

coupled with the chemical simplicity of threose relative to ribose,

have prompted careful consideration of TNA as a possible RNA

progenitor (Orgel, 2000). To explore this possibility, we and

others have attempted to identify polymerases that could be

used to examine the functional properties of TNA by in vitro

selection. Numerous primer-extension assays were performed

to identify DNA polymerases that could copy DNA templates

into TNA (Chaput and Szostak, 2003; Kempeneers et al., 2003)

and other polymerases that could copy TNA back into DNA

(Chaput et al., 2003). By identifying enzymes that could perform

these functions, we aimed to develop a replication system for

TNA that allowed pools of TNA molecules to be reverse-tran-

scribed into DNA, amplified by PCR, and forward-transcribed

back into TNA. From this body of work, it was discovered that

therminator DNA polymerase, an engineered variant of the 9� NDNA polymerase bearing the A485L mutation functions as an

efficient DNA-dependent TNA polymerase (Horhota et al.,

2005; Ichida et al., 2005). Under optimal conditions, therminator

was found to copy a long DNA template into TNA with high effi-

ciency and fidelity but failed to copy a pool of DNA sequences

into TNA (Ichida et al., 2005). We subsequently determined

that stretches of G-nucleotides in the DNA template sequence

lead to polymerase pausing and chain termination. By designing

DNA libraries that either eliminated all G-residues in the template

or presented G nucleotides at sufficiently low frequency so they

would occur as isolated residues, we were able to efficiently

transcribe DNA libraries into TNA (Yu et al., 2012). Consequently,

1366 Chemistry & Biology 19, November 21, 2012 ª2012 Elsevier Ltd

the resulting TNA sequences were either devoid of cytidine or

contained cytidine at reduced frequency.

To determine whether TNA could fold into structures with

discrete ligand-binding properties, a DNA display strategy

similar to the hairpin strategy developed by Liu and colleagues

was used to evolve TNA aptamers with affinity to human

thrombin (Figure 4) (Brudno et al., 2010). This approach provides

a solution to the problem of how to evolve XNA polymers when

an RT is not available to convert the pool of XNA molecules

back into DNA for amplification by PCR. Accordingly, a pool of

self-priming DNA templates was extended with TNA using ther-

minator DNA polymerase to copy the DNA library into TNA. The

DNA region of the TNA-DNA chimeric hairpins was made

double-stranded by extending a separate DNA primer annealed

to the step-loop structure with dNTPs. The resulting TNA-dsDNA

molecules produced the genotype-phenotype linkage necessary

to perform iterative rounds of in vitro selection and amplification

since TNA molecules could be selected on the basis of function

and amplified on the basis of their attached DNA sequence.

After several rounds of in vitro selection and amplification, TNA

aptamers emerged with high binding affinity and specificity to

human thrombin. The best aptamer identified in the selection

bound to thrombin with an equilibrium dissociation constant of

200 nM, which is similar to DNA and RNA aptamers previously

evolved to bind human thrombin. The isolation of TNA aptamers

from a pool of random TNAmolecules demonstrated, for the first

time, that the problem of ligand binding is not unique to the

natural polymers of RNA and DNA or close structural analogs

thereof (Yu et al., 2012). Indeed, one could imagine that purely

chemical constraints, like a shorter backbone repeat unit, might

preclude the ability for TNA to fold into structures with a desired

function. Since TNA does not appear to be limited in this regard,

it seems reasonable that similar selections for catalytic function

could be used to isolate novel TNA enzymes or ‘‘threozymes’’ by

in vitro evolution. As these studies progress, it will be interesting

to see how the functional properties of TNA compare to other

XNA molecules.

All rights reserved

OB

O

OPO

O

R=F, OH, etc O

OBO

BOPO

O

O

OB

O

OPO

OR

O

PO

O

BOPO

O

OOH

O

PO OB

OH

Cyclohexene Nucleic Acid Altritol Nucleic Acid Glycerol Nucleic Acid

Locked Nucleic Acid Apio Nucleic Acid(Fluoro) Arabino Nucleic Acid

Figure 5. Promising XNA Systems with Potential Application in Synthetic GeneticsThe future of synthetic genetics will likely include a wide range of alternative genetic systems that are each capable of undergoing Darwinian evolution. Theseexamples highlight a few promising candidates for future XNA polymers.

Chemistry & Biology

Review

Hexitol Nucleic Acid

Hexitol nucleic acid (HNA) was the second XNA polymer to be

studied in the context of synthetic genetics (Figure 3) (Pinheiro

et al., 2012). Developed by Herdewijn and colleagues as an anti-

sense reagent (Verheggen et al., 1993, 1995), HNA has a back-

bone structure composed of 10,50-anhydrohexitol, which is

a six-membered pyranosyl ring structure. The nucleobase is

located at the 20 position and phosphodiester linkages occur

between the 40 and 60 carbon atoms. Unlike most other RNA

analogs with six-membered carbohydrate moieties (Herdewijn,

2010), such as homo-DNA and pyranosyl-RNA (Krishnamurthy

et al., 1996), HNA is capable of base pairing opposite comple-

mentary strands of itself and RNA and DNA (Hendrix et al.,

1997a, 1997b). The rigid anhydrohexitol ring adopts a chair

conformation that causes HNA to form an A-type helical struc-

ture (Declercq et al., 2002; Lescrinier et al., 2000; Maier et al.,

2005). Like PNA, HNA is also more specific for a given target

sequence than DNA or RNA and HNA is able to distinguish

base pair mismatches more easily than a natural oligonucleotide

of the same sequence (Hendrix et al., 1997b).

To explore the functional properties of HNA, Holliger and

coworkers used a compartmentalized self-tagging strategy to

evolve DNA polymerases capable of copying DNA templates

into HNA (Pinheiro et al., 2012). The authors constructed a

library of polymerase variants based on the DNA enzyme

Thermococcus gorgonarius (Tgo) polymerase by randomly

incorporating genetic mutations into the parent gene. The library

was then used to identify mutant polymerases capable of

synthesizing HNA polymers on DNA templates. The selection

produced several enzymes that exhibited broad substrate

specificity, including polymerases that could transcribe HNA,

cyclohexenyl nucleic acids, locked nucleic acids (Campbell

and Wengel, 2011), TNA, arabinose nucleic acid (Noronha

et al., 2000) and 20-fluoro arabinose nucleic acid (Wilds and

Chemistry & Biology 1

Damha, 2000). To reverse-transcribe the XNA polymers back

into DNA, the authors evolved several novel RTs that could

copy HNA and other XNA polymers back into DNA. Their starting

library utilized statistical correlation analysis to identify an

allosteric network responsible for template recognition and

RNA-RT activity in the polB family of DNA polymerases. Neigh-

boring residues in the vicinity of possible hits were mutated

and screened in a polymerase activity assay. This approach

identified a new TgoTmutant, called RT521, that was a proficient

HNA RT. Together, the two enzymes allowed for the in vitro

evolution of HNA aptamers (Figure 4). HNA aptamers were

generated against HIV trans-activating response RNA and the

protein hen egg lysozyme. Both aptamers were found to bind

their targets with high affinity and specificity, consistent with

the interpretation that HNA, like TNA, can fold into shapes with

very specific ligand binding sites (Yu et al., 2012). Perhaps

even more importantly, however, the Holliger study provided

a general approach for creating XNA polymerases with broad

application in synthetic genetics.

Moving Beyond DNA and RNA: Implications for

Exobiology, Molecular Medicine, and Synthetic Biology

Armed with a new generation of polymerases and a large collec-

tion of nucleic acid modifications, the field of synthetic genetics

plans to expand the frontier of in vitro genetics to include

a growing list of structurally diverse XNA polymers (Figure 5).

Motivation for this work comes, in part, from a strong desire to

understand basic physical properties that may have contributed

to nature’s choice of ribofuranosyl nucleic acids (DNA and RNA)

as life’s genetic material (Eschenmoser, 1999). Since it is impos-

sible to rewind the evolutionary clock of time, experiments of this

type provide an important model for evaluating the fitness of

a given genetic polymer for a particular chemical function.

Although all forms of life (including those that are known to

have existed but are now extinct) are based on organisms that

9, November 21, 2012 ª2012 Elsevier Ltd All rights reserved 1367

Chemistry & Biology

Review

stored genetic information in DNA genomes and catalyzed

reactions with protein enzymes, earlier forms of life may have

relied on one biopolymer, not DNA or protein per se, but maybe

something else (Joyce, 2002). One possibility is that life evolved

from simple organisms that were based on RNA (Crick, 1968;

Orgel, 1968; Woese, 1967). This postulate, commonly referred

to as the RNA world hypothesis (Gilbert, 1986), received wide-

spread popularity following the discovery that RNA catalysts

exist in nature (Kruger et al., 1982). Since then,many other exam-

ples of RNA enzymes, or ribozymes, have been discovered and

others with functions more relevant to the RNA world have been

created by in vitro selection (Joyce, 2004; Wilson and Szostak,

1999). Despite these observations, it is difficult to envision how

a molecule as complicated as RNA could have emerged on the

early Earth and remained in the environment long enough to

replicate and produce offspring. This concern has led some to

suggest that RNA was not the first genetic material, but rather

an important evolutionary intermediate in the path to extant life

(Engelhart and Hud, 2010; Joyce et al., 1987).

If RNA was not the first genetic material, then what genetic

polymers came before RNA? Whatever prebiotic chemistry

gave rise to RNA would have almost certainly produced other

RNA analogs, some of which could have preceded or competed

directly with RNA. This hypothetical period in evolutionary history

is referred to as the pre-RNAworld—a timewhen chemistry gave

rise to simple self-replicating polymers that stored genetic infor-

mation and catalyzed chemical reactions (Lazcano and Miller,

1996). What these structures looked like and how they were

produced is a mystery that has challenged the minds of many

great chemists. Suggestions include PNAs (Nielsen et al.,

1991b), a class of RNA molecules with a protein-like backbone;

TNA (Schoning et al., 2000), an unnatural genetic system with

a four-carbon threose sugar in place of the natural five-carbon

ribose sugar; and glycerol nucleic acid (GNA) (Zhang et al.,

2005), an acyclic derivative of TNA. These alternative genetic

systems are interesting because they are: (1) physically realistic,

meaning that they can be constructed by chemical synthesis; (2)

capable of storing genetic information via Watson-Crick base

pairing; and (3) chemically simpler than RNA,meaning that prebi-

otic chemistry could, in theory, have favored their synthesis over

RNA. With the advent of synthetic genetics, it should be possible

to evaluate the functional properties for at least a subset of

synthetic genetic polymers. Since some of these molecules are

also strong candidates for early RNA progenitors, these studies

provide abasis for comparingmolecular functions across a range

of biopolymer systems. As these studies progress, an important

goal will be to develop XNA enzymes that can catalyze their own

synthesis. If successful, these self-replicating molecules could

be used to develop synthetic life forms with unnatural genomes.

In addition to addressing basic questions related to life’s first

genetic system, synthetic genetics also provides an opportunity

to develop nuclease-resistant molecules for biotechnology and

molecular medicine. Nucleic acid molecules with specific target

recognition and catalytic properties have become valuable tools

in many areas of basic and applied research (Krishnan and

Simmel, 2011; Mascini et al., 2012; Tombelli and Mascini,

2009). Unfortunately, most of the molecules created thus far

are composed of DNA and RNA, which are susceptible to

nuclease degradation and therefore not suitable for most biolog-

1368 Chemistry & Biology 19, November 21, 2012 ª2012 Elsevier Ltd

ical applications. In some cases, this problem can be overcome

by chemically modifying the nucleic acid backbone after the

selection is complete; however, structural changes of this type

are often expensive to implement and come with uncertain

outcomes. Synthetic genetics offers an alternative solution to

this problem by producing functional molecules with nucleic

acid backbones that are not recognized by natural enzymes.

This property of molecular invisibility represents a possible para-

digm shift, as functional molecules will now be cloaked behind

an unnatural nucleic acid backbone. Whether XNA polymers

will be able to supplant their DNA and RNA counterparts will

depend on how well these molecules function in biological envi-

ronments and the availability of XNA triphosphates and enzymes

needed to produce these molecules by in vitro selection.

Beyond therapeutic and diagnostic applications, synthetic

genetics could find widespread use in future synthetic biology

projects. Many practitioners of synthetic biology treat gene

networks found in living organisms as electronic circuitry that

can be ‘‘rewired’’ to produce engineered organisms with new

metabolic pathways embedded in their genomes (Benner and

Sismour, 2005). This process of reading and writing genetic

information relies on the fact that DNA is a universal language

recognized by all organisms. Consequently, synthetic biologists

often view DNA, as a plug-and-play technology where genetic

pathways are seamlessly transplanted from one organism into

another. Unfortunately, in practice, DNA does not always func-

tion in a predictable manner, and unexpected problems can

arise due to such genetic properties as epistasis and pleiotropy,

which occur when the effects of one gene are modified by other

genes or when a single gene exhibits multiple phenotypic traits,

respectively. Another disadvantage of using DNA is that safety

measures are difficult to engineer, as wild-type and engineered

genes cannot be distinguished by the cellular machinery respon-

sible for transcribing and translating genetic information. This

problem raises the concern that aberrant mutations in engi-

neered organisms could have unintended consequences that

are harmful to society if the organism escapes into the wild.

One way to mitigate the risks of bioengineering is to use genetic

polymers that are not recognized by natural enzymes (Herdewijn

and Marliere, 2009). Using this approach, engineered organisms

that escape into the wild would pose a minimal risk to society

because XNA substrates and XNA enzymes do not exist natu-

rally. In this light, XNA polymers represent an attractive genetic

material for synthetic biology as their molecular invisibility

provides the orthogonality needed to establish a genetic firewall

between nature and synthetic biology (Schmidt, 2010). Although

widespread use of synthetic genetic polymers may be years

away, the potential for XNA polymers to function as safe and

effective tools in synthetic biology warrants continued techno-

logical development.

ConclusionsThe emerging field of synthetic genetics provides a rich opportu-

nity to explore the functional landscape of artificial genetic

polymers by in vitro evolution. Currently, the most promising

xenonucleic acids are TNA and HNA—two unnatural genetic

polymers capable of undergoing Darwinian evolution in the labo-

ratory. We anticipate that the continued development of nucleic

acid chemistry and molecular biology techniques will enable the

All rights reserved

Chemistry & Biology

Review

creation of additional examples of XNA polymers that exhibit

heredity and evolution (Deleavey and Damha, 2012; Herdewijn

and Marliere, 2012). Further pursuit of these polymers provides

an exciting opportunity to develop novel genetic polymers with

significant downstream applications in exobiology, molecular

medicine, and synthetic biology.

ACKNOWLEDGMENTS

The authors thank G. Joyce, P. Herdewijn, and members of the Chaput labfor helpful comments and suggestions. This work was supported by theBiodesign Institute at Arizona State University.

REFERENCES

Appella, D.H. (2009). Non-natural nucleic acids for synthetic biology.Curr. Opin. Chem. Biol. 13, 687–696.

Appella, D.H. (2010). Directed evolution: overcoming biology’s limitations. Nat.Chem. Biol. 6, 87–88.

Aurup, H., Tuschl, T., Benseler, F., Ludwig, J., and Eckstein, F. (1994).Oligonucleotide duplexes containing 20-amino-20-deoxycytidines: thermalstability and chemical reactivity. Nucleic Acids Res. 22, 20–24.

Benner, S.A., and Sismour, A.M. (2005). Synthetic biology. Nat. Rev. Genet. 6,533–543.

Betz, K., Malyshev, D.A., Lavergne, T., Welte, W., Diederichs, K., Dwyer, T.J.,Ordoukhanian, P., Romesberg, F.E., and Marx, A. (2012). KlenTaq polymerasereplicates unnatural base pairs by inducing a Watson-Crick geometry. Nat.Chem. Biol. 8, 612–614.

Brudno, Y., and Liu, D.R. (2009). Recent progress toward the templatedsynthesis and directed evolution of sequence-defined synthetic polymers.Chem. Biol. 16, 265–276.

Brudno, Y., Birnbaum, M.E., Kleiner, R.E., and Liu, D.R. (2010). An in vitrotranslation, selection and amplification system for peptide nucleic acids.Nat. Chem. Biol. 6, 148–155.

Burmeister, P.E., Lewis, S.D., Silva, R.F., Preiss, J.R., Horwitz, L.R., Pender-grast, P.S., McCauley, T.G., Kurz, J.C., Epstein, D.M., Wilson, C., and Keefe,A.D. (2005). Direct in vitro selection of a 20-O-methyl aptamer to VEGF. Chem.Biol. 12, 25–33.

Burmeister, P.E., Wang, C., Killough, J.R., Lewis, S.D., Horwitz, L.R.,Ferguson, A., Thompson, K.M., Pendergrast, P.S., McCauley, T.G., Kurz, M.,et al. (2006). 20-Deoxy purine, 20-O-methyl pyrimidine (dRmY) aptamers ascandidate therapeutics. Oligonucleotides 16, 337–351.

Campbell, M.A., and Wengel, J. (2011). Locked vs. unlocked nucleic acids(LNA vs. UNA): contrasting structures work towards common therapeuticgoals. Chem. Soc. Rev. 40, 5680–5689.

Carothers, J.M., Oestreich, S.C., Davis, J.H., and Szostak, J.W. (2004). Infor-mational complexity and functional activity of RNA structures. J. Am. Chem.Soc. 126, 5130–5137.

Carothers, J.M., Oestreich, S.C., and Szostak, J.W. (2006). Aptamers selectedfor higher-affinity binding are not more specific for the target ligand. J. Am.Chem. Soc. 128, 7929–7937.

Chaput, J.C., and Switzer, C. (1999). A DNA pentaplex incorporating nucleo-base quintets. Proc. Natl. Acad. Sci. USA 96, 10614–10619.

Chaput, J.C., and Szostak, J.W. (2003). TNA synthesis by DNA polymerases.J. Am. Chem. Soc. 125, 9274–9275.

Chaput, J.C., Ichida, J.K., and Szostak, J.W. (2003). DNA polymerase-mediated DNA synthesis on a TNA template. J. Am. Chem. Soc. 125, 856–857.

Chelliserrykattil, J., and Ellington, A.D. (2004). Evolution of a T7 RNApolymerase variant that transcribes 20-O-methyl RNA. Nat. Biotechnol. 22,1155–1160.

Crick, F.H. (1968). The origin of the genetic code. J. Mol. Biol. 38, 367–379.

Chemistry & Biology 1

Darwin, C.R. (1859). On the Origin of Species by Means of Natural Selection,or the Preservation of Favoured Races in the Struggle for Life (London:John Murray).

Declercq, R., Van Aerschot, A., Read, R.J., Herdewijn, P., and VanMeervelt, L.(2002). Crystal structure of double helical hexitol nucleic acids. J. Am. Chem.Soc. 124, 928–933.

Delaney, J.C., Gao, J., Liu, H., Shrivastav, N., Essigmann, J.M., and Kool, E.T.(2009). Efficient replication bypass of size-expanded DNA base pairs in bacte-rial cells. Angew. Chem. Int. Ed. Engl. 48, 4524–4527.

Demidov, V.V., Potaman, V.N., Frank-Kamenetskii, M.D., Egholm, M.,Buchard, O., Sonnichsen, S.H., and Nielsen, P.E. (1994). Stability of peptidenucleic acids in human serum and cellular extracts. Biochem. Pharmacol.48, 1310–1313.

Deleavey, G.F., and Damha, M.J. (2012). Designing chemically modifiedoligonucleotides for targeted gene silencing. Chem. Biol. 19, 937–954.

Dias, N., and Stein, C.A. (2002). Antisense oligonucleotides: basic conceptsand mechanisms. Mol. Cancer Ther. 1, 347–355.

Ellington, A.D., and Szostak, J.W. (1990). In vitro selection of RNA moleculesthat bind specific ligands. Nature 346, 818–822.

Engelhart, A.E., and Hud, N.V. (2010). Primitive genetic polymers. Cold SpringHarb. Perspect. Biol. 2, a002196.

Eschenmoser, A. (1999). Chemical etiology of nucleic acid structure. Science284, 2118–2124.

Eyetech Study Group. (2003). Anti-vascular endothelial growth factor therapyfor subfoveal choroidal neovascularization secondary to age-related maculardegeneration: phase II study results. Ophthalmology 110, 979–986.

Fa, M., Radeghieri, A., Henry, A.A., and Romesberg, F.E. (2004). Expandingthe substrate repertoire of a DNA polymerase by directed evolution. J. Am.Chem. Soc. 126, 1748–1754.

Famulok, M., Hartig, J.S., and Mayer, G. (2007). Functional aptamers andaptazymes in biotechnology, diagnostics, and therapy. Chem. Rev. 107,3715–3743.

Ghadessy, F.J., Ramsay, N., Boudsocq, F., Loakes, D., Brown, A., Iwai, S.,Vaisman, A., Woodgate, R., and Holliger, P. (2004). Generic expansion ofthe substrate spectrum of a DNA polymerase by directed evolution. Nat.Biotechnol. 22, 755–759.

Gilbert, W. (1986). The RNA world. Nature (London) 319, 618.

Guatelli, J.C., Whitfield, K.M., Kwoh, D.Y., Barringer, K.J., Richman, D.D., andGingeras, T.R. (1990). Isothermal, in vitro amplification of nucleic acids bya multienzyme reaction modeled after retroviral replication. Proc. Natl. Acad.Sci. USA 87, 1874–1878.

Hall, A.H., Wan, J., Shaughnessy, E.E., Ramsay Shaw, B., and Alexander, K.A.(2004). RNA interference using boranophosphate siRNAs: structure-activityrelationships. Nucleic Acids Res. 32, 5991–6000.

Hall, A.H., Wan, J., Spesock, A., Sergueeva, Z., Shaw, B.R., and Alexander,K.A. (2006). High potency silencing by single-stranded boranophosphatesiRNA. Nucleic Acids Res. 34, 2773–2781.

Hawthorne, M.F. (1993). The role of chemistry in the development of boronneutron capture therapy of cancer. Angew. Chem. Int. Ed. Engl. 32, 950–984.

Hendrix, C., Rosemeyer, H., DeBouvere, B., VanAerschot, A., Seela, F., andHerdewijn, P. (1997a). 10,50-anhydrohexitol oligonucleotides: Hybridisationand strand displacement with oligoribonucleotides, interaction with RNase Hand HIV reverse transcriptase. Chemistry 3, 1513–1520.

Hendrix, C., Rosemeyer, H., Verheggen, I., Seela, F., VanAerschot, A., andHerdewijn, P. (1997b). 10,50-anhydrohexitol oligonucleotides: Synthesis, basepairing and recognition by regular oligodeoxyribonucleotides and oligoribonu-cleotides. Chemistry 3, 110–120.

Henry, A.A., and Romesberg, F.E. (2003). Beyond A, C, G and T: augmentingnature’s alphabet. Curr. Opin. Chem. Biol. 7, 727–733.

Herdewijn, P. (2010). Nucleic acids with a six-membered ‘carbohydrate’ mimicin the backbone. Chem. Biodivers. 7, 1–59.

9, November 21, 2012 ª2012 Elsevier Ltd All rights reserved 1369

Chemistry & Biology

Review

Herdewijn, P., and Marliere, P. (2009). Toward safe genetically modifiedorganisms through the chemical diversification of nucleic acids. Chem.Biodivers. 6, 791–808.

Herdewijn, P., andMarliere, P. (2012). Redesigning the leaving group in nucleicacid polymerization. FEBS Lett. 586, 2049–2056.

Hollenstein, M., Hipolito, C.J., Lam, C.H., and Perrin, D.M. (2009a). ADNAzyme with three protein-like functional groups: enhancing catalytic effi-ciency of M2+-independent RNA cleavage. ChemBioChem 10, 1988–1992.

Hollenstein, M., Hipolito, C.J., Lam, C.H., and Perrin, D.M. (2009b). Aself-cleaving DNA enzyme modified with amines, guanidines and imidazolesoperates independently of divalent metal cations (M2+). Nucleic Acids Res.37, 1638–1649.

Horhota, A., Zou, K., Ichida, J.K., Yu, B., McLaughlin, L.W., Szostak, J.W., andChaput, J.C. (2005). Kinetic analysis of an efficient DNA-dependent TNApolymerase. J. Am. Chem. Soc. 127, 7427–7434.

Ichida, J.K., Horhota, A., Zou, K.Y., McLaughlin, L.W., and Szostak, J.W.(2005). High fidelity TNA synthesis by Therminator polymerase. Nucleic AcidsRes. 33, 5219–5225.

Jhaveri, S., Olwin, B., and Ellington, A.D. (1998). In vitro selection of phosphor-othiolated aptamers. Bioorg. Med. Chem. Lett. 8, 2285–2290.

Joyce, G.F. (1994). In vitro evolution of nucleic acids. Curr. Opin. Struct. Biol. 4,331–336.

Joyce, G.F. (2002). The antiquity of RNA-based evolution. Nature 418,214–221.

Joyce, G.F. (2004). Directed evolution of nucleic acid enzymes. Annu. Rev.Biochem. 73, 791–836.

Joyce, G.F. (2007). Forty years of in vitro evolution. Angew. Chem. Int. Ed.Engl. 46, 6420–6436.

Joyce, G.F. (2012). Evolution. Toward an alternative biology. Science 336,307–308.

Joyce, G.F., Schwartz, A.W., Miller, S.L., and Orgel, L.E. (1987). The case foran ancestral genetic system involving simple analogues of the nucleotides.Proc. Natl. Acad. Sci. USA 84, 4398–4402.

Kang, J., Lee, M.S., Watowich, S.J., and Gorenstein, D.G. (2007). Combinato-rial selection of a RNA thioaptamer that binds to Venezuelan equine encepha-litis virus capsid protein. FEBS Lett. 581, 2497–2502.

Kang, J., Lee, M.S., Copland, J.A., 3rd, Luxon, B.A., and Gorenstein, D.G.(2008). Combinatorial selection of a single stranded DNA thioaptamer target-ing TGF-b1 protein. Bioorg. Med. Chem. Lett. 18, 1835–1839.

Kang, M., Heuberger, B., Chaput, J.C., Switzer, C., and Feigon, J. (2012).Solution structure of a parallel-stranded oligoisoguanine DNA pentaplexformed by d(T(iG)4T) in the presence of Cs+ ions. Angew. Chem. Int. Ed.Engl. 51, 7952–7955.

Keefe, A.D., and Cload, S.T. (2008). SELEX with modified nucleotides. Curr.Opin. Chem. Biol. 12, 448–456.

Keefe, A.D., Pai, S., and Ellington, A. (2010). Aptamers as therapeutics. Nat.Rev. Drug Discov. 9, 537–550.

Kempeneers, V., Vastmans, K., Rozenski, J., and Herdewijn, P. (2003). Recog-nition of threosyl nucleotides by DNA and RNA polymerases. Nucleic AcidsRes. 31, 6221–6226.

King, D.J., Ventura, D.A., Brasier, A.R., and Gorenstein, D.G. (1998). Novelcombinatorial selection of phosphorothioate oligonucleotide aptamers.Biochemistry 37, 16489–16493.

Klussmann, S., Nolte, A., Bald, R., Erdmann, V.A., and Furste, J.P. (1996).Mirror-image RNA that binds D-adenosine. Nat. Biotechnol. 14, 1112–1115.

Kool, E.T. (2002a). Active site tightness and substrate fit in DNA replication.Annu. Rev. Biochem. 71, 191–219.

Kool, E.T. (2002b). Replacing the nucleobases in DNA with designer mole-cules. Acc. Chem. Res. 35, 936–943.

1370 Chemistry & Biology 19, November 21, 2012 ª2012 Elsevier Ltd

Krishnamurthy, R., Pitsch, S., Minton, M., Miculka, C., Windhab, N., andEschenmoser, A. (1996). Pyranosyl-RNA: Base pairing between homochiraloligonucleotide strands of opposite sense of chirality. Angew. Chem. Int. Ed.Engl. 35, 1537–1541.

Krishnan, Y., and Simmel, F.C. (2011). Nucleic acid based molecular devices.Angew. Chem. Int. Ed. Engl. 50, 3124–3156.

Krueger, A.T., and Kool, E.T. (2007). Model systems for understanding DNAbase pairing. Curr. Opin. Chem. Biol. 11, 588–594.

Krueger, A.T., Peterson, L.W., Chelliserry, J., Kleinbaum, D.J., and Kool, E.T.(2011). Encoding phenotype in bacteria with an alternative genetic set. J.Am. Chem. Soc. 133, 18447–18451.

Kruger, K., Grabowski, P.J., Zaug, A.J., Sands, J., Gottschling, D.E., andCech, T.R. (1982). Self-splicing RNA: autoexcision and autocyclization of theribosomal RNA intervening sequence of Tetrahymena. Cell 31, 147–157.

Lato, S.M., Ozerova, N.D., He, K., Sergueeva, Z., Shaw, B.R., and Burke, D.H.(2002). Boron-containing aptamers to ATP. Nucleic Acids Res. 30, 1401–1407.

Lavergne, T., Malyshev, D.A., and Romesberg, F.E. (2012). Major groovesubstituents and polymerase recognition of a class of predominantly hydro-phobic unnatural base pairs. Chemistry 18, 1231–1239.

Lazcano, A., and Miller, S.L. (1996). The origin and early evolution of life:prebiotic chemistry, the pre-RNA world, and time. Cell 85, 793–798.

Lescrinier, E., Esnouf, R., Schraml, J., Busson, R., Heus, H.A., Hilbers, C.W.,and Herdewijn, P. (2000). Solution structure of a HNA-RNA hybrid. Chem.Biol. 7, 719–731.

Lesnik, E.A., Guinosso, C.J., Kawasaki, A.M., Sasmor, H., Zounes, M.,Cummins, L.L., Ecker, D.J., Cook, P.D., and Freier, S.M. (1993). Oligodeoxynu-cleotides containing 20-O-modified adenosine: synthesis and effects onstability of DNA:RNA duplexes. Biochemistry 32, 7832–7838.

Leva, S., Lichte, A., Burmeister, J., Muhn, P., Jahnke, B., Fesser, D., Erfurth, J.,Burgstaller, P., and Klussmann, S. (2002). GnRH binding RNA and DNA Spie-gelmers: a novel approach toward GnRH antagonism. Chem. Biol. 9, 351–359.

Li, N.S., Frederiksen, J.K., and Piccirilli, J.A. (2011). Synthesis, properties, andapplications of oligonucleotides containing an RNA dinucleotide phosphoro-thiolate linkage. Acc. Chem. Res. 44, 1257–1269.

Li, P., Sergueeva, Z.A., Dobrikov, M., and Shaw, B.R. (2007). Nucleoside andoligonucleoside boranophosphates: chemistry and properties. Chem. Rev.107, 4746–4796.

Lin, Y., Qiu, Q., Gill, S.C., and Jayasena, S.D. (1994). Modified RNA sequencepools for in vitro selection. Nucleic Acids Res. 22, 5229–5234.

Liu, H., Gao, J., Lynch, S.R., Saito, Y.D., Maynard, L., and Kool, E.T. (2003). Afour-base paired genetic helix with expanded size. Science 302, 868–871.

Loakes, D., Gallego, J., Pinheiro, V.B., Kool, E.T., and Holliger, P. (2009).Evolving a polymerase for hydrophobic base analogues. J. Am. Chem. Soc.131, 14827–14837.

Maier, T., Przylas, I., Strater, N., Herdewijn, P., and Saenger, W. (2005). Rein-forced HNA backbone hydration in the crystal structure of a decameric HNA/RNA hybrid. J. Am. Chem. Soc. 127, 2937–2943.

Malyshev, D.A., Seo, Y.J., Ordoukhanian, P., and Romesberg, F.E. (2009).PCR with an expanded genetic alphabet. J. Am. Chem. Soc. 131, 14620–14621.

Mascini, M., Palchetti, I., and Tombelli, S. (2012). Nucleic acid and peptide ap-tamers: fundamentals and bioanalytical aspects. Angew. Chem. Int. Ed. Engl.51, 1316–1332.

Mayer, G. (2009). The chemical biology of aptamers. Angew. Chem. Int. Ed.Engl. 48, 2672–2689.

Mi, J., Liu, Y., Rabbani, Z.N., Yang, Z., Urban, J.H., Sullenger, B.A., and Clary,B.M. (2010). In vivo selection of tumor-targeting RNA motifs. Nat. Chem. Biol.6, 22–24.

Mills, D.R., Peterson, R.L., and Spiegelman, S. (1967). An extracellularDarwinian experiment with a self-duplicating nucleic acid molecule. Proc.Natl. Acad. Sci. USA 58, 217–224.

All rights reserved

Chemistry & Biology

Review

Mitsui, T., Kitamura, A., Kimoto, M., To, T., Sato, A., Hirao, I., and Yokoyama,S. (2003). An unnatural hydrophobic base pair with shape complementaritybetween pyrrole-2-carbaldehyde and 9-methylimidazo[(4,5)-b]pyridine. J.Am. Chem. Soc. 125, 5298–5307.

Moran, S., Ren, R.X.-F., Rumney, S., and Kool, E.T. (1997). Difluorotoluene,a nonpolar isostere for thymine, codes specifically and efficiently for adeninein DNA replication. J. Am. Chem. Soc. 119, 2056–2057.

Nielsen, P.E. (1995). DNA analogues with nonphosphodiester backbones.Annu. Rev. Biophys. Biomol. Struct. 24, 167–183.

Nielsen, P.E., Egholm, M., Berg, R.H., and Buchardt, O. (1991a). Sequence-selective recognition of DNA by strand displacement with a thymine-substituted polyamide. Science 254, 1497–1500.

Nielsen, P.E., Egholm, M., Berg, R.H., and Buchardt, O. (1991b). Sequence-selective recognition of DNA by strand displacement with a thymine-substituted polyamide. Science 254, 1497–1500.

Nimjee, S.M., Rusconi, C.P., and Sullenger, B.A. (2005). Aptamers: anemerging class of therapeutics. Annu. Rev. Med. 56, 555–583.

Nolte, A., Klussmann, S., Bald, R., Erdmann, V.A., and Furste, J.P. (1996).Mirror-design of L-oligonucleotide ligands binding to L-arginine. Nat. Bio-technol. 14, 1116–1119.

Noronha, A.M., Wilds, C.J., Lok, C.N., Viazovkina, K., Arion, D., Parniak, M.A.,and Damha, M.J. (2000). Synthesis and biophysical properties of arabi-nonucleic acids (ANA): circular dichroic spectra, melting temperatures, andribonuclease H susceptibility of ANA.RNA hybrid duplexes. Biochemistry 39,7050–7062.

Olea, C., Jr., Horning, D.P., and Joyce, G.F. (2012). Ligand-dependentexponential amplification of a self-replicating L-RNA enzyme. J. Am. Chem.Soc. 134, 8050–8053.

Orgel, L. (2000). Origin of life. A simpler nucleic acid. Science 290, 1306–1307.

Orgel, L.E. (1968). Evolution of the genetic apparatus. J. Mol. Biol. 38,381–393.

Piccirilli, J.A., Krauch, T., Moroney, S.E., and Benner, S.A. (1990). Enzymaticincorporation of a new base pair into DNA and RNA extends the geneticalphabet. Nature 343, 33–37.

Pinheiro, V.B., Taylor, A.I., Cozens, C., Abramov, M., Renders, M., Zhang, S.,Chaput, J.C., Wengel, J., Peak-Chew, S.Y., McLaughlin, S.H., et al. (2012).Synthetic genetic polymers capable of heredity and evolution. Science 336,341–344.

Purschke, W.G., Radtke, F., Kleinjung, F., and Klussmann, S. (2003). A DNASpiegelmer to staphylococcal enterotoxin B. Nucleic Acids Res. 31, 3027–3032.

Robertson, D.L., and Joyce, G.F. (1990). Selection in vitro of an RNA enzymethat specifically cleaves single-stranded DNA. Nature 344, 467–468.

Ruckman, J., Green, L.S., Beeson, J., Waugh, S., Gillette, W.L., Henninger,D.D., Claesson-Welsh, L., and Janji�c, N. (1998). 20-Fluoropyrimidine RNA-based aptamers to the 165-amino acid form of vascular endothelial growthfactor (VEGF165). Inhibition of receptor binding and VEGF-induced vascularpermeability through interactions requiring the exon 7-encoded domain. J.Biol. Chem. 273, 20556–20567.

Saiki, R.K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A., andArnheim, N. (1985). Enzymatic amplification of b-globin genomic sequencesand restriction site analysis for diagnosis of sickle cell anemia. Science 230,1350–1354.

Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T.,Mullis, K.B., and Erlich, H.A. (1988). Primer-directed enzymatic amplificationof DNA with a thermostable DNA polymerase. Science 239, 487–491.

Sassanfar, M., and Szostak, J.W. (1993). An RNAmotif that binds ATP. Nature364, 550–553.

Schmidt, M. (2010). Xenobiology: a new form of life as the ultimate biosafetytool. Bioessays 32, 322–331.

Schoning, K., Scholz, P., Guntha, S., Wu, X., Krishnamurthy, R., andEschenmoser, A. (2000). Chemical etiology of nucleic acid structure: thea-threofuranosyl-(30—>20 ) oligonucleotide system. Science 290, 1347–1351.

Chemistry & Biology 1

Shiraishi, T., and Nielsen, P.E. (2011). Improved cellular uptake of antisensepeptide nucleic acids by conjugation to a cell-penetrating peptide and a lipiddomain. Methods Mol. Biol. 751, 209–221.

Somasunderam, A., Ferguson, M.R., Rojo, D.R., Thiviyanathan, V., Li, X.,O’Brien,W.A., andGorenstein, D.G. (2005). Combinatorial selection, inhibition,and antiviral activity of DNA thioaptamers targeting the RNase H domain ofHIV-1 reverse transcriptase. Biochemistry 44, 10388–10395.

Szostak, J.W. (1992). In vitro genetics. Trends Biochem. Sci. 17, 89–93.

Tombelli, S., and Mascini, M. (2009). Aptamers as molecular tools forbioanalytical methods. Curr. Opin. Mol. Ther. 11, 179–188.

Trevino, S.G., Zhang, N., Elenko, M.P., Luptak, A., and Szostak, J.W. (2011).Evolution of functional nucleic acids in the presence of nonheritable backboneheterogeneity. Proc. Natl. Acad. Sci. USA 108, 13492–13497.

Tuerk, C., and Gold, L. (1990). Systematic evolution of ligands by exponentialenrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249,505–510.

Vaish, N.K., Fraley, A.W., Szostak, J.W., and McLaughlin, L.W. (2000).Expanding the structural and functional diversity of RNA: analog uridinetriphosphates as candidates for in vitro selection of nucleic acids. NucleicAcids Res. 28, 3316–3322.

Vaught, J.D., Bock, C., Carter, J., Fitzwater, T., Otis, M., Schneider, D.,Rolando, J., Waugh, S., Wilcox, S.K., and Eaton, B.E. (2010). Expanding thechemistry of DNA for in vitro selection. J. Am. Chem. Soc. 132, 4141–4151.

Verheggen, I., Van Aerschot, A., Toppet, S., Snoeck, R., Janssen, G., Balzarini,J., De Clercq, E., and Herdewijn, P. (1993). Synthesis and antiherpes virusactivity of 1,5-anhydrohexitol nucleosides. J. Med. Chem. 36, 2033–2040.

Verheggen, I., Van Aerschot, A., Van Meervelt, L., Rozenski, J., Wiebe, L.,Snoeck, R., Andrei, G., Balzarini, J., Claes, P., De Clercq, E., et al. (1995).Synthesis, biological evaluation, and structure analysis of a series of new1,5-anhydrohexitol nucleosides. J. Med. Chem. 38, 826–835.

Wilds, C.J., and Damha, M.J. (2000). 20-Deoxy-20-fluoro-b-D-arabinonucleo-sides and oligonucleotides (20F-ANA): synthesis and physicochemical studies.Nucleic Acids Res. 28, 3625–3635.

Williams, K.P., Liu, X.H., Schumacher, T.N., Lin, H.Y., Ausiello, D.A., Kim, P.S.,and Bartel, D.P. (1997). Bioactive and nuclease-resistant L-DNA ligand ofvasopressin. Proc. Natl. Acad. Sci. USA 94, 11285–11290.

Wilson, D.S., and Szostak, J.W. (1999). In vitro selection of functional nucleicacids. Annu. Rev. Biochem. 68, 611–647.

Wittung, P., Kajanus, J., Edwards, K., Nielsen, P., Norden, B., andMalmstrom,B.G. (1995). Phospholipid membrane permeability of peptide nucleic acid.FEBS Lett. 365, 27–29.

Woese, C.R. (1967). The genetic code: Themolecular basis for genetic expres-sion (New York: Harper & Row).

Wohler, F. (1828). Ueber kunstliche Bildung des Harnstoffs. Ann. Chim.Physiol. 37, 330–333.

Xu, Y., and Kool, E.T. (1998). Chemical and enzymatic properties of bridging50-S-phosphorothioester linkages in DNA. Nucleic Acids Res. 26, 3159–3164.

Yang, Z., Chen, F., Chamberlin, S.G., and Benner, S.A. (2010). Expandedgenetic alphabets in the polymerase chain reaction. Angew. Chem. Int. Ed.Engl. 49, 177–180.

Yu, H., Zhang, S., and Chaput, J.C. (2012). Darwinian evolution of an alterna-tive genetic system provides support for TNA as an RNA progenitor. Nat.Chem. 4, 183–187.

Zhang, L., Peritz, A., and Meggers, E. (2005). A simple glycol nucleic acid. J.Am. Chem. Soc. 127, 4174–4175.

Zhang, S., Switzer, C., and Chaput, J.C. (2010). The resurgence of acyclic nu-cleic acids. Chem. Biodivers. 7, 245–258.

Zhou, J., Swiderski, P., Li, H., Zhang, J., Neff, C.P., Akkina, R., and Rossi, J.J.(2009). Selection, characterization and application of new RNA HIV gp 120aptamers for facile delivery of Dicer substrate siRNAs into HIV infected cells.Nucleic Acids Res. 37, 3094–3109.

9, November 21, 2012 ª2012 Elsevier Ltd All rights reserved 1371