Phosphonomethyl Oligonucleotides as Backbone-Modified Artificial Genetic Polymers

Article abstract of DOI:10.1021/jacs.8b03447

Although several synthetic or xenobiotic nucleic acids (XNAs) have been shown to be viable genetic materials in vitro, major hurdles remain for their in vivo applications, particularly orthogonality. The availability of XNAs that do not interact with natu

Full text of DOI:10.1021/jacs.8b03447

Subscriber access provided by Kaohsiung Medical University
Article
Phosphonomethyl Oligonucleotides as
Backbone Modified Artifi-cial Genetic Polymers
Chao Liu, Christopher Cozens, Faten Jaziri, Jef Rozenski, Amandine Maréchal, Shrinivas G.
Dumbre, Valérie Pezo, philippe marliere, Vitor B. Pinheiro, Elisabetta Groaz, and Piet Herdewijn
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b03447 • Publication Date (Web): 03 May 2018
Downloaded from http://pubs.acs.org on May 3, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 15
Journal of the American Chemical Society
1
2
3
4
5
6
Phosphonomethyl Oligonucleotides as Backbone Modified Artificial
Genetic Polymers
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†,
§,⊥
‡
†
§
†
⊥
Chao Liu, Christopher Cozens, Faten Jaziri, Jef Rozenski, Amandine Marechal, Shrinivas Dumbre, Valéꢀ
‡
‡
§, ,
†
†,‡,
∥
rie Pezo, Philippe Marlière, Vitor B. Pinheiro, * Elisabetta Groaz, and Piet Herdewijn *
†
Medicinal Chemistry, Rega Institute for Medical Research, KU Leuven, Herestraat 49, 3000 Leuven, Belgium
§
University College London, Gower Street, London WC1E 6BT, UK
‡
iSSB, Genopole, CNRS, UEVE, Université ParisꢀSaclay, 5 rue Henri Desbruères, 91030 Evry Cedex, France
∥Institute of Structural and Molecular Biology, Birkbeck College, University of London, Malet Street, WC1E 7HX, UK
⊥These authors contributed equally to this work.
KEYWORDS: XNA; Phosphono nucleic acid; Polymerase evolution; In vivo transliteration; HꢀPhosphinates; Xenobiology
ABSTRACT: Although several synthetic or xenobiotic nucleic acids (XNAs) have been shown to be viable genetic materials in
vitro, major hurdles remain for their in vivo applications, particularly orthogonality. The availability of XNAs that do not interact
with natural nucleic acids and are not affected by natural DNA processing enzymes, as well as specialized XNA processing enꢀ
zymes that do not interact with natural nucleic acids, are essential. Here, we report 3’–2’ phosphonomethylꢀthreosyl nucleic acid
(tPhoNA) as a novel XNA genetic material and a prime candidate for in vivo XNA applications. We established routes for the
chemical synthesis of phosphonate nucleic acids and phosphorylated monomeric building blocks, and we demonstrated that DNA
duplexes were destabilized upon replacement with tPhoNA. We engineered a novel tPhoNA synthetase enzyme and, with a previꢀ
ously reported XNA reverse transcriptase, demonstrated that tPhoNA is a viable genetic material (with an aggregate error rate of
ꢀ
3
approximately 17 x 10 per base) compatible with the isolation of functional XNAs. In vivo experiments to test tPhoNA orthogoꢀ
nality showed that the E. coli cellular machinery had only very limited potential to access genetic information in tPhoNA. Our work
is the first report of a synthetic genetic material modified in both sugar and phosphate backbone moieties and represents a signifiꢀ
cant advance in biorthogonality towards the introduction of XNA systems in vivo.
further restrict in vivo applications of XNAs, highlighting the
need for alternative designs in this area.
INTRODUCTION
In RNA and DNA, life on Earth has settled on uniquely efꢀ
ficient molecules to store and, facilitated by highly efficient
enzymes, propagate genetic information. While other moleꢀ
Among all the XNA genetic systems studied to date,
(3′,2′)ꢀαꢀLꢀthreose nucleic acid (TNA, Figure 1) has generated
1
considerable interest as a possible RNA progenitor or competiꢀ
tor because of the chemical simplicity of threose relative to
ribose and the ability of TNA to form stable duplexes with
cules (XNAs, xenobiotic nucleic acids) can fulfil the function
of those natural genetic polymers, establishing them as bona
fide genetic materials is challenging. It requires the efficient
chemical synthesis of unnatural precursors as well as isolation
of polymerases that can, at least, transfer genetic information
RNA and DNA, despite having a repeating backbone unit one
14ꢀ16
atom shorter than that of natural nucleic acids.
Storage of
2,3
genetic information in TNA has been established through
synthesis of TNA oligomers by Therminator DNA polymerꢀ
in and out of a natural system efficiently. It is especially
important to acquire efficient chemical methods to synthesize
primers for polymerase selection as well as specific oligonuꢀ
cleotide sequences in order to establish whether XNA polyꢀ
mers could support duplex formation with a complementary
XNA, DNA, or RNA strand and exhibit canonical Watsonꢀ
Crick pairing. While XNAs are established as synthetic genetꢀ
17
ase and reverse transcription by several natural and engiꢀ
4, 18, 19
neered enzymes.
Previously, our group demonstrated that Therminator polꢀ
ymerase was also able to catalyze the condensation of the
diphosphate derivatives of both phosphonomethylthreosylꢀ
4
adenine
(PMTApp)
and
5’ꢀOꢀphosphonomethylꢀ2’ꢀ
ic polymers in vitro, only very few studies have been conꢀ
5
ꢀ9
deoxyadenine (PMdApp) with the formation of phosphonate
ducted to evaluate backbone modified XNAs in vivo. So far,
the replication of such XNA systems has been accomplished
through a DNA intermediate, which is not compatible with the
in vivo generation of an orthogonal XNA episome. Notably,
the study of backbone modified XNAs remains underexplored
oligonucleotide stretches, which were found to be resistant to
20,21
degradation by snakeꢀvenom phosphodiesterase.
The solidꢀ
phase synthesis of 5’ꢀOꢀphosphonomethylꢀdeoxyribosyl oliꢀ
gonucleotides (dPhoNA, Figure 1) was first reported by Rosꢀ
2
2
1
0ꢀ13
enberg et al. via the phosphotriester method. The asꢀ
compared to that of base modified XNAs.
limited metabolic stability, lack of orthogonality, and toxicity
Moreover, a
synthesized phosphonate oligonucleotides exhibited high
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 15
stability against nucleases of L1210 cell free extract and posꢀ Alternatively, the adenine Hꢀphosphinate building block
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
sessed the ability to form triplexes with natural counterparts.
In a recent study, oligothymidylates containing various ratios
of 5’ꢀOꢀmethylphosphonate internucleotide linkages were
found to be able to induce RNA cleavage by ribonuclease H
containing a benzoyl protected adenine base, compound 3e,
could be obtained by directly introducing the Hꢀphosphinate
functionality into the sugar moiety of threose nucleoside inꢀ
2
6
termediate 7, as described by Kostov et al. (Scheme 1). Seꢀ
23
(RNase H).
lective hydrolysis of the benzoyl group at the 2’ꢀposition of
24
fully protected nucleoside 4 under basic conditions gave
compound 5 in 82% yield. Protection of the 2’ꢀhydroxyl group
of 5 with MMTrCl afforded nucleoside 6 in 50% yield, along
with the recovery of 42% of the starting material. The subseꢀ
quent removal of the 3’ꢀtertꢀbutyldimethylsilyl protecting
group was carried out by treatment of 6 with triethylamine
trifluoride in THF. Next, compound 7 was reacted with sodiꢀ
um 4ꢀtolueneꢀsulfonyloxymethylꢀ(H)ꢀphosphinate 8 in the
presence of sodium hydride in DMF to give Hꢀphosphinate
nucleoside 3e in 74% yield. Finally, all these Hꢀphosphinate
monomers were transformed to their DBU salts prior to solidꢀ
phase synthesis. In addition, for comparison with phosphoꢀ
nateꢀlinkage modified oligonucleotides including a natural
sugar in the backbone, 5’ꢀOꢀphosphonomethylꢀdeoxyribosyl
oligonucleotides (dPhoNA) were also prepared and likewise
The main objective of this study was to investigate the potenꢀ
tial of 3’–2’ phosphonomethylꢀthreosyl nucleic acid (tPhoNA)
as novel XNA genetic material for in vivo applications, particꢀ
ularly in comparison with dPhoNA. The presumed enzymatic
and chemical stability of the phosphonate linkage, which
would avoid degradation by cellular enzymes, along with a
significant advance in biorthogonality due to the concurrent
modification of both sugar and phosphate backbone moieties
are significant advantages of those chemistries. First, synthetic
routes were established for the preparation of phosphonate
oligonucleotides as well as their corresponding phosphorylated
building blocks required for chemical and enzymatic syntheꢀ
sis. We then characterized the influence of tPhoNA and
dPhoNA on duplex stability, establish tPhoNA as a synthetic
genetic polymer through polymerase engineering, permitting
enzymatic tPhoNA synthesis and reverse transcription, and
finally demonstrated tPhoNA’s orthogonality by in vivo transꢀ
literation experiments.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6
2
’ꢀDeoxyribonucleosideꢀ5’ꢀOꢀmethylꢀ(H)ꢀphosphinates (S4b-
e) were synthesized starting from deoxyribonucleosides in a
similar manner as that described for threosyl Hꢀphosphinate
monomers (Scheme S1).
B
B
B
B
O
O
O
O
O
O
O
O
O
Scheme 1. Synthetic routes for the preparation of threosyl
nucleoside H-phosphinate monomers 3b-e required for the
solid-phase synthesis of tPhoNA containing oligonucleo-
tides. Similar synthetic steps were employed for the prepa-
ration of deoxyribosyl nucleoside H-phosphinate building
blocks (S4b-e) for dPhoNA synthesis, as detailed in the SI
(Scheme S1).
O
P
O
P
O
P
O
P
O-
O
O
Oꢀ
O
Oꢀ
B
O
B
B
O
-
B
O
O
O
O
O
O
O
O
O
O
O
DNA
TNA
tPhoNA
dPhoNA
Figure 1. Chemical structures of phosphono nucleic acids (tPhoꢀ
NA and dPhoNA) as compared to DNAꢀ and TNAꢀbased phosꢀ
phodiester backbones.
O
O
O
P
H
iPrO
P
O
B
MMTrCl
iPrO
P
O
B
1. LiAlH4, TMSCl HO
2. H
65ꢀ74%
O
B
O
O
O
OiPr
AgNO3
47ꢀ90%
OiPr
2
O
2
OH
OMMTr
OMMTr
2a B = ABz
b B = T
3a B = A
3b B = T
3c B = Hx
a B = ABz
b B = T
1
1
NH
MeOH
8%
3
in
2 2
Me NCH(OMe)
2
2c B = Hx
2d B = A
69%
Dmf
1c B = Hx
3d B = A
8
Bz
e B = A
3
RESULTS AND DISCUSSION
O
P
H
Phosphonate Nucleic Acid Synthesis. Chemical Synthesis
of Nucleoside HꢀPhosphinate Monomers. As depicted in
Scheme 1, threosyl Hꢀphosphinate monomers 3b-d were preꢀ
pared starting from 3’ꢀdiisopropylphosphonomethyl threosyl
NaO
OTs
NaH
74%
8
ABz
O
TBSO
A
Bz
ABz
ABz
TBSO
TBSO
HO
O
O
O
NaOH
82%
MMTrCl
AgNO
0%
3 3
Et N— HF
3
83%
OBz
OH
OMMTr
OMMTr
5
Bz
4
5
6
7
nucleosides 1a (B = benzoyl adenine, A ), 1b (B = thymine,
T), and 1c (B = hypoxanthine, Hx), respectively, which were
Synthesis of tPhoNAꢀDNA and dPhoNAꢀDNA Chimeras.
Recently, Rosenberg et al. reported the preparation of a fully
modified 7ꢀmer dPhoNA oligomer via Hꢀphosphinate chemisꢀ
in turn synthesized following literature procedures starting
2
4,25
from Lꢀascorbic acid.
Compounds 1a-c were reacted with
MMTrCl in the presence of AgNO and pyridine to afford the
3
try
employing
2ꢀchloroꢀ5,5’ꢀdimethylꢀ1,3,2ꢀ
corresponding 2’ꢀOꢀtritylated nucleoside analogues 2a-c in
dioxaphosphorinaneꢀ2ꢀoxide (NEPꢀCl) and an anhydrous
mixture (CCl –methanol–Et N–MeIm) as coupling and oxidizꢀ
4
7–90% yield. In view of the incompatibility of the benzoyl
4
3
protecting group with the conditions required by the following
26
ing agents, respectively. Here, we opted for the synthesis of
PhoNA oligonucleotides and their DNA chimeras using a
combination of Hꢀphosphinate and conventional phosphoꢀ
ramidite chemistries. However, different conditions were
adopted along the synthesis cycle compared to those previousꢀ
ly reported, as detailed in the Supporting information (Scheme
S3, Table S1).
LiAlH mediated reduction step, compound 2a was first deproꢀ
4
6
tected at the N ꢀposition by treatment with methanolic ammoꢀ
nia to give adenine derivative 2d. Next, the diisopropyl ester
phosphonate functionality of 2b-d was reduced to the correꢀ
sponding phosphine stage in the presence of LiAlH ꢀTMSCl,
4
followed by oxidation with hydrogen peroxide to afford threoꢀ
syl Hꢀphosphinate nucleosides 3a-c in 65–70% yield over two
steps. The free exocyclic amino group of adenineꢀcontaining
nucleoside 3a was then reprotected with a dimethylformaꢀ
midino (dmf) group to provide Hꢀphosphinate monomer 3d.
Specifically, considering that the Hꢀphosphinate linkage is
sensitive to conventional waterꢀcontaining oxidizers (I2 in
pyridine/H O), it was immediately oxidized to a methyl group
2
protected phosphonate linkage after the condensation step.
ACS Paragon Plus Environment

Page 3 of 15
Journal of the American Chemical Society
with A, T, C (cytosine), and G (guanine) bases required for
Second, the oxidation step was followed by a capping step
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
using phenoxyacetic anhydride to avoid failure sequences.
Third, 33% aqueous ammonia was preferred as final deprotecꢀ
tion reagent over previously reported conditions (gaseous
polymerase engineering (Figure 2) were obtained from the
corresponding nucleoside phosphonic acids (S5a-d, Scheme
2
4,27
S2), which were synthesized according to literature reports.
26
NH3).
Previously, the diphosphate derivative of phosphonomethylꢀ
threosyl adenine (PMTApp) was isolated in poor yield (10%)
As proof of principle, a fully modified polythymidylate
4ꢀmer dPhoNA oligomer could be obtained by using this
2
0
by using the N,N′ꢀcarbonyldiimidazole (CDI) approach.
1
Therefore, we decided to prepare all PMTNpp using an alterꢀ
native dicyclohexylcarbodiimide (DCC) mediated condensaꢀ
modified synthesis protocol. Stability studies on this oligomer
revealed that the phosphonate linkage (PꢀCꢀO) was stable
against aqueous alkaline deprotection conditions (Figure S1).
Next, up to six successive threose phosphonate nucleotides
were successfully incorporated into DNA oligomers by this
method. In addition, two 10ꢀmer sequences each flanked by
one DNA unit at both ends were likewise obtained (Table 1,
ON9 and ON10). The mass analysis and sequences of the
synthesized oligomers are listed in Table 1. OligonucleotidDemsf
ON 1, ON 3, ON 11, and ON 14 were prepared using A
protected nucleotides, while ON 5–10 and ON 13 were atꢀ
28
tion method, which pleasingly led to an improvement of the
diphosphorylated phosphonate product yields (30–40%).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NH
2
O
N
N
NH
N
O
P
O
P
O
P
O
P
O
P
O
P
HO
O
O
O
N
HO
O
O
O
N
O
O
O
OH OH OH
OH OH OH
OH
OH
PMTApp
PMTTpp
Bz
tained using A protected nucleotides.
NH2
O
N
N
N
NH
N NH2
O
P
O
P
O
O
P
O
P
O
P
HO
O
O
P
O
N
O
HO
O
O
O
Table 1. Mass Analysis of PhoNA-DNA Chimeras Used for
O
O
a
OH OH OH
OH OH OH
Tm Studies.
OH
OH
PMTCpp
PMTGpp
ON
Sequence
Calc.
Obs.
ON 1
5’ꢀTTT TAA ATA TAAꢀ3’
3640.7
3640.6
Figure 2. Chemical structures of diphosphates of threosyl nucleoꢀ
side phosphonates (PMTNpp) serving as substrates for engineerꢀ
ing of a dedicated tPhoNA polymerase through Tgo directed
evolution.
ON 2
ON 3
5’ꢀTTA TAT TTA AAAꢀ3’
3640.7
5617.9
3640.9
5618.2
5’PꢀTGC ATG GCA CGG
CGC TAGꢀ3’
Hybridization Properties of tPhoNA and dPhoNA. Hetꢀ
eroduplex Stability with DNA. In order to evaluate the hybridiꢀ
zation properties of phosphonateꢀlinked modified oligonucleoꢀ
tides, the thermal stabilities of duplexes composed of PhoNAꢀ
DNA chimeras (ONs 1–14) and complementary strands were
evaluated by temperatureꢀdependent UV spectroscopy. The Tm
values of the corresponding unmodified duplexes were deterꢀ
mined for comparison.
ON 4
ON 5
ON 6
ON 7
ON 8
5’PꢀCTA GCG CCG TGC
5537.9
5457.0
5537.0
5471.0
5520.0
5538.1
5457.0
5537.0
5471.0
5520.0
CAT GCAꢀ3’
5’ꢀCTA GCG CCG TAT CAT
GCAꢀ3’
5’ꢀTGC ATG ATA CGG CGC
TAGꢀ3’
5’ꢀCTA GCG ATT TAT CAT
Generally, the introduction of a 3’ꢀOꢀphosphonomethylꢀ
threosyl unit into a DNA sequence resulted in a considerable
destabilization of the DNA/DNA duplex relative to the correꢀ
GCAꢀ3’
5’ꢀTGC ATG ATA AAT CGC
TAGꢀ3’
sponding unmodified duplex with ∆T values ranging from ꢀ
m
ON 9
ON 10
ON 11
ON 12
ON 13
ON 14
5’ꢀTTA TAT TTA AAAꢀ3’
5’ꢀTTT TAA ATA TAAꢀ3’
5’ꢀTTT TAA aTA TAAꢀ3’
5’ꢀTTA TAt TTA AAAꢀ3’
5’ꢀTtt taa ata taaꢀ3’
3640.7
3640.7
3654.7
3654.7
3794.8
3794.8
3640.7
3640.7
3654.8
3654.8
3794.9
3794.9
1.6 to ꢀ7.0 °C (Table 1, ONs 1–4). As the number of modificaꢀ
tions increased from one to three and six (ONs 5–8), the Tm
dropped within a range from ꢀ0.6 to ꢀ5.2 °C, while showing a
modest reduction in ∆Tm per modification (ꢀ0.2 to ꢀ1.3
°
C/mod). When 10 modifications were introduced to produce
a T/A 12 mer (ON 9 and 10), the T decreased below the
m
measurable threshold. In comparison, the introduction of a 5’ꢀ
Oꢀphosphonomethylꢀdeoxyribosyl unit containing an adenine
base (ON 11) resulted in a thermal stability similar to that of
the unmodified counterpart (∆Tm = +0.1 °C/mod.), although
the insertion of one 5’ꢀOꢀphosphonomethylꢀdeoxyribosyl unit
bearing a thymine base (ON 12) resulted in a considerable
destabilization of the DNA/DNA duplex (∆Tm = ꢀ2.4
5’ꢀTta tat tta aaaꢀ3’
a
Modified units are bolded and underlined. tPhoNA and dPhoNA
nucleotides are shown as blue capital and magenta lowercase
letters, respectively.
°
C/mod.). The fully modified 5’ꢀphosphonate 12ꢀmer oligoꢀ
mers (ON 13 and ON 14) showed no hybridization with a
complementary DNA strand.
Chemical Synthesis of Diphosphates of Threosyl Nucleoꢀ
side Phosphonates (PMTNpp). The four PMT diphosphates
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 15
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. T Data for Duplexes of tPhoNA and dPhoNA Containing Oligonucleotides with their DNA Complements.
m
tPhoNA
a
b
Duplex with DNA
ON 1
N° of modifications
Tm (°C)
31.2
26.9
66.5
71.2
64.2
67.4
55.0
56.0
< 5
∆T (°C)
∆T /mod (°C)
m
m
1
1
ꢀ 1.6
ꢀ 5.9
ꢀ 1.6
ꢀ 5.9
ON 2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ON 3
1
ꢀ 7.0
ꢀ 7.0
ON 4
1
ꢀ 2.3
ꢀ 2.3
ON 5
3
ꢀ 3.8
ꢀ 1.3
ON 6
3
ꢀ 0.6
ꢀ 0.2
ON 7
6
ꢀ 5.2
ꢀ 0.9
ON 8
6
ꢀ 2.0
ꢀ 0.3
ON 9
10
10
> ꢀ 27.8
> ꢀ 27.8
> ꢀ 2.8
> ꢀ 2.8
ON 10
< 5
dPhoNA
a
b
Duplex with DNA
ON 11
N° of modifications
Tm (°C)
32.9
30.4
< 5
∆T (°C)
∆T /mod (°C)
m
m
1
1
+ 0.1
ꢀ 2.4
+ 0.1
ꢀ 2.4
ON 12
ON 13
11
11
> ꢀ 27.8
> ꢀ 27.8
> ꢀ 2.5
> ꢀ 2.5
ON 14
< 5
dsDNA Reference Duplexes and Tm
Tm (°C)
Sequence
Sequence
Tm (°C)
5
3
’ꢀTTTTAAATATAAꢀ3’
’ꢀAAAATTTATATTꢀ5’
5’ꢀCTA GCG CCG TAT CAT GCAꢀ3’
3’ꢀGAT CGC GGC ATA GTA CGTꢀ5’
32.8
3.5
68.0
5
’pꢀTGC ATG GCA CGGꢀCGC TAGꢀ3’
5’ꢀCTA GCG ATT TAT CAT GCAꢀ3’
7
60.2
3
’ꢀACG TAC CGT GCC GCG ATCꢀp5’
3’ꢀGAT CGC TAA ATA GTA CGTꢀ5’
a
b
c
22
T measurements represent an average of at least two experiments. ꢁT versus dsDNA duplexes. Rejman et al. described
m
m
*
*
ꢁT /mod between ꢀ0.2 °C (T :dA ) and ꢀ2.4 °C (A :dT ) for homopolymers with a triple helix motif.
m
15
15
15
15
Heteroduplex Stability with RNA. Hybridization studies on
bilization of the DNA/DNA duplex (∆T = ꢀ7.2 °C and ꢀ7.5
m
tPhoNAꢀDNA chimeras with RNA demonstrated that the ON
°C, respectively). However, extension of the modifications
5
/RNA duplex (T = 65.1 °C, Table 3) was more stable relaꢀ
from one to three and six pairs (ON 5/ON 6 and ON 7/ON 8)
m
tive to its ON 5/DNA duplex (64.2 °C). As observed for the
did not lead to a steep decline in T . Instead, a similar reducꢀ
m
ON 5/RNA duplex, the ∆T was ꢀ0.8 °C for three modificaꢀ
tion in T was observed with ∆T values of ꢀ1.1 and ꢀ1.5 °C
m
m
m
tions introduced when compared to the corresponding
DNA/RNA duplex. Increasing the number of modifications
from three to six (ON 7, Table 3) resulted in a moderate deꢀ
cline in T from 54.8 to 53.7 °C with a ∆T value of ꢀ0.2 °C
per modification, respectively. On the contrary, the introducꢀ
tion of one pair of dPhoNA nucleotides (ON 11/ON 12) reꢀ
sulted in a moderate destabilization of the DNA/DNA duplex
(∆T = ꢀ0.9 °C), while the fully modified dPhoNA oligomer
m
m
m
per modification. On the contrary, the fully modified dPhoNA
oligomer (ON 14, Table 3) could form a duplex with a comꢀ
ON 14 could not form a duplex with its complementary strand.
The 10ꢀmer tPhoNA (ON 9/ON 10) displayed a broad melting
curve, suggesting that the melting process is not an allꢀorꢀnone
melting profile. The association between both complementary
oligonucleotides (ON 9/ON 10) is of undefined nature.
plementary strand of RNA (T = 29.0 °C) that was more stable
m
as compared to the DNA/RNA duplex (Tm = 19.6 °C) and
slightly less stable compared to the RNA/RNA duplex (T =
m
3
1.7 °C). Remarkably, when the sequence was inverted beꢀ
The above Tm data obtained for dPhoNA and dPhoNAꢀ
DNA chimeras with complementary strands of DNA and RNA
is generally in accordance with previous data reported by
tween both strands (ON 13, Table 3), the duplex with RNA
became less stable.
22,23
Homoduplex Formation of tPhoNA and dPhoNA. In addiꢀ
tion, in order to establish whether phosphonate oligonucleoꢀ
tides could form selfꢀcomplementary duplexes, the hybridizaꢀ
tion properties of PhoNAꢀDNA chimeras with complementary
PhoNAꢀDNA strands were studied. The results showed that
the introduction of one pair of phosphonate nucleotides (ON
Rosenberg et al.
However, in both cases (tPhoNAꢀDNA
and dPhoNAꢀDNA chimeras), a higher binding affinity for a
complementary RNA stand as opposed to a DNA strand was
observed. This could arise from the different conformations of
the resultant PhoNAꢀDNA/DNA and PhoNAꢀDNA/RNA
29,30
duplexes.
Based on experiments with T/A mixed sequencꢀ
1
/ON 2 and ON 3/ON 4, Table 4) led to a considerable destaꢀ
es, it is not clear whether tPhoNA could form stable homoduꢀ
ACS Paragon Plus Environment

Page 5 of 15
Journal of the American Chemical Society
plexes, although the properties of this association was not
analyzed.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 15
Table 3. T Data for Duplexes of tPhoNA and dPhoNA Containing Oligonucleotides with their RNA Complements.
m
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tPhoNA
a
Duplex with RNA
N° of modifications
Tm (°C)
ꢀ
Tm (°C)
ꢀ
Tm/mod (°C)
ꢀ 0.3
ON 5
ON 7
ON 9
ON 10
3
6
65.1
53.7
< 5
ꢀ 0.8 (b)
ꢀ1.1 (b)
ꢀ 0.2
10
10
> ꢀ 26.7 (c)
> ꢀ 26.7 (c)
> ꢀ 2.7
> ꢀ 2.7
< 5
dPhoNA
a
b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Duplex with RNA
N° of modifications
Tm (°C)
ꢀ
Tm (°C)
ꢀ
Tm/mod (°C)
ꢀ 0.6
ON 13
ON 14
11
11
12.7
29.0
ꢀ 6.4 (b)
+ 9.4 (b)
+ 0.85
DNA/RNA and dsRNA Reference Duplexes and T (°C)
m
Sequence
’ꢀd(TTTTAAATATAA)ꢀ5’
’ꢀr(AAAAUUUAUAUU)ꢀ5’
’ꢀd(TTA TAT TTA AAA)ꢀ3’
Tm (°C)
6.2
19.6
1.7
Sequence
Tm (°C)
5
5’ꢀd(CTA GCG CCG TAT CAT GCA)ꢀ3’
3’ꢀr(GAU CGC GGC AUA GUA CGU)ꢀ5’
5’ꢀd(CTA GCG ATT TAT CAT GCA)ꢀ3’
3’ꢀr(GAU CGC UAA AUA GUA CGU)ꢀ5’
1
65.9
3
5
54.8
3
5
’ꢀr(AAU AUA AAU UUU)ꢀ5’
’ꢀr(UUU UAA AUA UAA)ꢀ3’
3
3’ꢀr(AAA AUU UAU AUU)ꢀ5’
a
b
Tm measurements represent an average of at least two experiments. ꢁT versus DNA/RNA duplexes, (c) ꢁT versus RNA/RNA duplexꢀ
m
m
es.
Table 4. T Data for Duplexes of tPhoNA and dPhoNA Containing Oligonucleotides: Self-Hybridization of Chimeric
m
Duplexes Containing DNA.
tPhoNA
a
b
N° of modifications
Tm (°C)
ꢀ
Tm (°C)
ꢀ 7.2
ꢀ
ꢀ
Tm/mod (°C)
ꢀ 7.2
ON 1/ON 2
ON 3/ON 4
ON 5/ON 6
ON 7/ON 8
ON 9/ON 10
1
1
25.6
66.0
64.6
51.4
ꢀ 7.5
ꢀ 7.5
3
ꢀ 3.4
ꢀ 1.1
6
ꢀ 8.8
ꢀ 1.5
10
31.7 (c)
ꢀ 1.1
ꢀ 0.1
dPhoNA
a
b
N° of modifications
Tm (°C)
ꢀ
Tm (°C)
Tm/mod (°C)
ꢀ 0.9
ON 11/ON 12
ON 13/ON 14
a
1
31.9
< 5
ꢀ 0.9
11
> ꢀ 27.8
> ꢀ 2.5
b
c
T measurements represent an average of at least two experiments. ꢁT versus dsDNA. broad melting profile.
m
m
Establishing tPhoNA as a Xenobiotic Genetic Materi-
al. Many in vitro and in vivo applications can only be realꢀ
ized if XNA oligomers are proven to be viable genetic mateꢀ
rials, capable of storing and propagating genetic information
gorgonarius) DNA polymerase variant harboring four mutaꢀ
32
33
tions: V93Q, D141A, E143A, and A485L. Since archaeal
Bꢀfamily polymerases have been extensively engineered for
34
XNA synthesis and reverse transcription, we chose a stepꢀ
3
1
accessible by XNA polymerases. As PMTNpp were poor
substrates for commercial DNA polymerases (Figure S2),
and in view of the significant basal incorporation activity for
wise approach to polymerase engineering, testing previously
reported mutations and screening residues known to be imꢀ
plicated in XNA substrate specificity (summarized in Figure
S3).
2
0
PMTApp exhibited by Therminator (an A485L mutant of
the archaeal 9°N DNA polymerase), we decided to engineer
a tPhoNA synthase from the same starting enzyme previousꢀ
ly used for the directed evolution of HNA synthases (TgoT,
First, we introduced mutations known to improve sugarꢀ
4
modified XNA synthesis (L403P) and enhance polymerase
35
processivity (H147E). L403 is adjacent to the highly conꢀ
served polymerase A motif (D404 – P410) and the L403P
4
a Therminator homologue). TgoT is a Tgo (Thermococcus
ACS Paragon Plus Environment

Page 7 of 15
Journal of the American Chemical Society
mutation has been shown to improve incorporation of arabiꢀ
no (and fluoroarabino) nucleic acids. The H147E mutation
affects the probability of nascent strand switching between
polymerase and exonuclease active sites, reducing exonucleꢀ
ase and increasing primer extension activities. Both L403P
of this systematic engineering has been that every mutation
(apart from I521L) introduced to improve tPhoNA synthesis
resulted in reduced DNA synthesis capability (Figure S4). It
remains to be shown if this shift in substrate is the result of a
specialist DNA polymerase being converted into a generalist
XNA polymerase, or if Tgo_EPFLH engineering is demonꢀ
strating the first steps towards the engineering of orthogonalꢀ
ity.
4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
TgoT_P) and the double mutant (TgoT_EP: H147E, L403P)
displayed improved tPhoNA synthesis compared to TgoT,
and were capable of fullꢀlength synthesis (57 PMTpp incorꢀ
porations) from a mixedꢀsequence template (TempN, Table
S6) under forcing conditions (high polymerase concentraꢀ
tion, long extensions, supplemented with manganese), as
shown in Figure S3.
Having synthesized tPhoNA, we screened available reꢀ
verse transcriptases (RTs) that could synthesize DNA against
XNA templates. The previously described T.RT521 (TgoT:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
I521L) and a close homologue harboring equivalent mutaꢀ
We then sought to improve polymerase processivity by
optimizing nascent nucleic acid binding. Mutations to resiꢀ
due E664 on the polymerase thumb have been shown to be
tions K.RTꢀ521K (KOD DNA polymerase: V93E, D141A,
E143A, A485L, I521L, E664K) were both efficient tPhoNA
RTs capable of DNA synthesis from both DNA and RNA
primers (Figure 3B and Figure S6), despite the poor hybridiꢀ
zation observed between tPhoNA and DNA in the absence of
polymerase, akin to the previously observed reverse tranꢀ
3
6
3
essential for efficient RNA (E664K) and TNA (E664I)
synthesis as well as improved 2’ꢀ5’ backbone modified DNA
37
synthesis. Screening all possible 20 E664 variants for
tPhoNA synthesis identified E664H (TgoT_EPH) as a clear
improvement over TgoT_EP (Figure 3 and S5).
41
scription of glycerol nucleic acids.
Together, Tgo_EPFLH and both RTs tested enabled us to
transfer genetic information from DNA into tPhoNA and
recover that information back to DNA (Figure 3C), with an
Next, we introduced I521L, a mutation originally isolatꢀ
4
ed for XNA reverse transcription but shown to enhance
38
ꢀ3
XNA synthesis when added to some XNA synthetases. The
resulting mutant (TgoT_EPLH) showed modest gains in
activity for both PMTNpp (Figure S5) and dNTP incorporaꢀ
tion (Figure S4).
approximate aggregate error rate of 17ꢀ20 x 10 per incorpoꢀ
ration (K.RT521K and T.RT521, respectively – Figure S9); a
degree of fidelity compatible with the development of apꢀ
4,38
tamers and aptazymes based on this chemistry.
Finally, comparing recent KOD DNA polymerase terꢀ
Analysis of the errors (Figure S9), showed a significant
bias towards the misincorporation of PMTGpp (or misincorꢀ
poration of dCTP during RT). A similar effect has been
observed in TNA synthesis with T.RT521 and could be
minimized through the use of biased unnatural nucleotide
39
nary complex structures with natural (dATP) and unnatural
40
threosyl (tATP) substrates suggested that Tyr409 was not
wellꢀoriented to provide stabilizing stacking interactions
with the threosyl sugar of the incoming tATP nucleotide. We
speculated that larger hydrophobic moieties at L408 could
better position Y409 for stacking against threosyl sugars,
stabilizing PMTNpp binding and improving its incorporaꢀ
tion. Both mutations tested (L408I and L408F) increased
tPhoNA synthetase activity, especially L408F. The resulting
polymerase (TgoT_EPFLH: H147E, L403P, L408F, I521L,
E664H) was found to be an efficient tPhoNA synthetase
4
pools lower in the guanosine analogue.
(
Figure 3A), capable of synthesizing a 57ꢀmer (from TempN
template) in less than five minutes (Figure S7) using low
polymerase (0.2 µM) and manganese (100 µM) concentraꢀ
tions. In fact, fullꢀlength tPhoNA synthesis could be
achieved even in the absence of manganese (Figure S5). The
synthesized polymer showed higher mobility in denaturing
gels (Figure S7) and increased exonuclease resistance (Figꢀ
ure S8) compared to DNA. Enzymatic synthesis of tPhoNA
was confirmed by far infrared spectroscopy, where an obꢀ
served frequency change of the (PO ꢀ) νas from 1225ꢀ1240 to
2
ꢀ
1
1
205 cm appeared as one of several consistent markers for
the bond change from PꢀOꢀC in DNA to PꢀCꢀO in tPhoNA
(
Figure 3D and Figure S10).
TgoT_EPFLH retained some of the biases observed in
the parental polymerase, such as poor PMTTpp homopolyꢀ
meric incorporation (Figure S7A). In mixed sequences, two
sequential PMTTpp incorporations reproducibly led to polꢀ
ymerase stalling (Figure S7B) and were avoided from subseꢀ
quent experiments. The aggregate fidelity of PMTTpp incorꢀ
poration did not significantly differ from PMTApp incorpoꢀ
ration, which showed no stalling pattern, suggesting that the
stalling is due to problems accommodating consecutive
PMTTpp incorporations, rather than catastrophic loss of
fidelity (Figure S9A and S9C). An unintended consequence
Figure 3. tPhoNA as a genetic polymer. (A) Impact of the
systematic engineering of the starting polymerase (TgoT) for the
synthesis of tPhoNA under forcing conditions (Figure S5 for
details). Intermediates in the engineering process including
TgoT_P (TgoT: L403P), TgoT_EP (TgoT_P: H147E),
Tgo_EPH (TgoT_EP: E664H), TgoT_EPLH (TgoT_EPH:
I521L) as well as the isolated tPhoNA synthase TgoT_EPFLH
(TgoEPH: L408F) are shown. (B) Reverse transcription of
tPhoNA to DNA. Both T.RT521 and K.RT521K were capable
of templateꢀdependent DNA synthesis (green) against a chimerꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 15
ic tPhoNA template (20 nt DNA from the original synthesis
interpret the decrease of active thyA genes among the clones
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
primer with 39 PMT incorporations based on TempN2), which
is shown in red, from an RNA primer. (C) PCR amplification of
the tPhoNA RT products was only possible when both tPhoNA
synthesis and reverse transcription were carried out, confirming
that tPhoNA can store and propagate genetic information. (MW:
NEB Low molecular weight DNA ladder). (D) ATRꢀFTIR
transformed upon extending the phosphonate nucleotide
stretch as evidence of inefficient copying by E. coli polyꢀ
merases.
ꢀ
1
spectra of DNA (black) and tPhoNA (red) in the 1300ꢀ900 cm
ꢀ
range. Frequency changes of the (PO ) νas from 1225ꢀ1240 to
2
ꢀ
1
ꢀ
1205 cm , a shift of the (PO ) symmetric stretching mode (νs
)
2
ꢀ
1
ꢀ1
at 1085 cm , and the (CꢀO) ribose stretch at 1062 cm in DNA
to 1074 and 1051 cm in tPhoNA can be observed as well as
signal reductions at 1009 cm (assigned to a ribose ring mode)
and the characteristic riboseꢀphosphodiester skeletal motion
mode of DNA at 965 cm . Complete spectra are shown in Figꢀ
ꢀ
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ
1
ꢀ1
ure S10.
In Vivo PhoNA-Dependent DNA Synthesis. To evaluꢀ
ate the ability of phosphonate modified nucleic acids to serve
as template for DNA synthesis in E. coli, six 5’ꢀ
phosphorylated DNAꢀPhoNAꢀDNA chimera oligonucleoꢀ
tides (ON 4 and 15-19, Table S3) were synthesized and
tested using the established gappedꢀvector assay based on the
restoration of the active site of thymidylate synthase (Figure
Figure 4. (A) Selection screening of PhoNA containing oligoꢀ
nucleotides capable of templating DNA synthesis in E. coli
5,42
based on the restoration of the thyA gene. In vivo propagation
of DNAꢀtPhoNAꢀDNA (B) and DNAꢀdPhoNAꢀDNA (C) chiꢀ
meras. T* and I* indicate Phoꢀmodified units with a thymine
and hypoxanthine base, respectively. The normalized ratios
correspond to the experimentally derived average number of
5,42
4). This enzyme catalyzes the conversion of deoxyuridine
monophosphate to thymidine monophosphate, and is essenꢀ
tial for E. coli growth in a medium lacking either thymine or
thymidine (dt). The six 18ꢀbase long 5’ꢀphosphorylated
DNAꢀPhoNAꢀDNA chimera oligonucleotides were ligated
into a gapped heteroduplex plasmid where 6 codons surꢀ
rounding the catalytic Cys146 of the thyA gene had been
deleted. The resulting ligation products were then transꢀ
formed into an E. coli strain lacking thyA. Transformants are
only able to survive in thymidineless media when a PhoNA
oligonucleotide chimera is recognized by the bacterial repliꢀ
cation machinery and utilized as template for DNA syntheꢀ
sis, restoring the thymidylate synthase active site. The ratio
between bacterial colony numbers in media with or without
thymidine indicates the extent of successful templating.
+
thymidine prototrophic colonies (thyA ) from the average total
+
number of colonies (bla ).
CONCLUSION
For in vivo applications, it is desirable that XNAs are
both chemically and biologically orthogonal: that is that
neither the polymers nor the building blocks interact with
natural nucleic acids or proteins, and that XNAꢀsynthesizing
and binding proteins do not synthesize or bind natural nucleꢀ
ic acids. Significant progress has been made in recent years
in describing XNA oligomers and establishing them as geꢀ
netic polymers through engineered polymerases that can
synthesize and reverse transcribe to transfer genetic inforꢀ
mation from DNA to XNA and back. However, most of
these XNAs retain significant affinity for DNA and RNA
and substantial efficacy in in vivo transliteration experiꢀ
ments. Furthermore, all engineered XNA polymerases so far
described also retain substantial DNA polymerase capability.
As expected, ligation with no oligonucleotide or of a
Cys146ꢀdeleted oligonucleotide did not yield any protoꢀ
trophic transformants. The replacement of a DNA unit by
either a tPhoNA or dPhoNA building block (ON 4 and 17,
respectively) yielded a similar 2.5ꢀfold decrease of the numꢀ
ber of prototrophic transformants compared to the positive
controls (Tables S4 and S5, Figure 4B and 4C, respectively).
A further 2ꢀ and 6.5ꢀfold drop in the yield of prototrophic
transformants resulted from extension of the tPhoNA stretch
from one to two (ON 15) and three (ON 16) oligonucleoꢀ
tides, respectively. Thus, the successive addition of tPhoNA
nucleotides significantly diminished DNA propagation in
vivo. Even more appreciable yield drops were observed upon
extension of the 2’ꢀdeoxyriboseꢀ5’ꢀphosphonate stretch from
one to two oligonucleotides (ON 18, 76ꢀfold decrease) and
from one to three oligonucleotides (ON 19, 217ꢀfold deꢀ
crease). These results suggest that the dPhoNA backbone
modification caused less productive in vivo transliteration
than tPhoNA by E. coli DNA polymerases. This is presumaꢀ
bly due to the difference in length of the internucleotide
linkage between tPhoNA (6 bonds per repeating unit, in
analogy to DNA) and dPhoNA (7 bonds per repeating unit),
as depicted in Figure 1. Since replacement of guanine in the
Cys146 codon (TGC/T) by an hypoxanthine is well tolerated
Thus, we set out to evaluate for the first time chemically
modified nucleic acids featuring a phosphonate internucleoꢀ
tide linkage for their potential as orthogonal genetic materiꢀ
als. Specifically, 3’–2’ phosphonomethylꢀthreosyl (tPhoNA)
oligonucleotides, along with 5’–3’ phosphonomethyl–
deoxyribosyl (dPhoNA) sequences as well as their DNA
chimeras were conveniently prepared by Hꢀphosphinate and
phosphoramidate solidꢀphase syntheses upon careful adaptaꢀ
tion of the synthesis cycle conditions.
Both chemistries showed significant levels of orthogoꢀ
nality at both chemical (oligonucleotide properties) and
biological (recognition by natural protein) levels. Our meltꢀ
ing analyses demonstrated that heavily modified tPhoNA
and dPhoNA (ON 9/10 and 13/14, respectively) did not
exhibit detectable hybridization to complementary DNA
(and, for tPhoNA, RNA) strands. Significantly, tPhoNA
retained some potential to form homoduplexes (ON 9/10,
Table 4), at least for ATꢀrich sequences.
tPhoNA also showed signs of biological orthogonality.
PMTNpps were demonstrably poor substrates for natural
polymerases (Figure S1), yet tPhoNA could be efficiently
5
and does not interfere with the correct copying of DNA, we
ACS Paragon Plus Environment

Page 9 of 15
Journal of the American Chemical Society
concentration was determined by measuring the absorbance
synthesized by the engineered polymerase TgoT_EPFLH
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(Figure 3). Importantly, as we engineered TgoT for better
tPhoNA synthesis, we observed a noticeable drop in its DNA
synthetase function (Figures S3 and S4). Given the orthogoꢀ
nality demonstrated both chemically (oligonucleotide anꢀ
nealing) and in vivo (transliteration), it is likely that
TgoT_EPFLH’s broadened substrate specificity can be furꢀ
ther engineered to develop an orthogonal polymerase.
in MilliꢀQ water at 260 nm and 80 °C, and assuming that the
phosphonate nucleosides have the same extinction coeffiꢀ
cients per base moiety in the denatured state as their natural
nucleoside counterparts (A, ε = 15060; T, ε = 8560; C, ε =
7100; G, ε = 12180; T, ε = 8560; I, ε = 7500; U, ε = 9660).
For all experiments, the concentration for each strand was 4
ꢂM. Melting curves were determined with a Varian Cary 100
BIO spectrophotometer. Cuvettes were maintained at a conꢀ
stant temperature by water circulation through the cuvette
holder. The temperature of the solution was measured with a
thermistor that was directly immersed in the cuvette. Temꢀ
perature control and data acquisition were carried out autoꢀ
matically with an IBMꢀcompatible computer by using the
Cary WinUV thermal application software. A quick heating
and cooling cycle was carried out to allow proper annealing
Together, these data suggest that a fully orthogonal geꢀ
netic system based on tPhoNA and specialist tPhoNA polyꢀ
merases, and with minimal interaction with natural dNTPs,
nucleic acids or polymerases, is achievable.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
METHODS
General Chemistry Details. Detailed procedures for the
synthesis of threosyl and deoxyribosyl nucleoside Hꢀ
phosphinates as well as diphosphates of threosyl nucleoside
phosphonates are described in detail in the Supporting Inꢀ
formation. All reactions were carried out under an argon
atmosphere using ovenꢀdried glassware. Chemical reagents
were purchased from commercial sources and used as reꢀ
of both strands. The samples were then heated from 10 to 80
ꢀ1
°
C at a rate of 0.2 °C min and were cooled again at the
same speed. Melting temperatures were determined by plotꢀ
ting the first derivative of the absorbance as a function of
temperature; data plotted were the average of two runs.
Generally, up and down curves showed identical T values.
1
13
31
m
ceived. H, C, and P spectra were recorded on a 300, 500,
or 600 MHz Bruker Avance spectrometer. 2D NMR experiꢀ
ments (HꢀCOSY, HSQC, and HMBC) were used for the
characterization of all the intermediates and final comꢀ
pounds. High resolution mass spectra were recorded on a
quadrupole orthogonal acceleration timeꢀofꢀflight mass
spectrometer (Synapt G2 HDMS, Waters, Milford, MA).
Preparative HPLC purification was performed on a Phenomꢀ
enex Gemini 110A column (C18, 10 ꢂm, 21.2 mm × 250
mm). Column chromatography was performed using silica
gel 60 Å, 0.035−0.070 mm (Acros Organics).
Primer Extensions. Primer extension reactions were
typically carried out as 8 µl reactions with 0.2 µM primer
tag01F3, 0.4 µM template TempN or TempNv3, 50 µM each
PMTNpp, 0.2 µM TgoT_EPFLH in 1x Thermopol buffer.
MnCl was added to a final concentration of 0.6 mM for
2
earlier polymerase variants. A typical thermal cycling proꢀ
gram was 1 min at 94 °C, 5 min at 50 °C, and 2 h at 65 °C.
For time courses, reaction mixtures were made up without
polymerase and annealed by heating to 95 °C for 5 min,
cooling to 4 °C at 0.1°C/s and held at 4 °C for 3ꢀ5 min. Polꢀ
ymerase was then added to the desired final concentration,
typically from 20x stocks (i.e., 5 µl 10 µM polymerase stock
added to 95 µl reaction mix for 0.5 µM final concentration).
Reactions were aliquoted to 6 µl reactions in individual 0.2
ml tubes. These were placed simultaneously on a PCR block
preheated to either 50 or 65 °C and tubes were removed from
the heat and immediately quenched by addition of two volꢀ
umes of 98% formamide and 10 mM EDTA at the desired
timepoints. Primer extensions were analyzed by denaturing
polyacrylamide gel electrophoresis (8 M urea, 16% acrylaꢀ
mide, 1x TBE). Large scale syntheses and purification are
described in the Supplementary Information.
Solid-Phase Synthesis. The solidꢀphase synthesis of
PhoNA oligonucleotides as well as PhoNAꢀDNA chimeras
was accomplished with an Expedite DNA synthesizer (Apꢀ
plied Biosystems) by using either an adapted Hꢀphosphinate
approach or its combination with the standard phosphoꢀ
ramidite method (Scheme S3, Table S1). After synthesis, the
support was dried in vacuo, treated with a freshly prepared
thiophenolꢀEt NꢀDMF mixture (23:32:45; v/v) for 4 h to
3
remove the methyl protecting groups at the phosphonate
linkages, rinsed with dry acetonitrile, and further dried in
vacuo. The oligomers were deprotected and cleaved from the
solid support by treatment with aqueous ammonia (33%)
overnight. After gel filtration on a NAPꢀ25 column (Seꢀ
phadex G25ꢀDNA grade; Pharmacia) using water as eluent,
the crude mixture was analyzed through a MonoꢀQ HR 5/5
anion exchange column, after which purification was
achieved by using a MonoꢀQ HR 10/100 GL column (Pharꢀ
macia) with the following gradient systems: A = 20 mM
TrisꢀHCl, 10 mM NaClO in 15% CH CN, pH 7.4; B = 20
tPhoNA Reverse Transcription and PCR. Reverse
transcription reactions for sequencing consisted of 1x Therꢀ
mopol buffer (NEB), 0.25 µM primer outnest2_test7, 100
µM each dNTP, 30 nM T.RT521 or K.RT521K. A masꢀ
termix without template or polymerase was made up and
divided into 17 µl aliquots. 2 µl gelꢀpurified tPhoNA was
added and primers annealed by heating to 5 min for 95 °C
and cooling to 4 °C at 0.1 °C/s. After 3ꢀ5 min at 4 °C, 1 µl
600 nM polymerase was added (final concentration 30 nM)
and incubated for 30 s at 50 °C, followed by 2 h at 65 °C.
For RT gels, conditions were the same except primer
IR700_test7 and template TempN2 were used and incubation
was 5 min at 50 °C followed by 1 h at 65 °C.
4
3
mM TrisꢀHCl, 600 mM NaClO in 15% CH CN, pH 7.4.
4
3
The lowꢀpressure liquid chromatography system consisted of
a Merck–Hitachi Lꢀ6200A intelligent pump, MonoꢀQ HR
10/100 GL column (Pharmacia), Uvicord SII 2138 UV deꢀ
tector (PharmaciaꢀLKB), and recorder. The productꢀ
containing fraction was desalted on a NAPꢀ25 column and
lyophilized. The presence of the desired product was conꢀ
firmed by mass spectrometry.
After incubation, excess primers were removed by addꢀ
ing 0.5 µl (10U) exonuclease I (NEB) and incubating at 37
°
C for 30 min. Reactions were then purified using Monarch
UV Melting Experiments. Oligomers were dissolved in
a buffer solution containing NaCl (0.1 or 1 M), potassium
phosphate (0.02 M, pH 7.5), and EDTA (0.1 mM). The
PCR & DNA Cleanup kits (NEB), following the protocol for
short PCR products.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 15
PCR amplification was carried out using MyTaq HS in a 1x DNA ligase T4 reaction buffer (NEB) for 20 ꢂL per
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DNA Polymerase (Bioline, London, UK) using primers
outnest1 and outnest2 in reactions consisting 1x MyTaq
buffer, 0.4 µM each primer, and 0.025 U/µl MyTaq DNA
Polymerase. PCR reactions were typically run for 1 min at
8 °C, then 18 cycles of 15 s at 98 °C, 15 s at 55 °C, and 10
s at 72 °C, with a final 1 min at 68 °C polishing step.
sample. The mixture was then denaturated at 85 °C and
cooled as before. Ligation was performed by adding 1 mM
ATP and 400 U T4 DNA ligase (NEB) to the samples before
overnight incubation at 16 °C. The ligated mixtures were
dialyzed as before, and transformed by electroporation into a
fresh electrocompetent E. coli K12 strain (ꢁthyA:aad). Incuꢀ
bation of the electroporated bacteria was performed at 37 °C
for 1 h, before plating 100 ꢂL of a serial dilution of the susꢀ
pension onto a MullerꢀHinton (MH) media containing 100
9
Fourier Transform Infrared (FTIR) Spectroscopy.
tPhoNA chimeras were synthesized in reactions consisting of
x Thermopol buffer, 1 µM primer TagBspQI, 2 µM temꢀ
plate TempN_MS2nick, 50 µM each PMTNpp, 0.1 mM
1
−1
0
−1
ꢂg mL ampicillin (spreading the 10 and 10 dilutions) and
onto the same media additionally supplemented with 0.3 mM
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
MnCl , 1.3 µM TgotT_EPFLH and incubated in 24 x 40 µl
2
−3
−4
thymidine (10 and 10 dilutions).
aliquots in 0.2 ml 8ꢀwell PCR strips for 2 min at 94 °C, 10
min at 50 °C, and 4 h at 65 °C. DNA controls were syntheꢀ
sized in reactions consisting 1x MyTaq buffer, 1 µM primer
TagBspQI, 1.8 µM template TempN_MS2nick, 0.01 U/µl
My Taq HS DNA Polymerase (Bioline) and incubated for 2
min at 98 °C and 5 cycles of 15 s at 98 °C, 15 s at 50 °C, and
ASSOCIATED CONTENT
Supporting Information
The supporting information is available free of charge on
the ACS Publications websites at DOI:
1
5 s at 72 °C. Purification was carried out as described in
Detailed synthetic procedures, characterization inforꢀ
mation, NMR spectra, supplementary methods, figures, and
references.
Supplementary Information for the largeꢀscale tPhoNA
synthesis and purification, except that the templates were 3'ꢀ
biotinylated, meaning beads were washed in 2 x 100 µl 30
mM NaOH and the eluate contained the synthesized strand.
An equivalent volume of formamide/EDTA loading buffer
was added and the whole sample was loaded directly onto
gels for purification. Yield was 287 pmol for tPhoNA and
192 pmol for DNA. Samples were washed 5 times in water
using Vivaspin 500 centrifugal concentrators (Sartorius
Stedim Biotech GmbH, Germany) with 10 kD MWCO.
AUTHOR INFORMATION
Corresponding Author
*Eꢀmail:
piet.herdewijn@kuleuven.be;
v.pinheiro@ucl.ac.uk
Notes
The authors declare no competing financial interest.
Next, 10 µL of the resulting 10 µM solution of DNA or
tPhoNA (prepared in double distilled water) were placed on
a 3 mm diamond prism with KRS5 optics (1ꢀreflection Duꢀ
raSampIIR II, SensIR/Smith Detection). After drying with a
ACKNOWLEDGEMENTS
C.L. acknowledges the China Scholarship Council (CSC) for
funding (grant 201306220065). The research leading to these
results has received funding from the European Research Counꢀ
cil under the European Union’s Seventh Framework Program
stream of N gas, IR spectra were recorded at 295 K and 4
2
ꢀ
1
cm resolution in attenuated total reflectance (ATR) mode
on a Bruker Vertex 80v spectrometer equipped with a liquid
nitrogenꢀcooled MCTꢀC detector and a KBr beamsplitter.
The optics compartment was kept under vacuum (<2 hPa)
during data acquisition. All frequencies cited have an accuꢀ
(
2
(
FP7/2007–2013)/ERC (Grant agreement no ERCꢀ2012ꢀADG
0120216/ 320683). We also wish to thank FWO Vlaanderen
G0A8816N) and ERASynBio (in vivo XNA) for financial
support. Mass spectrometry was conducted with the support of
the Hercules Foundation of the Flemish Government (grant
20100225ꢀ7). We wish to express our gratitude to Guy
Schepers, Luc Baudemprez, and Lia Margamuljana for excellent
technical assistance. V.B.P. and C.C. were supported by the UK
Biotechnology and Biosciences Research Council (grant number
BB/N01023X/1). V.B.P. was also supported by the European
Research Council [ERCꢀ2013ꢀStG project 336936 (HNAepiꢀ
some)]. A.M. was supported by the Medical Research Council
UK (Career Development Award MR/M00936X/1).
ꢀ
1
racy of ± 1 cm . For each IR spectrum recorded, 500 interꢀ
ferograms were averaged before Fourier transformation. A
reference spectrum of the clean ATR prism was recorded
and used as a background for the sample spectra.
In Vivo Propagation Assay of DNA-PhoNA-DNA
Chimeras. Oligonucleotides were dissolved in MilliꢀQ water
to reach 1 mM concentration and then diluted tenꢀfold before
the assay. Oligonucleotides were tested inside a gapped
heteroduplex generated through enzymatic modification,
denaturation, and hybridization of the ampicillin resistance
4
3
gene containing pAK1 and 2 plasmids. The form of this
5,42
REFERENCES
heteroduplex is described in details elsewhere.
A mix of
equimolar (25 ng each) purified NheI and NsiI cut pAK1,
purified EcoRI cut, and dephosphorylated pAK 2 were dilutꢀ
ed in 10 mM TrisꢀHCl pH 7.5 with 100 mM NaCl. The
mixture was denaturated at 95 °C for 5 min before cooling to
ambient temperature over 2 h to achieve hybridization, folꢀ
lowed by water dialysis through a 0.05 ꢂM nitrocellulose
filter (Millipore) for 30 min. The oligonucleotides (100
pmoles) as well as a positive control (oligonucleotide with
an intact catalytic residue codon) and two negative controls
(1) Herdewijn, P.; Marliere, P. Chem. Biodiv. 2009, 6, 791ꢀ808.
(2) Chaput, J. C.; Yu, H. Y.; Zhang, S. Chem. Biol. 2012, 19, 1360ꢀ
1
371.
(3) Larsen, A. C.; Dunn, M. R.; Hatch, A.; Sau, S. P.; Youngbull, C.;
Chaput, J. C. Nat. Commun. 2016, 7, 11235.
(
4) Pinheiro, V. B.; Taylor, A. I.; Cozens, C.; Abramov, M.; Renders,
M.; Zhang, S.; Chaput, J. C.; Wengel, J.; PeakꢀChew, S. Y.;
McLaughlin, S. H.; Herdewijn, P.; Holliger, P. Science 2012, 336,
3
41ꢀ344.
(5) Pezo, V.; Liu, F. W.; Abramov, M.; Froeyen, M.; Herdewijn, P.;
Marliere, P. Angew. Chem. Int. Ed. 2013, 52, 8139ꢀ8143.
(
water and oligomer with a deleted catalytic residue codon)
were added separately to the dialyzed heteroduplex mixture
(
6) Bande, O.; Abu El Asrar, R.; Braddick, D.; Dumbre, S.; Pezo, V.;
Schepers, G.; Pinheiro, V. B.; Lescrinier, E.; Holliger, P.;
Marliere, P.; Herdewijn, P. Chem. Eur. J. 2015, 21, 5009ꢀ5022.
ACS Paragon Plus Environment

Page 11 of 15
Journal of the American Chemical Society
(7) Bande, O.; Braddick, D.; Agnello, S.; Jang, M.; Pezo, V.;
(25) Liu, C.; Dumbre, S. G.; Pannecouque, C.; Korba, B.; De Jonghe,
Schepers, G.; Rozenski, J.; Lescrinier, E.; Marlière, P.; Herdewijn,
P. Chem. Sci. 2016, 7, 995ꢀ1010.
8) Pezo, V.; Schepers, G.; Lambertucci, C.; Marlière, P.; Herdewijn,
S.; Herdewijn, P. Org. Biomol. Chem. 2017, 15, 5513ꢀ5528.
(26) Kostov, O.; Pav, O.; Budesinsky, M.; Liboska, R.; Simak, O.;
Petrova, M.; Novak, P.; Rosenberg, I. Org. Lett. 2016, 18, 2704ꢀ
2707.
(27) Wu, T. F.; Froeyen, M.; Kempeneers, V.; Pannecouque, C.; Wang,
J.; Busson, R.; De Clercq, E.; Herdewijn, P. J. Am. Chem. Soc.
2005, 127, 5056ꢀ5065.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
(
(
P. ChemBioChem 2014, 15, 2255ꢀ2258.
9) Inoue, N.; Shionoya, A.; Minakawa, N.; Kawakami, A.; Ogawa,
N.; Matsuda, A. J. Am. Chem. Soc. 2007, 129, 15424ꢀ15425.
10) Marlière, P.; Patrouix, J.; Döring, V.; Herdewijn, P.; Tricot, S.;
Cruveiller, S.; Bouzon, M.; Mutzel, R. Angew. Chem. Int. Ed.
2011, 50, 7109ꢀ7114.
11) Mehta, A. P.; Li, H.; Reed, S. A.; Supekova, L.; Javahishvili, T.;
Schultz, P. G. J. Am. Chem. Soc. 2016, 138, 14230ꢀ14233.
12) Eremeeva, E.; Abramov, M.; Margamuljana, L.; Rozenski, J.;
Pezo, V.; Marliere, P.; Herdewijn, P. Angew. Chem. Int. Ed. 2016,
(28) Moffatt, J. G.; Khorana, H. G. J. Am. Chem. Soc. 1961, 83, 649ꢀ
658.
(
(29) Lesnik, E. A.; Freier, S. M. Biochemistry 1995, 34, 10807ꢀ10815.
(30) Pav, O.; Kosiova, I.; Barvik, I.; Pohl, R.; Budesinsky, M.;
Rosenberg, I. Org. Biomol. Chem. 2011, 9, 6120ꢀ6126.
(31) Pinheiro, V. B.; Loakes, D.; Holliger, P. Bioessays 2013, 35, 113ꢀ
122.
(32) Fogg, M. J.; Pearl, L. H.; Connolly, B. A. Nat. Struct. Biol. 2002,
9, 922ꢀ927.
(33) Gardner, A. F.; Jack, W. E. Nucleic Acids Res. 2002, 30, 605ꢀ613.
(34) Houlihan, G.; ArangundyꢀFranklin, S.; Holliger, P. Curr. Opin.
Biotechnol. 2017, 48, 168ꢀ179.
(35) Kuroita, T.; Matsumura, H.; Yokota, N.; Kitabayashi, M.;
Hashimoto, H.; Inoue, T.; Imanaka, T.; Kai, Y. J. Mol. Biol. 2005,
351, 291ꢀ298.
(36) Cozens, C.; Pinheiro, V. B.; Vaisman, A.; Woodgate, R.; Holliger,
P. Proc. Nat. Acad. Sci. U S A 2012, 109, 8067ꢀ8072.
(37) Cozens, C.; Mutschler, H.; Nelson, G. M.; Houlihan, G.; Taylor,
A. I.; Holliger, P. Angew. Chem. Int. Ed. 2015, 54, 15570ꢀ15573.
(38) Taylor, A. I.; Pinheiro, V. B.; Smola, M. J.; Morgunov, A. S.;
PeakꢀChew, S.; Cozens, C.; Weeks, K. M.; Herdewijn, P.;
Holliger, P. Nature 2015, 518, 427ꢀ430.
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
5, 7515ꢀ7519.
(
13) Zhang, Y.; Ptacin, J. L.; Fischer, E. C.; Aerni, H. R.; Caffaro, C.
E.; San Jose, K.; Feldman, A. W.; Turner, C. R.; Romesberg, F. E.
Nature 2017, 551, 644ꢀ647.
14) Schoning, K. U.; Scholz, P.; Guntha, S.; Wu, X.; Krishnamurthy,
R.; Eschenmoser, A. Science 2000, 290, 1347ꢀ1351.
15) Yu, H. Y.; Zhang, S.; Chaput, J. C. Nat. Chem. 2012, 4, 183ꢀ187.
16) Schoning, K. U.; Scholz, P.; Wu, X. L.; Guntha, S.; Delgado, G.;
Krishnamurthy, R.; Eschenmoser, A. Helv. Chim. Acta 2002, 85,
(
(
(
4
111ꢀ4153.
(
17) Ichida, J. K.; Horhota, A.; Zou, K. Y.; McLaughlin, L. W.;
Szostak, J. W. Nucleic Acids Res. 2005, 33, 5219ꢀ5225.
(18) Yu, H. Y.; Zhang, S.; Dunn, M. R.; Chaput, J. C. J. Am. Chem.
Soc. 2013, 135, 3583ꢀ3591.
19) Dunn, M. R.; Chaput, J. C. ChemBioChem 2016, 17, 1804ꢀ1808.
20) Renders, M.; Emmerechts, G.; Rozenski, J.; Krecmerova, M.;
Holy, A.; Herdewijn, P. Angew. Chem. Int. Ed. 2007, 46, 2501ꢀ
(
(
(39) Kropp, H. M.; Betz, K.; Wirth, J.; Diederichs, K.; Marx, A. PLoS
One 2017, 12, 1ꢀ20.
(40) Chim, N.; Shi, C. H.; Sau, S. P.; Nikoomanzar, A.; Chaput, J. C.
Nat. Commun. 2017, 8, 1810.
(41) Tsai, C. H.; Chen, J. Y.; Szostak, J. W. Proc. Nat. Acad. Sci. U S
A 2007, 104, 14598ꢀ14603.
(42) Pochet, S.; Kaminski, P. A.; Van Aerschot, A.; Herdewijn, P.;
Marliere, P. C. R. Biol. 2003, 326, 1175ꢀ1184.
2
504.
(
21) Renders, M.; Lievrouw, R.; Krecmerova, M.; Holy, A.;
Herdewijn, P. ChemBioChem 2008, 9, 2883ꢀ2888.
(22) Rejman, D.; Snasel, J.; Liboska, R.; Tocik, Z.; Paces, O.;
Kralikova, S.; Rinnova, M.; Kois, P.; Rosenberg, I. Nucleosides
Nucleotides Nucleic Acids 2001, 20, 819ꢀ823.
23) Sipova, H.; Springer, T.; Rejman, D.; Simak, O.; Petrova, M.;
Novak, P.; Rosenbergova, S.; Pav, O.; Liboska, R.; Barvik, I.;
Stepanek, J.; Rosenberg, I.; Homola, J. Nucleic Acids Res. 2014,
(
(43) Tchou, J.; Kasai, H.; Shibutani, S.; Chung, M. H.; Laval, J.;
Grollman, A. P.; Nishimura, S. Proc. Natl. Acad. Sci. U. S. A.
1991, 88, 4690ꢀ4694.
4
2, 5378ꢀ5389.
(
24) Liu, C.; Dumbre, S. G.; Pannecouque, C.; Huang, C. S.; Ptak, R.
G.; Murray, M. G.; De Jonghe, S.; Herdewijn, P. J. Med. Chem.
2016, 59, 9513ꢀ9531.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 15
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table of Contents
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
12
ACS Paragon Plus Environment

Page 13 of 15
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
tPhoNA as a genetic polymer. (A) Impact of the systematic engineering of the starting polymerase (TgoT)
for the synthesis of tPhoNA under forcing conditions (Figure S5 for details). Intermediates in the engineering
process including TgoT_P (TgoT: L403P), TgoT_EP (TgoT_P: H147E), Tgo_EPH (TgoT_EP: E664H),
TgoT_EPLH (TgoT_EPH: I521L) as well as the isolated tPhoNA synthase TgoT_EPFLH (TgoEPH: L408F) are
shown. (B) Reverse transcription of tPhoNA to DNA. Both T.RT521 and K.RT521K were capable of templateꢀ
dependent DNA synthesis (green) against a chimeric tPhoNA template (20 nt DNA from the original
synthesis primer with 39 PMT incorꢀporations based on TempN2), which is shown in red, from an RNA
primer. (C) PCR amplification of the tPhoNA RT products was only possible when both tPhoNA synthesis and
reverse transcription were carried out, confirming that tPhoNA can store and propagate genetic information.
(
MW: NEB Low molecular weight DNA ladder). (D) ATRꢀFTIR spectra of DNA (black) and tPhoNA (red) in the
ꢀ
1
ꢀ
ꢀ1
1300ꢀ900 cm range. Frequency changes of the (PO
2
) νas from 1225ꢀ1240 to 1205 cm , a shift of the
sup>ꢀ
ꢀ1
ꢀ1
(PO
1
2
) symmetric stretching mode (ν
s
) at 1085 cm , and the (CꢀO) ribose stretch at 1062 cm in DNA to
ꢀ
1
ꢀ1
074 and 1051 cm in tPhoNA can be observed as well as signal reductions at 1009 cm (assigned to a
ꢀ
1
ribose ring mode) and the characteristic riboseꢀphosphodiester skeletal motion mode of DNA at 965 cm .
Comꢀplete spectra are shown in Figure S10.
3
75x355mm (168 x 168 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 15
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
A) Selection screening of PhoNA containing oligonucleotides capable of templating DNA synthesis in E. coli
based on the restoration of the thyA gene.5,42 In vivo propagation of DNA-tPhoNA-DNA (B) and DNA-
dPhoNA-DNA (C) chimeras. T* and I* indicate Pho-modified units with a thymine and hypoxan-thine base,
respectively. The normalized ratios correspond to the experimentally derived average number of thymidine
prototrophic colonies (thyA+) from the average total number of colonies (bla+).
3
78x179mm (96 x 96 DPI)
ACS Paragon Plus Environment

Page 15 of 15
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TOC
16x7mm (300 x 300 DPI)
ACS Paragon Plus Environment