Synthesis and properties of GuNA purine/pyrimidine nucleosides and oligonucleotides

Article abstract of DOI:10.1039/d0ob01970d

We recently designed guanidine-bridged nucleic acids (GuNA), and GuNA bearing a thymine (T) nucleobase was synthesized and successfully incorporated into oligonucleotides. The GuNA-T-modified oligonucleotides possessed high duplex-forming ability towards their complementary single-stranded RNAs and were highly stable against 3-exonuclease. Therefore, GuNA is a promissing artificial nucleic acid for therapeutic antisense oligonucleotides. We herein report the facile synthesis of GuNA phosphoramidites bearing adenine (A), guanine (G), and 5-methylcytosine (mC) nucleobases and a robust method for the preparation of GuNA-modified oligonucleotides, even with sequences having acid-sensitive purine nucleobases. Oligonucleotides modified with GuNA-A,-G, or-mC possessed high duplex-forming ability, similar to those modified with GuNA-T. Moreover, some of the GuNA-modified oligonucleotides were revealed to have high base discriminating ability compared with that of their natural counterparts. GuNA nucleosides exhibited no genotoxicity in bacterial reverse mutation assays. Thus, all GuNAs (GuNA-T,-A,-G, and-mC) are now available to be examined in therapeutic applications. This journal is

Full text of DOI:10.1039/d0ob01970d

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Synthesis and properties of GuNA purine/
pyrimidine nucleosides and oligonucleotides†
Cite this: DOI: 10.1039/d0ob01970d
Shinji Kumagai, a Hiroaki Sawamoto,a Tomo Takegawa-Araki,a Yuuki Arai,a
a
Shuhei Yamakoshi,a Katsuya Yamada,a Tetsuya Ohta,a Eiji Kawanishi,
b,c
Naohiro Horie,b Takao Yamaguchi b and Satoshi Obika
*
We recently designed guanidine-bridged nucleic acids (GuNA), and GuNA bearing a thymine (T) nucleo-
base was synthesized and successfully incorporated into oligonucleotides. The GuNA-T-modified oligo-
nucleotides possessed high duplex-forming ability towards their complementary single-stranded RNAs
and were highly stable against 3’-exonuclease. Therefore, GuNA is a promissing artificial nucleic acid for
therapeutic antisense oligonucleotides. We herein report the facile synthesis of GuNA phosphoramidites
bearing adenine (A), guanine (G), and 5-methylcytosine (mC) nucleobases and a robust method for the
preparation of GuNA-modified oligonucleotides, even with sequences having acid-sensitive purine
nucleobases. Oligonucleotides modified with GuNA-A, -G, or -mC possessed high duplex-forming ability,
similar to those modified with GuNA-T. Moreover, some of the GuNA-modified oligonucleotides were
revealed to have high base discriminating ability compared with that of their natural counterparts. GuNA
nucleosides exhibited no genotoxicity in bacterial reverse mutation assays. Thus, all GuNAs (GuNA-T, -A,
-G, and -mC) are now available to be examined in therapeutic applications.
Received 25th September 2020,
Accepted 5th November 2020
DOI: 10.1039/d0ob01970d
rsc.li/obc
carbonyl (Boc) groups prior to the oligomerization reaction.
Use of cationic motifs such as guanidine and amine for the
Introduction
Several antisense oligonucleotides (ASOs) have been modification of oligonucleotides has attracted extensive atten-
approved.1 The core technology of ASOs, used in clinical and tion, and several types of cationically modified oligonucleo-
preclinical applications, is a chemical modification that tides have been developed.4 In our previous studies, the
improves the biophysical and pharmacokinetic properties of GuNA-T-modified oligonucleotides showed high duplex-
ASOs.2 The commonly used phosphorothioate backbone modi- forming ability toward the complementary ssRNAs and a
fication increases the enzymatic stability and cellular uptake of greatly increased enzymatic stability compared to that of the
ASOs. On the other hand, sugar modifications such as 2′-O- 2′,4′-BNA/LNA-modified oligonucleotides. Based on these
methoxyethyl-RNA (MOE) and 2′,4′-bridged nucleic acids/ results, GuNA is believed to be a promising modification for
locked nucleic acids (Fig. 1, 2′,4′-BNA/LNA) increase the enzy- therapeutic ASOs. However, two issues need to be resolved
matic stability and duplex-forming ability toward the target before its applications. The first is the preparation of GuNA-
single-stranded RNA (ssRNA).
purines, since purine nucleosides generally require multistep
We recently designed a guanidine-bridged nucleic acid reactions to afford 2′-amino-substituted products.5 The second
(Fig. 1, GuNA),3 and GuNA bearing a thymine (T) nucleobase is the use of 75% trifluoroacetic acid (TFA) for the removal of
was successfully synthesized and incorporated into oligonu- the Boc protective groups, because such acidic conditions
cleotides by protecting the guanidine moiety with tert-butoxy- usually cause undesired depurination side reactions (elimin-
aSohyaku. Innovative Research Division, Mitsubishi Tanabe Pharma Corporation,
Shonan Health Innovation Park, 2-26-1 Muraoka-Higashi, Fujisawa, Kanagawa 251-
8555, Japan; 1000 Kamoshida, Aoba-ku, Yokohama 227-0033, Japan
bGraduate School of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka,
Suita, Osaka 565-0871, Japan. E-mail: obika@phs.osaka-u.ac.jp;
Fax: +81 6 6879 8204; Tel: +81 6 6879 8200
cNational Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN), 7-6-
8 Saito-Asagi, Ibaraki, Osaka 567-0085, Japan
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0ob01970d
Fig. 1 Structures of 2’,4’-BNA/LNA and GuNA nucleosides.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 2 Synthetic plan for GuNA-T, -A, -G, and -mC phosphoramidites
(DMTr: 4,4’-dimethoxytrityl).
ation of purine nucleobases) and thus limit the use of ASOs to
pyrimidine-rich sequences.
Recently, we developed a novel synthetic route for 2′-amino-
LNA,6,7 in which transglycosylation was used as a key reaction
(Fig. 2).8 This synthetic route allowed us to easily prepare the
intermediates for GuNA. We herein describe the facile syn-
thesis of GuNA phosphoramidites bearing adenine (A),
Scheme 1 Preparation of 1A, 1G, and 1mC from 1T via transglycosyla-
tion reactions. Reagents and conditions: (a) N,O-bistrimethyl-
silylacetamide, TMSOTf, 1,2-dichloroethane, 70%; (b) TBAF, THF, 0 °C,
guanine (G), and 5-methylcytosine (mC) nucleobases and a 97% (for 1A), 87% (for 1G), 99% (for 1mC).
robust preparation method for GuNA-modified oligonucleo-
tides, even with sequences having acid-sensitive purine nucleo-
bases, by using a base-labile protective group for the guanidine
moiety. The duplex-forming ability and base discrimination of
the given GuNA-modified oligonucleotides and safety assess-
ment data of the GuNA nucleosides are also described.
Results and discussion
Synthesis of GuNA-T, A, -G, and -mC phosphoramidites bearing
various types of protective groups on the guanidine moiety
First, 5′-O-(4,4′-dimethoxytrityl)-2′-amino-LNA 2A, 2G, and 2mC
were prepared from the corresponding thymine analogue 1T
using transglycosylation conditions (Scheme 1), as we had
recently described.8 5-Methylcytosine nucleobase was selected
instead of natural cytosine considering the potential risk of
immune response.9 The trimethylsilyl (TMS) groups were then
removed by using tetra-n-butylammonium fluoride (TBAF) to
afford 1A, 1G, and 1mC in good yields.
We selected five protective groups, Boc, acetyl (Ac), isobu-
tyryl (iB), 9-fluorenylmethyloxycarbonyl (Fmoc), and 2-(tri-
methylsilyl)ethoxycarbonyl (Teoc), for the guanidine moiety of
GuNAs.10 S-Methylisothioureas having these protective
groups4d,11–13 were used for the guanidination reactions of 1T,
1A, 1G, and 1mC to afford the precursors of the GuNA phos-
phoramidites (Scheme 2). For 3T, five compounds that were
differently protected were obtained. Among them, carbamate-
type 3T-Boc, 3T-Teoc, and 3T-Fmoc were successfully converted
into the corresponding phosphoramidites 4T-Boc,3 4T-Teoc,
and 4T-Fmoc. However, we could not isolate 4T-Ac and 4T-iB
Scheme 2 Synthesis of GuNA phosphoramidites. Reagents and con-
ditions: (a) N,N’-protected-S-methylisothiourea,4d,11–13 AgOTf, DIPEA,
THF, rt; (b) (i-Pr)2NP(Cl)O(CH2)2CN, DIPEA, CH2Cl2, rt or [(i-Pr)2N]2PO
(CH2)2CN, diisopropylammonium tetrazolide, MeCN, rt.
in acceptable yields and purities because these amidites are 1T by using a monoacetyl-protected S-methylisothiourea,14 but
very unstable (probably due to the hydrolysis of one of the two no guanidylated product was obtained probably due to the
acyl protective groups). We also performed guanidination of instability of the mono-acetyl-guanidination reagent. From 1A,
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
1G, and 1mC, we achieved the syntheses of phosphoramidites 50% triethylamine in acetonitrile, respectively. Unlike the Boc
4A, 4G, and 4mC bearing three types of protective groups.15 method, the Teoc and Fmoc methods successfully afforded
The synthetic route described here is very useful because all ON4 and ON5 in 9–17% yields. Finally, we successfully incor-
GuNA phosphoramidites are prepared rapidly in four steps porated GuNA into sequences with purine bases that had been
from the same starting compound 1T.
problematic because of the limitation of the protective groups,
and oligonucleotides with GuNA-T, -A, -G, and -mC were given
in acceptable yields using the Teoc and Fmoc methods.
Synthesis of oligonucleotides containing GuNA-T, -A, -G, and
-mC
Duplex-forming ability of GuNA-T-, GuNA-A-, GuNA-G-, and
GuNA-mC-modified oligonucleotides toward fully-matched/
one-base-mismatched ssRNAs or ssDNAs
Solid phase synthesis of oligonucleotides bearing GuNA(Boc/
Fmoc/Teoc)-A/G/mC/T was performed on an automated DNA
synthesizer. We confirmed from trityl monitoring that all of
these GuNA amidites 4 were successfully incorporated into oli-
gonucleotides despite each having a sterically bulky protective
group on its guanidine moiety.
To evaluate the thermal stability of the duplexes containing
GuNAs, we synthesized GuNA-modified oligonucleotides with
a sequence of 5′-d(GCGTTXTTTGCT)-3′, where X represents
GuNA-T, -A, -G, or -mC (ON8–ON11). These oligonucleotides
were successfully produced in 8–31% yields by using Fmoc-
protected GuNA phosphoramidites. The given oligonucleotides
were hybridized with fully matched or one-base-mismatched
strands, and their stabilities were determined by UV-melting
curves. The obtained Tm values are summarized in Table 2.
Consistent with previous results observed in poly-T/poly-A
duplexes containing GuNA-T,3 GuNA-T-modified ON8 exhibi-
ted significantly higher Tm values toward complementary
ssRNA and ssDNA (53 and 56 °C, respectively) than its natural
counterpart. The Tm value of ON8 toward complementary
ssRNA (Y = A) was close to that of its LNA-T-modified counter-
part. Importantly, GuNA-A-, GuNA-G-, and GuNA-mC-modified
oligonucleotides (ON9–ON11) also exhibited high Tm values
toward their complementary strands. Hence, the duplex-stabi-
lizing effects of GuNAs were confirmed regardless of the type
of GuNA nucleobase.
Deprotection of Boc, Teoc, and Fmoc groups was performed
prior to cleavage from the polystyrene (PS) support to avoid tri-
azine formation.4d When Boc was used as a protective group
(Boc method), the PS support was treated with 20% TFA in di-
chloromethane, followed by ammonia treatment. As a result,
for ON2 and ON3 containing one or three GuNA-A monomer
(s) in the poly-T sequence, respectively, we succeeded in
obtaining the target oligonucleotide. We also successfully
obtained ON1, a reported oligonucleotide.3 In contrast, ON4
and ON5 containing GuNA-A monomers in the poly-A
sequence were not obtained because of degradation by TFA
treatment (Table 1). Thus, the Boc method was suited for the
synthesis of pyrimidine-rich sequences, but not for the syn-
thesis of purine-rich sequences. Next, we examined the Teoc
and Fmoc methods. Before cleavage from the PS support by
ammonia treatment, the Teoc group and Fmoc group were
removed by treatment with 0.2 M TBAF in tetrahydrofuran and
It was very interesting that most of the GuNA-modified oli-
gonucleotides showed higher base discrimination than the
natural oligonucleotides did. For example, the base discrimi-
nation value (difference in Tm between fully matched and the
most stable mismatched duplex) of ON8 toward ssRNA was
11 °C, which was 8 °C higher than that of the natural oligo-
nucleotide (X = DNA-T). All the GuNA-modified oligonucleo-
tides showed excellent base discrimination, especially toward
ssRNA. It is also noteworthy that GuNA-T-modified ON8 had
far superior base discrimination to the corresponding LNA-T-
modified oligonucleotide, which also possessed the 2′,4′-
bridge structure (base discrimination values: 11 °C for ON8
and 5 °C for LNA-T-modified oligonucleotide). Thus, GuNA is
believed to be a valuable modification for therapeutic anti-
sense oligonucleotides in which decreased binding to off-
target RNAs is considered to be important for reducing tox-
icity. From the base discrimination studies, we also found that
GuNA-T has destabilizing effects on mismatch pairing with
guanine, which mainly contributed to the high base discrimi-
nation ability of ON8 toward ssRNA. We believe that the guani-
dine moiety of GuNA causes a steric/electronic clash with the
2-amino group of the facing guanine when forming a T·G
wobble base pair (Fig. 3). GuNA-G·rU and GuNA-G·T wobble
base pairs did not show such a decrease in Tm, which would
support our GuNA-T·G destabilization model in Fig. 3. Another
Table 1 Prepared GuNA-modified oligonucleotides
MALDI-TOF-MSb
Oligonucleotide
sequencea
Yield
Method [%]
Calcd
Found
5′-d(TTTTT
̲
Boc
64c
14
43
8
3045.530 3046.698
3055.541 3055.781
3211.630 3211.822
3136.645 3136.871
3274.711 3274.999
Teoc
Fmoc
Boc
Teoc
Fmoc
Boc
Teoc
Fmoc
Boc
Teoc
Fmoc
Boc
5′-d(TTTTA̲TTTTT)-3′ (ON2)
29
38
14
19
22
5′-d(TTTA̲A̲A̲TTTT)-3′ (ON3)
d
5′-d(AAAAA
̲
—
17
10
d
5′-d(AAAA
̲
̲
—
Teoc
Fmoc
Teoc
Fmoc
13
9
11
38
9
5′-d(TTTTG
3071.536 3071.569
3045.545 3045.692
5′-d(TTTTmC
Fmoc
18
a T
exact [M − H]−. c Data from ref. 3. d Decomposition.
̲
= GuNA-T, A
̲
= GuNA-A, G
̲
= GuNA-G, and mC
̲
= GuNA-mC. b m/z of
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
Genotoxicity evaluation of GuNA nucleosides
Since oligonucleotides can be metabolized in vivo to form
monomers,16 it is desirable that the nucleoside monomers
used for oligonucleotide therapeutics are not genotoxic.17 To
assess whether the GuNA monomers are genotoxic, we syn-
thesized monomers 5 from 3-Boc (Scheme 3). The 3T-Boc com-
pound was treated with TFA to remove both the Boc and DMTr
groups to obtain the target GuNA monomer 5T. The A, G, and
mC derivatives of 3-Boc were treated with ammonia to remove
the protective group(s) at the base moiety prior to Boc and
DMTr deprotection with TFA. All deprotected nucleosides were
synthesized successfully as test articles for genotoxicity.
Bacterial reverse mutation assays18,19 were carried out on
the GuNA monomers 5 by using five tester strains (Table 4). All
monomers were not mutagenic in all tester strains, regardless
of metabolic activation.
Fig. 3 Plausible destabilization mechanism of T·G mismatched wobble
base pairing.
Table 3 Thermal stability of the duplexes formed between multiple
GuNA-modified oligonucleotides and their complimentary ssDNAsa
Oligonucleotide
Tm (ΔTm/mod) [°C]
5′-d(TTTTTTTTTT)-3′
22
5′-d(TTTT
5′-d(AAAAAAAAAA)-3′
5′-d(AAAA AAAA)-3′ (ON5)
5′-d(TTTGGGTTTT)-3′
5′-d(TTTG TTTT)-3′ (ON13)
5′-d(TTTCCCTTTT)-3′
5′-d(TTTmCmCm
TTTT)-3′ (ON14)
̲
̲
̲
52
22
38
36
57
31
51
(+10)
̲
̲
̲
(+6)
(+7)
(+7)
̲G̲G̲
Conclusions
̲
̲ C̲
We synthesized GuNA-T, -A, -G, and -mC phosphoramidites
with different protective groups and investigated their suit-
ability for oligonucleotide synthesis. The Boc protective group
was useful for the synthesis of oligonucleotides with a pyrimi-
dine-rich sequence, but we found that sequences having
several purine nucleobases could only be synthesized using
GuNAs protected with Teoc or Fmoc protective groups. The
new synthetic route enabled us to prepare GuNA-modified oli-
gonucleotides with no sequence limitations.
Consistent with previous results obtained with GuNA-T-
modified oligonucleotides, the GuNA-A-, GuNA-G-, and
GuNA-mC-modified oligonucleotides showed excellent duplex-
stabilizing ability toward their complementary ssRNAs and
ssDNAs. Moreover, some of the GuNA-modified oligonucleo-
tides were found to have high base discriminating ability. Four
GuNA nucleosides exhibited no genotoxicity in bacterial
reverse mutation assays. Therefore, GuNA-T, -A, -G, and -mC
are now available to examine in therapeutic applications.
a Conditions: 10 mM sodium phosphate buffer (pH 7.0), 100 mM
NaCl, and 4 µM each oligonucleotide. The Tm values reflect the average
of three measurements. ΔTm/mod: the change in Tm value (ΔTm) per
modification compared to the natural oligonucleotides. T
̲ = GuNA-T, A̲
= GuNA-A, G
̲
= GuNA-G, and mC = GuNA-mC.
̲
important property of GuNA-T is a cumulative effect on the Tm
values by multiple modifications.3 This was also confirmed
with GuNA-A-, GuNA-G-, and GuNA-mC-modified oligonucleo-
tides (ON5, ON13, and ON14) (Table 3).
Experimental
General
Scheme 3 Synthesis of GuNA nucleosides without protective groups.
1H NMR, 13C NMR, and 31P NMR spectra were recorded at a
400 MHz NMR spectrometer. Chemical shift values are
expressed as δ values (ppm) relative to internal tetramethyl-
Reagents and conditions: (a) TFA, CH2Cl2, rt, then DOWEX™ 1 × 2 Cl−
form, 73% (b) 7 M NH3 in MeOH, rt; TFA, CH2Cl2, rt, then DOWEX™ 1 × 2
Cl− form, 63–98%.
Table 4 Results of bacterial reverse mutation assay with and without metabolic activation
Tester strain
Compound
TA98
TA100
TA1535
TA1537
WP2uvrA
5T
5A
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
5G
5mC
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
silane (0.00 ppm) or residual CHCl3 (7.26 ppm) for 1H NMR, General procedure for desilylation (synthesis of 1A)
internal tetramethylsilane or 3-(trimethylsilyl)propionic-2,2,3,3-
2A (110 mg, 0.15 mmol) and tetrahydrofuran (2 mL) were
mixed, 1 M tetra-n-butylammonium fluoride in tetrahydro-
furan (0.19 mL, 0.19 mmol) was added dropwise under ice
cooling, and the mixture was stirred on ice for 50 minutes.
d4 acid, sodium salt (0.00 ppm) or chloroform-d (δ = 77.0 ppm)
for 13C NMR, and 0.05% aqueous H3PO4 (δ = 0.00 ppm) for 31P
NMR. High resolution mass spectra of all new compounds were
measured on a Thermo Scientific LTQ Orbitrap XL. For column
Chloroform (10 mL) and saturated brine (2 mL) were added to
chromatography, silica gel (Wakogel® 60N, FUJIFILM Wako,
the mixture, which was then stirred and extracted. The organic
Osaka, Japan) or pre-packed silica gel column (Hi-Flash-M, L, or
layer was washed with saturated brine (2 mL) and then dried
3L, YAMAZEN, Osaka, Japan) was used. The progress of reac-
over sodium sulfate. Insoluble material was removed by fil-
tions was monitored by analytical thin layer chromatography
tration, and the filtrate was dried by evaporation. The resulting
(TLC) on glass plates (TLC Silica gel 60 F254), and the products
residue was then purified by silica gel chromatography (chloro-
form/methanol, 100/0 to 91/9) to obtain compound 1A (95 mg,
yield 97%) as a white powder.
were visualized by UV light or LC/MS. The oligonucleotides were
analyzed for purity by reverse-phase HPLC on a Shimadzu HPLC
system with a Waters XBridge™ OST C18 2.5 μm (4.6 × 50 mm)
column. The eluent was 400 mM hexafluoroisopropanol/15 mM
triethylammonia aqueous solution and methanol, and the
eluate was characterized by MALDI-TOF mass spectroscopy on a
Bulker Daltonics Autoflex TOF/TOF mass spectrometer.
1H NMR (CDCl3) δ: 8.97 (1H, s), 8.79 (1H, s), 8.32 (1H, s),
8.08–7.99 (2H, m), 7.65–7.21 (13H, m), 6.91–6.83 (4H, m), 6.05
(1H, s), 4.30 (1H, s), 3.91 (1H, s), 3.80 (6H, s), 3.58 (1H, d, J =
10.8 Hz), 3.52 (1H, d, J = 10.8 Hz), 3.16 (1H, d, J = 10.8 Hz),
3.09 (1H, d, J = 10.8 Hz), 2.33 (1H, brs); 13C-NMR (CDCl3) δ:
164.95, 158.60, 152.41, 150.55, 149.48, 144.46, 140.65, 135.62,
135.53, 133.61, 132.78, 130.04, 128.82, 128.10, 127.98, 127.96,
127.02, 123.44, 113.28, 88.05, 87.94, 86.53, 1.34, 62.27, 60.17,
55.23, 52.83, 50.22, 20.46, 13.82; HRMS (ESI) calcd for
C39H37N6O6 [M + H]+: 685.2769, found 685.2782.
Transglycosylation of 2′-amino-LNA-T to 2′-amino-LNA-A(bz)
(synthesis of 2A)
Compound 2A was prepared following our previous report.8
Synthesis of 2G
Compound 2G was prepared following our previous report.8
Synthesis of 1G
Using a similar procedure as for 1A, compound 1G (163 mg,
yield 87%) was prepared as a white powder from 2G (210 mg,
0.22 mmol).
Synthesis of 2mC
1H NMR (CDCl3) δ: 8.20 (1H, s), 8.08 (1H, s), 7.49–7.18
(19H, m), 6.85–6.80 (m, 4H), 5.93 (1H, s), 4.48 (1H, s), 4.01
(1H, s), 3.77 (6H, s), 3.49 (1H, d, J = 10.8 Hz), 3.45 (1H, d, J =
10.8 Hz), 3.19 (1H, d, J = 10.8 Hz), 3.13 (1H, d, J = 10.8 Hz),
2.80 (1H, brs), 1.24 (3H, d, J = 6.7 Hz), 1.23 (3H, d, J = 6.7 Hz);
13C-NMR (CDCl3) δ: 175.7, 158.5, 155.9, 153.6, 151.6, 150.6,
144.6, 142.4, 141.7, 135.6, 130.0, 129.3, 129.2, 128.1, 127.9,
126.9, 121.7, 121.0, 117.8, 113.2, 87.9, 87.8, 86.3, 72.2, 62.3,
60.5, 55.2, 50.4, 36.1, 19.3, 19.1; HRMS (ESI) calcd for
C49H48N7O8 [M + H]+: 862.3559, found 862.3567.
The mixture of 1T (2.00 g, 3.50 mmol), N-(5-methyl-2-oxo-1H-
pyrimidin-4-yl)benzamide (2.40 g, 10.50 mmol), and N,O-bis
(trimethylsilyl)acetamide (7.7 mL, 31.50 mmol) in 1,2-dichlor-
oethane (35 mL) was stirred at 40 °C (external temperature) for
ten minutes. Thereafter, trimethylsilyl triflate (0.10 mL,
0.52 mmol) was added dropwise for five minutes, and the
mixture was stirred at the same temperature for ten minutes.
The mixture was poured into ice-cold, saturated aqueous
sodium bicarbonate solution, stirred for five minutes, and then
separated using a separatory funnel. The aqueous layer was
extracted with chloroform twice, and the combined organic
layer was washed with brine and dried over sodium sulfate.
Insoluble material was removed by filtration, and the filtrate was
Synthesis of 1mC
dried in vacuo. The resulting residue was purified by silica gel Using a similarly procedure as for 1A, compound 1mC (1.00 g,
chromatography (chloroform/methanol, 99/1 to 95/5) to obtain yield 99%) was prepared as a pale yellow powder from 2mC
compound 2mC (1.84 g, yield 70%) as a white powder.
(1.12 g, 1.50 mmol).
1H NMR (CDCl3) δ: 13.43 (1H, brs), 8.32 (2H, d, J = 8.3 Hz),
1H NMR (CDCl3) δ: 13.25 (1H, brs), 8.26–8.24 (2H, m), 7.80
8.04 (1H, s), 7.55–7.24 (13H, m), 6.88–6.82 (4H, m), 5.58 (1H, (1H, m), 7.64–7.20 (15H, m), 6.87–6.83 (4H, m), 5.75 (1H, brs),
s), 4.27 (1H, s), 3.81–3.80 (6H, m), 3.59 (1H, s), 3.53 (1H, d, J = 4.38 (1H, brs), 4.15 (1H, brs), 3.83–3.72 (1H, m), 3.78 (3H, s),
10.8 Hz), 3.23 (1H, d, J = 10.8 Hz), 2.97 (1H, d, J = 10.3 Hz), 3.77 (3H, s), 3.70 (1H, brd, J = 11.0 Hz), 3.52 (1H, brd, J = 11.0
2.78 (1H, d, J = 10.3 Hz), 2.12 (1H, brs), 1.87 (3H, s), 0.06 (9H, Hz), 3.50 (1H, brs), 3.33 (1H, brd, J = 11.4 Hz), 1.76 (3H, s);
s); 13C-NMR (CDCl3) δ: 179.5, 159.9, 158.6, 147.5, 144.4, 137.1, 13C-NMR (CDCl3) δ: 171.1, 158.8, 158.8, 144.4, 135.3, 134.1,
136.3, 135.5, 132.3, 130.0, 129.9, 129.8, 128.3, 128.0, 127.9, 132.5, 131.0, 130.1, 130.0, 129.9, 128.4, 128.1, 127.2, 113.4,
127.0, 113.2, 113.1, 111.0, 89.1, 88.9, 86.3, 70.4, 62.3, 59.1, 88.1, 87.5, 87.0, 77.2, 69.8, 65.9, 59.0, 55.3, 51.6, 13.6; HRMS
55.2, 50.6, 13.7, 0.0; HRMS (ESI) calcd for C42H47N4O7Si [M + (ESI) calcd for
C39H39N4O7 [M +
H]+: 675.2813, found
H]+: 747.3209, found 747.3210.
675.2817.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
General procedure for guanidine formation (synthesis of
3T-Teoc)
135.63, 135.52, 130.07, 129.22, 129.15, 128.10, 127.81, 127.69,
127.65, 127.58, 127.48, 127.42, 127.35, 127.26, 127.05, 126.89,
126.76, 126.61, 121.60, 113.31, 113, 14, 86.56, 86.15, 74.30,
65.16, 63.85, 61.15, 55.19, 36.71, 19.27, 19.17, 17.64, 17.35,
−1.54; HRMS (ESI) calcd for C62H74N9O12Si2 [M + H]+:
1192.4990, found 1192.5009.
A solution of compound 1T (1266 mg, 2.22 mmol) and 2-tri-
methylsilylethyl
N-[methylsulfanyl-(2-trimethyl-
silylethoxycarbonylamino)methylene]carbamate (1258 mg,
3.32 mmol) in tetrahydrofuran (7 mL) was stirred, and then tri-
ethylamine (1.23 mL, 8.86 mmol) and silver(I) triflate (854 mg,
3.32 mmol) were added successively under ice cooling. The Synthesis of 3mC-Teoc
mixture was warmed to room temperature and stirred for four
Using a similar procedure as for 3T-Teoc, compound 3mC-Teoc
hours. Under ice cooling, saturated brine was then added, and
the mixture was extracted with ethyl acetate. The organic layer
was dried over anhydrous sodium sulfate, and insoluble
material was removed by filtration. The solvent was evaporated
from the filtrate, and the resulting residue was purified by
silica gel chromatography (hexane/ethyl acetate, 55/45 to 45/
55) to obtain 3T-Teoc (1533 mg, yield 77%) as a white powder.
1H NMR (CDCl3) δ: 10.36 (1H, brs), 8.26 (1H, s), 7.59 (1H,
s), 7.49–7.42 (2H, m), 7.39–7.20 (7H, m), 6.89–6.80 (4H, m),
5.62 (1H, s), 4.71 (1H, s), 4.36–4.07 (5H, m), 3.84–3.70 (7H, m),
3.62 (1H, d, J = 11.8 Hz), 3.57–3.46 (2H, m), 1.67 (3H, s),
1.11–0.97 (4H, m), 0.04 (18H, s); 13C NMR (CDCl3) δ: 163.80,
158.20, 149.80, 144.61, 135.26, 135.00, 134.73, 134.60, 129.82,
129.73, 127.95, 127.65, 126.86, 113.29, 108.64, 87.52, 63.31,
62.13, 55.05, 17.16, 12.24, −1.50, −1.55; HRMS (ESI) calcd for
C45H60N5O11Si2 [M + H]+: 902.3822, found 902.3821.
(870 mg, yield 66%) was prepared as a white powder from 1mC
(890 mg, 1.32 mmol).
1H NMR (CDCl3) δ: 13.41 (1H, brs), 10.32 (1H, brs),
8.32–8.30 (2H, m), 7.75 (1H, s), 7.54–7.23 (12H, m), 6.88–6.85
(4H, m), 5.67 (1H, s), 4.78 (1H, s), 4.35 (1H, brd, J = 1.7 Hz),
4.25–4.13 (5H, m), 3.80 (s, 3H), 3.80 (s, 3H), 3.76 (1H, d, J =
11.7 Hz), 3.64 (1H, d, J = 11.7 Hz), 3.57–3.51 (2H, m), 1.86 (3H,
s), 1.08–1.03 (4H, m), 0.04 (18H, s); 13C NMR (DMSO-d6) δ:
179.77, 162.84, 159.59, 158.73, 153.88, 153.10, 147.44, 144.39,
137.05, 135.53, 35.34, 135.22, 132.55, 130.10, 129.95, 128.14,
128.07, 127.11, 113.35, 111.83, 89.15, 86.79, 69.47, 65.97,
64.58, 64.11, 59.22, 55.25, 53.63, 17.54, 17.35, 13.68, −1.53;
HRMS (ESI) calcd for C52H65N6O11Si2 [M + H]+: 1005.4244,
found 1005.4245.
Synthesis of 3A-Boc
Synthesis of 3A-Teoc
Using a similar procedure as for 3T-Boc,3 compound 3A-Boc
(1.70 g, yield 74%) was prepared as a white powder from 1A
(1.74 g, 2.54 mmol).
Using a similar procedure as for 3T-Teoc, compound 3A-Teoc
(110 mg, yield 82%) was prepared as a white powder from 1A
(92 mg, 0.13 mmol).
1H NMR (CDCl3) δ: 10.46 (1H, brs), 9.09 (1H, s), 8.74 (1H,
s), 8.26 (1H, s), 8.02–8.00 (2H, m), 7.63–7.20 (12H, m),
6.85–6.82 (4H, m), 6.23 (1H, s), 5.21 (1H, s), 4.51 (1H, s),
3.97–3.84 (2H, m), 3.79 (6H, s), 3.61–3.50 (3H, m), 1.49 (18H,
s); 13C-NMR (CDCl3) δ: 164.54, 158.64, 152.70, 150.62, 149.54,
144.29, 141.06, 135.46, 135.31, 133.57, 132.87, 130.05, 130.02,
128.92, 128.08, 127.99, 127.83, 128.03, 123.40, 113.30, 88.09,
86.59, 86.45, 77.24, 71.56, 65.21, 60.30, 55.25, 53.98, 28.12;
HRMS (ESI) calcd for C50H55N8O10 [M + H]+: 927.4036, found
927.4043.
1H NMR (CDCl3) δ: 10.66 (1H, brs), 9.01 (1H, s), 8.77 (1H,
s), 8.25 (1H, s), 8.08–7.98 (2H, m), 7.67–7.40 (5H, m), 7.39–7.17
(7H, m), 6.90–6.78 (4H, m), 6.28 (1H, s), 5.13 (1H, brs),
4.63–4.54 (1H, m), 4.32–4.15 (4H, m), 3.96 (1H, d, J = 11.3 Hz),
3.87–3.74 (7H, m), 3.62–3.50 (2H, m), 3.31–3.09 (1H, m),
1.16–0.98 (4H, m), 0.04 (18H, s); 13C NMR (CDCl3) δ: 165.59,
158.08, 151.68, 151.48, 151.45, 150.49, 146.22, 144.58, 135.38,
135.15, 133.32, 132.44, 129.68, 129.60, 128.46, 127.62, 126.71,
125.54, 113.21, 85.43, 63.48, 62.80, 62.25, 60.54, 57.56, 54.98,
53.25, 53.14, 38.88, 20.00, 17.14, 13.90, −1.54; HRMS (ESI)
calcd for C52H63N8O10Si2 [M
+
H]+: 1015.4200, found Synthesis of 3G-Boc
1015.4204.
Using a similar procedure as for 3T-Boc,3 compound 3G-Boc
(83 mg, yield 82%) was prepared as a white powder from 1G
(78 mg, 0.09 mmol).
Synthesis of 3G-Teoc
Using a similar procedure as for 3T-Teoc, compound 3G-Teoc
1H NMR (CDCl3) δ: 10.54 (1H, brs), 8.08 (2H, brs), 7.43–7.15
(114 mg, yield 84%) was prepared as a white powder from 1G (19H, m), 6.77–6.80 (4H, brd, J = 7.6 Hz), 6.15 (1H, brs), 5.09
(94 mg, 0.11 mmol). (2H, brs), 4.10–4.07 (2H, m), 3.86–3.75 (7H, m), 3.46–3.38 (2H,
1H NMR (CDCl3) δ: 10.87 (1H, brs), 8.12 (1H, s), 8.03 (1H, m), 2.66 (1H, brs), 1.92 (1H, brs), 1.48–1.41 (18H, m),
s), 7.52–7.33 (11H, m), 7.32–7.13 (8H, m), 6.81–6.75 (4H, m), 1.22–1.15 (6H, m); 13C-NMR (CDCl3) δ: 174.81, 162.48, 158.49,
6.24 (1H, brs), 5.18 (1H, s), 4.99 (1H, brs), 4.26–4.03 (6H, m), 156.16, 154.84, 153.54, 151.16, 150.38, 144.55, 144.05, 141.72,
3.93–3.81 (1H, m), 3.76 (3H, s), 3.76 (3H, s), 3.45 (1H, d, J = 135.66, 135.57, 130.06, 129.22, 128.11, 127.81, 126.78, 121.51,
10.8 Hz), 3.38 (1H, d, J = 10.8 Hz), 2.65–2.52 (1H, m), 1.22 (3H, 113.15, 86.87, 86.15, 82.58, 79.93, 77.36, 77.25, 77.04, 76.73,
d, J = 7.2 Hz), 1.17 (3H, d, J = 6.7 Hz), 1.13–0.87 (4H, m), 0.03 73.81, 61.01, 55.19, 54.62, 36.50, 28.16, 28.02, 19.29, 19.13;
(9H, brs), −0.01 (9H, brs); 13C-NMR (CDCl3) δ: 174.67, 163.35, HRMS ESI) calcd for C60H66N9O12 [M + H]+: 1104.4826, found
158.20, 156.20, 153.57, 151.0, 150.40, 144.26, 144.26, 141.67, 1104.4830.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Synthesis of 3mC-Boc
1H NMR (CDCl3) δ: 11.14 (1H, brs), 8.12 (1H, s), 7.98 (1H,
s), 7.75–7.60 (m, 6H), 7.60–7.15 (m, 19H), 6.89–6.85 (4H, m)
6.28–6.85 (1H, m), 5.30 (1H, d, J = 3.3 Hz), 5.64–4.90 (1H, m),
4.41–3.96 (9H, m), 3.74 (6H, brs), 3.49–3.41 (2H, m), 2.53—
2.47 (1H, m), 1.17 (3H, d, J = 6.9), 1.14 (3H, d, J = 6.9 Hz); 13C
NMR (CDCl3) δ: 158.48, 156.26, 153.50, 150.96, 150.36, 144.51,
144.19, 144.14, 142.95, 141.64, 141.19, 135.58, 135.48, 130.05,
129.35, 129.22, 128.08, 127.88, 127.83, 127.64, 127.16, 127.09,
126.81, 125.42, 125.24, 125.06, 120.00, 119.92, 113.14, 86.19,
77.22, 68.25, 60.94, 55.18, 46.93, 46.61, 36.83, 19.18; HRMS
(ESI) calcd for C80H70N9O12 [M + H]+: 1348.5139, found
1348.5140.
Using a similar procedure as for 3T-Boc,3 compound 3mC-Boc
(1467 mg, yield 61%) was prepared as a white powder from
1mC (1762 mg, 2.61 mmol).
1H NMR (CDCl3) δ: 13.39 (1H, brs), 10.10 (1H, brs),
8.32–8.29 (2H, m), 7.77 (1H, d, J = 0.9 Hz), 7.55–7.23 (12H, m),
6.88–6.83 (4H, m), 5.62 (1H, s), 4.72 (1H, s), 4.34 (1H, d, J = 2.7
Hz), 3.80(s, 3H), 3.80(s, 3H), 3.78 (1H, d), 3.58 (1H, d, J = 11.6
Hz), 3.54–3.48 (2H, m), 1.87 (3H, d, J = 0.9 Hz), 1.49 (18H, s);
13C-NMR (CDCl3) δ: 179.76, 161.86, 159.67, 158.72, 158.70,
153.52, 152.19, 147.46, 144.43, 137.06, 135.68, 135.39, 135.25,
132.54, 130.09, 129.94, 128.15, 128.08, 128.06, 127.09, 113.34,
111.75, 89.28, 86.90, 86.74, 83.70, 80.25, 69.40, 64.43, 59.24,
55.26, 53.42, 28.07, 13.73; HRMS (ESI) calcd for C50H57N6O11
[M + H]+: 917.4080, found 917.4074.
Synthesis of 3mC-Fmoc
Using
a similar procedure as for 3T-Fmoc, compound
3mC-Fmoc (1.63 g, yield 54%) was prepared as a white powder
from 1mC (1.77 g, 2.62 mmol).
Synthesis of 3T-Fmoc
1H NMR (CDCl3) δ: 13.46 (1H, brs), 10.56 (1H, brs), 8.32
(2H, d, J = 8.5 Hz), 7.82–7.70 (5H, m), 7.70–7.55 (4H, m),
7.55–7.40 (5H, m), 7.40–7.25 (16H, m), 6.87–6.85 (4H, m), 5.73
(1H, brs), 4.71 (1H, brs), 4.55–4.20 (7H, m), 3.81 (3H, s), 3.81
(3H, s), 3.81–3.63 (2H, m), 3.57 (2H, s), 1.87 (3H, s); 13C NMR
(CDCl3) δ: 159.62, 158.76, 158.74, 147.40, 147.35, 144.36,
141.24, 137.02, 135.46, 135.27, 135.18, 132.59, 130.10, 129.96,
128.17, 128.11, 128.06, 127.71, 127.17, 125.30, 125.18, 120.03,
113.37, 111.88, 77.21, 89.12, 86.88, 86.66, 69.38, 64.70, 59.07,
55.28, 53.70, 46.81, 46.59, 13.70; HRMS (ESI) calcd for
C70H61N6O11 [M + H]+: 1161.4393, found 1161.4395.
Using a similar procedure as for 3T-Teoc, compound 3T-Fmoc
(2.36 g, 2.23 mmol, yield 71%) was prepared as a white powder
from 1T (1.80 g, 3.15 mmol) and N,N′-bis(9-fluorenylmethoxy-
carbonyl)-S-methyl-isothiourea using pyridine instead of
triethylamine.
1H NMR (CDCl3) δ: 10.60 (1H, brs), 7.98 (1H, brs), 7.77–7.73
(4H, m), 7.69–7.65 (2H, m), 7.62 (1H, brs), 7.59–7.54(2H, m),
7.48–7.46 (2H, m), 7.40–7.22 (19H, m), 6.87–6.84 (4H, m), 5.67
(1H, brs), 4.71 (1H, brs), 4.50–4.30 (6H, m), 4.27–4.23 (1H, m), 3.80
(3H, s), 3.80 (3H, s), 3.65(1H, d, J = 11.4 Hz), 3.55 (2H, s), 1.67 (3H,
d, J = 0.9 Hz); 13C NMR (CDCl3) δ: 163.21, 162.89, 158.84, 156.33,
149.43, 144.36, 143.99, 143.03, 141.33, 135.36, 135.26, 134.38,
130.12, 130.10, 128.12, 128.00, 127.96, 127.70, 127.29, 127.24,
127.20, 127.13, 125.45, 125.36, 125.23, 125.10, 120.11, 119.95,
113.43, 110.68, 88.92, 86.96, 86.46, 77.73, 69.64, 68.94, 68.31,
65.00, 59.16, 55.27, 53.84, 46.92, 46.66, 12.51; HRMS (ESI) calcd
for C63H56N5O11 [M + H]+: 1058.3971, found 1058.3969.
Synthesis of 3T-Ac
Using a similar procedure as for 3T-Teoc, compound 3T-Ac
(73 mg, yield 60%) was prepared as a white powder from 1T
(100 mg, 0.18 mmol) and N,N′-diacetyl-S-methyl-isothiourea
(34 mg, 0.19 mmol).
1H NMR (DMSO-d6) δ: 7.53–7.39 (3H, m), 7.53–7.39 (7H, m),
6.93–6.904 (4H, m), 6.67–4.16 (3H, m), 3.75 (6H, s), 3.56–3.36
(3H, s), 2.22–1.50 (6H, m), 1.95 (3H, s); 13C NMR (CDCl3) δ:
172.28, 171.34, 163.75, 158.12, 149.74, 144.52, 135.17, 135.09,
134.90, 134.32, 129.73, 129.65, 127.88, 127.58, 126.77, 113.21,
108.48, 87.19, 86.69, 63.17, 59.65, 59.38, 59.15, 54.96, 52.62,
24.48, 24.26, 23.45, 22.41, 12.21; HRMS (ESI) calcd for
C37H40N5O9 [M + H]+: 698.2821, found 698.2816.
Synthesis of 3A-Fmoc
Using a similar procedure as for 3T-Fmoc, compound 3A-Fmoc
(704 mg, yield 61%) was prepared as a white powder from 1A
(668 mg, 0.98 mmol).
1H NMR (CDCl3) δ: 10.86 (1H, brs), 8.97 (1H, s), 8.49 (1H,
s), 8.26 (1H, s), 8.00, (2H, brd, J = 7.3 Hz)), 7.75–7.48 (12H, m),
7.36–7.16 (16H, m), 6.86–6.83 (4H, m), 6.29 (1H, brs), 5.05 (1H,
brs), 4.62 (1H, brs), 4.46–4.26 (6H, m), 4.00 (1H, m), 3.92–3.85
(2H, m), 3.78 (6H, s), 3.62–3.56 (2H, m), 3.41 (1H, brs); 13C
NMR (CDCl3) δ: 164.47, 158.69, 152.73, 150.73, 149.49, 144.21,
143.98, 141.26, 140.92, 135.35, 135.23, 133.45, 132.90, 130.03,
130.01, 129.04, 128.92, 128.23, 128.06, 127.14, 125.30, 125.17,
123.59, 120.20, 120.02, 113.35, 87.77, 86.79, 71.59, 68.60,
68.41, 60.07, 55.25, 46.88, 46.63; HRMS (ESI) calcd for
C70H59N8O10 [M + H]+: 1171.4349, found 1171.4352.
Synthesis of 3T-iB
Using a similar procedure as for 3T-Teoc, compound 3T-iB
(284 mg, yield quant.) was prepared from 1T (200 mg,
0.35 mmol) and N,N′-dibutyryl-S-methyl-isothiourea (97 mg,
0.42 mmol).
1H NMR (CDCl3) δ: 11.43 (1H, brs), 8.62 (1H, brs), 7.65 (1H,
brs), 7.48–7.46 (2H, m), 7.37–7.23 (7H, m), 6.87–6.83 (4H, m),
5.58 (1H, brs), 4.45 (1H, brs), 4.40 (1H, s), 3.80 (3H, s), 3.80
(3H, s), 3.61 (2H, brs), 3.55 (2H, brs), 2.55 (2H, m), 1.63 (3H,
s), 1.16 (12H, brs); 13C NMR (CDCl3) δ: 163.60, 158.74, 149.59,
Synthesis of 3G-Fmoc
Using a similarly procedure as for 3T-Fmoc, compound 149.56, 144.38, 135.37, 135.24, 134.49, 130.10, 128.07, 127.13,
3G-Fmoc (1.12 g, yield 71%) was prepared as a white powder 113.36, 110.59, 89.16, 86.81, 86.52, 77.23, 69.27, 64.62, 59.06,
from 1G (1.01 g, 1.18 mmol).
55.25, 53.57, 39.22, 36.95, 36.53, 19.28, 18.77, 12.54; HRMS
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
(ESI) calcd for C41H48N5O9 [M
754.3436.
+
H]+: 754.3447, found General procedure for amidite synthesis using diamidite
(synthesis of 4G-Teoc)
To a solution of 3G-Teoc (51 mg, 0.04 mmol) in acetonitrile
(1 mL) at room temperature were added diisopropyl-
ammonium tetrazolide (11 mg, 0.06 mmol) and 2-cyanoethyl
General procedure for amidite synthesis using chloridite
(synthesis of 4T-Teoc)
3T-Teoc (61 mg, 0.07 mmol) and dichloromethane (1.5 mL)
were mixed, and under ice cooling, diisopropylethylamine
(23 μL, 0.13 mmol) and 2-cyanoethyl N,N-diisopropyl-
chlorophosphoramide (22 μL, 0.10 mmol) were added drop-
wise successively. The mixture was stirred at room temperature
for four hours. The mixture was cooled again on ice and
additional diisopropylethylamine (23 μL, 0.13 mmol) and
2-cyanoethyl N,N-diisopropylchlorophosphoramide (22 μL,
0.10 mmol) were added. The mixture was then stirred at room
temperature for two hours. Under ice cooling, saturated
aqueous sodium bicarbonate solution was added, and the
mixture was extracted with chloroform. The organic layer was
washed with water and saturated brine successively, and dried
over anhydrous sodium sulfate. Insoluble material was
removed by filtration. The solvent was evaporated from the fil-
trate, and the resulting residue was purified by silica gel
chromatography (hexane/ethyl acetate, 65/35 to 50/50) to
obtain 4T-Teoc (40 mg, yield 54%) as a white amorphous solid.
31P NMR (CDCl3) δ: 149.46, 149.34, 148.89, 148.73; HRMS
(ESI) calcd for C54H77N7O12PSi2 [M + H]+: 1102.4901, found
1102.4910.
N,N,N′,N′-tetraisopropylphosphorodiamidite
(19
mg,
0.06 mmol). The mixture was stirred at room temperature for
one hour. Additional diisopropylammonium tetrazolide
(22 mg, 0.13 mmol) and 2-cyanoethyl N,N,N′,N′-tetraisopropyl-
phosphorodiamidite (38 mg, 0.13 mmol) were added, and the
mixture was stirred at room temperature for three hours.
Additional
2-cyanoethyl
N,N,N′,N′-tetraisopropyl-
phosphorodiamidite (6 mg, 0.02 mmol) was added, and the
mixture was stirred at room temperature for 20 minutes.
Chloroform (5 mL) and saturated brine (2 mL) were added on
ice. The mixture was stirred and then passed through a phase
separator to remove aqueous phase. After the solvent was evap-
orated, the residue was purified by silica gel chromatography
(hexane/ethyl acetate, 80/20 to 60/40) to obtain 4G-Teoc
(61 mg, yield 83%) as a white powder.
31P NMR (CDCl3) δ: 149.37, 149.28, 148.81; HRMS (ESI)
calcd for
1392.6069.
C71H91N11O13PSi2 [M +
H]+: 1392.6069, found
Synthesis of 4T-Fmoc
Using a similar procedure as for 4G-Teoc, compound 4T-Fmoc
(1.71 g, yield 61%) was prepared as a white powder from
3T-Fmoc (2.36 g, 2.23 mmol).
Synthesis of 4A-Boc
Using a similar procedure as for 4T-Teoc, compound 4A-Boc
(105 mg, yield 86%) was prepared as a white powder from
3A-Boc (100 mg, 0.11 mmol).
31P NMR (CDCl3) δ: 149.76, 149.47, 149.01; HRMS (ESI)
calcd for C72H73N7O12P [M + H]+: 1258.5049, found 1258.5053.
31P NMR (CDCl3) δ: 149.40, 149.25; HRMS (ESI) calcd for
C59H72N10O11P [M + H]+: 1127.5114, found 1127.5117.
Synthesis of 4A-Fmoc
Using a similar procedure as for 4G-Teoc, compound 4A-Fmoc
(1.33 g, yield 73%) was prepared as a white powder from
3A-Fmoc (1.55 g, 1.32 mmol).
Synthesis of 4A-Teoc
Using a similar procedure as for 4T-Teoc, compound 4A-Teoc
(67 mg, yield 53%) was prepared as a white powder from
3A-Teoc (105 mg, 0.10 mmol).
31P NMR (CDCl3) δ: 149.71, 149.56, 149.12; HRMS (ESI)
calcd for C79H76N10O11P [M
1371.5435.
+
H]+: 1371.5427, found
31P NMR (CDCl3) δ: 149.74, 149.45, 149.06; HRMS (ESI)
calcd for
1215.5286.
C61H80N10O11PSi2 [M +
H]+: 1215.5279, found
Synthesis of 4G-Boc
Synthesis of 4mC-Boc
Using a similar procedure as for 4G-Teoc, compound 4G-Boc
(44 mg, yield 84%) was prepared as a white powder from
3G-Boc (42 mg, 0.04 mmol).
Using a similar procedure as for 4T-Teoc, compound 4mC-Boc
(85 mg, yield 49%) was prepared as a white powder from
3mC-Boc (142 mg, 0.15 mmol).
31P NMR (CDCl3) δ: 148.92, 148.80, 148.36; HRMS (ESI)
calcd for C69H83N11O13P [M
1304.5910.
+
H]+: 1304.5904, found
31P NMR (CDCl3) δ: 149.59, 149.18, 148.59; HRMS (ESI)
calcd for C59H74N8O12P [M + H]+: 1117.5158, found 1117.5152.
Synthesis of 4mC-Teoc
Synthesis of 4G-Fmoc
Using a similar procedure as for 4T-Teoc, compound 4mC-Teoc Using a similar procedure as for 4G-Teoc, compound 4G-Fmoc
(31 mg, yield 52%) was prepared as a white powder from (1.07 g, yield 71%) was prepared as a white powder from
3mC-Teoc (50 mg, 0.05 mmol).
3G-Fmoc (1.32 mg, 0.98 mmol).
31P NMR (CDCl3) δ: 149.77, 149.56, 149.16, 148.96; HRMS
31P NMR (CDCl3) δ: 149.53, 149.07, 148.96, 148.72; HRMS
(ESI) calcd for C61H82N8O12PSi2 [M + H]+: 1205.5323, found (ESI) calcd for C89H87N11O13P [M + H]+: 1548.6217, found
1205.5328.
1548.6211.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Synthesis of 4mC-Fmoc
hours, and was then centrifuged (14 000 rcf, 4 °C, 30 minutes).
The precipitate was washed with cold 75% ethanol and then a
solution of 5% dimethyl sulfoxide in 10 mM disodium hydro-
gen phosphate buffer was added. The suspension was passed
through a gel permeation column (illustra™ NAP-10™ column
Sephadex™ G-25 DNA Grade, GE Healthcare), and then puri-
fied by reverse-phase HPLC with Waters XBridge™ C18 (4.6 ×
50 mm analytical and 10 × 50 mm preparative) columns using
a linear gradient of methanol in aqueous 400 mM hexafluoro-
isopropanol/15 mM triethylamine.
Using
a similar procedure as for 4G-Teoc, compound
4mC-Fmoc (1.128 g, yield 62%) was prepared as a white powder
from 3mC-Fmoc (1.53 g, 1.32 mmol).
31P NMR (CDCl3) δ: 149.99, 149.54, 149.28, 149.13; HRMS
(ESI) calcd for C79H78N8O12P [M + H]+: 1361.5471, found
1361.5480.
General procedure for solid phase synthesis of GuNA-modified
oligonucleotides
The synthesis of a 1–2 µmol scale of oligonucleotides modified
with GuNA was performed using the nS-8II oligonucleotide
synthesizer (GeneDesign) according to the standard phosphor-
amidite protocol, except that the coupling time was extended
to eight minutes. The synthesis was carried out in trityl off
mode. 5-Ethylthio-1H-tetrazole (0.25 M) or Activator 42®
(Sigma-Aldrich) was used as the activator. Primer Support 5G
350 (GE Healthcare Bio-Sciences AB) was used as the solid
support. The amidite solutions (0.1 M in acetonitrile) were de-
hydrated using Molecular Sieves Pack 3A (FUJIFILM Wako).
General procedure for deprotection and cleavage of Fmoc-
protected GuNA-modified oligonucleotides
After solid phase synthesis, the support matrix was treated
with 50% triethylamine in acetonitrile. After transferring the
support matrix from the column to a snap vial, 250 μL ethanol
and 750 μL of 28% ammonia was added and the mixture was
heated at 65 °C for eight hours. The solution was then concen-
trated in vacuo to one-half volume. The residue was passed
through a gel permeation column (illustra™ NAP-10™ column
Sephadex™ G-25 DNA Grade, GE Healthcare), and then puri-
fied by reverse-phase HPLC with Waters XBridge™ C18 (4.6 ×
50 mm analytical and 10 × 50 mm preparative) columns using
General procedure for deprotection with TFA and cleavage of
Boc-protected GuNA-modified oligonucleotides
a
linear gradient of methanol in aqueous 400 mM
After solid phase synthesis, the column was charged with
1 mL of 20% (v/v) trifluoroacetic acid in methylene chloride
and allowed to stand at room temperature for one hour. This
solution was then removed, and the column was washed with
methylene chloride and acetonitrile. One milliliter of a solu-
tion of 28% ammonia and ethanol (3 : 1) was added, and the
column was allowed to stand for one hour. This cleavage solu-
tion was transferred into a snap vial. The remaining support
matrix was transferred into a snap vial and 1 mL of the
3 : 1 mixture of 28% ammonia : ethanol was added. Both vials
were heated at 65 °C for five hours. The combined solution
was concentrated in vacuo to one-half volume. The resulting
residue was passed through a gel permeation column (illus-
tra™ NAP-10™ column Sephadex™ G-25 DNA Grade, GE
Healthcare), and then purified by reverse-phase HPLC with a
Waters XBridge™ C18 (10 × 50 mm) column using a linear gra-
dient of methanol in aqueous 400 mM hexafluoroisopropanol/
15 mM triethylamine.
Hexafluoroisopropanol/15 mM triethylamine.
UV melting experiment and melting profile
The UV melting experiments were carried out on a Shimadzu
UV-1650 spectrometer equipped with a Tm analysis accessory.
All
target
strands
were
purchased
(GeneDesign).
Equimolecular amounts of the target RNA or DNA strand and
oligonucleotide were dissolved in 10 mM sodium phosphate
buffer (pH 7.0) containing 100 mM sodium chloride to give a
final strand concentration of
4 µM. The samples were
denatured by heating at 95 °C followed by slow cooling to
room temperature to allow the strands to anneal. The melting
profile was recorded at 260 nm from 0 to 90 °C at a scan rate
of 0.5 °C min−1. The Tm values, the temperatures of half-dis-
sociation of the formed duplexes, were calculated using the
midline method.
Synthesis of 5T
3T-Boc (1220 mg, 1.50 mmol) was divided into two, and tri-
fluoroacetic acid (3.05 mL) was added to both at room temp-
erature and the mixtures were stirred for two hours. The tri-
General procedure for deprotection and cleavage of Teoc-
protected GuNA-modified oligonucleotides
After solid phase synthesis, the support matrix was transferred fluoroacetic acid was removed under reduced pressure. The
from the column to a snap vial. Five hundred microliters of residues were merged and dissolved in water and washed with
0.2 M tetra-n-butylammonium fluoride in tetrahydrofuran was ethyl acetate. The organic layer was extracted with water, and
added, and the mixture was heated at 65 °C for eight hours. the combined aqueous layer was lyophilized. The crude
After cooling to room temperature, 250 μL ethanol and 750 μL mixture was purified by reverse-phase column chromatography
of 28% ammonia solution was added, and the mixture was using Biotage® SNAP Ultra C18 60 g as the column and 0.1%
heated at 65 °C for eight hours. The solution was concentrated aqueous trifluoroacetic acid and methanol as the eluent (90/10
in vacuo to one-half volume. Seventy microliters of 5 M sodium to 40/60) to obtain the trifluoroacetic acid salt of the GuNA
acetate buffer (pH 5.2) and 100 μL of 3 M sodium chloride nucleoside as a white solid. This was subjected to ion
were added, and the mixture was diluted with ethanol. The exchange using the Cl− form of DOWEX™ 1 × 2 100–200 mesh
resulting suspension was allowed to stand at −20 °C for three anion exchange resin and then lyophilized to obtain the hydro-
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
gen chloride salt of the GuNA-T nucleoside 5T (381 mg, yield NMR (D2O) δ: 162.39, 158.73, 151.07, 142.78, 106.60, 92.32,
74%) as a white solid. 88.71, 71.87, 65.78, 59.77, 55.17, 14.84; HRMS (ESI) calcd for
1H NMR (D2O) δ: 7.68 (1H, d, J = 1.3 Hz), 5.74 (1H, s), 4.69 C12H19N6O4 [M + H]+: 311.1462, found 311.1458; elemental
(1H, d, J = 1.3 Hz), 4.37 (1H, s), 4.06 (2H, s), 3.76 (1H, d, J = analysis calcd for C12H18N6O4·2.1 HCl·1.5 H2O: C, 34.82; H,
10.0 Hz), 3.76 (1H, d, J = 10.0 Hz), 1.83 (3H, d, J = 1.0 Hz); 13C 5.63; N, 20.30; Cl, 17.99, found: C, 35.12; H, 5.56; N, 20.29; Cl,
NMR (D2O) δ: 169.34, 158.64, 153.93, 138.60, 113.61, 91.82, 17.84; HPLC purity: 99.8%.
88.30, 71.89, 66.03, 59.81, 55.16, 14.55; HRMS (ESI) calcd for
C12H185O5 [M + H]+: 312.1303, found 312.1298. Elemental ana-
Reverse mutation assay
lysis calcd for C12H17N5O5·HCl·1.2 H2O C, 39.02; H, 5.57; N,
18.96; Cl, 9.60, found: C, 38.96; H, 5.51; N, 18.81; Cl, 9.40;
pre-incubation method with and without metabolic activation
HPLC purity: 99.7%.
Bacterial reverse mutation assays were performed using the
[phenobarbital and 5,6-benzoflavone-induced rat liver S9 frac-
tion (S9 mix)] at concentrations from 9.77 to 5000 μg per plate
based on the Guidelines on Genotoxicity Tests for Drugs.
Synthesis of 5A
3A-Boc (4.13 g, 4.46 mmol) was dissolved in 7 M ammonia in
Briefly, 0.5 mL of 0.1 M sodium phosphate buffer (pH 7.4)
methanol (41 mL) and the solution was stirred at 40 °C for
(without metabolic activation) or 0.5 mL of S9 mix (with meta-
four hours. The solvent was removed in vacuo. The residue was
then dissolved in ethyl acetate and washed with water and 5%
bolic activation) and 0.1 mL of the test substrate solution were
added to a test tube. Then, 0.1 mL of bacterial suspension was
sodium chloride solution successively. The combined aqueous
added and incubated for 20 minutes at 37 °C. After incubation,
layer was extracted with ethyl acetate. The combined organic
the mixture was mixed with 2 mL of top agar and poured onto
layer was dried over sodium sulfate, filtered, and concentrated
a minimal glucose agar plate. After incubated for 48 hours at
in vacuo. The resulting residue was purified by column chrom-
atography (hexane/ethyl acetate, 80/20 to 60/40) to obtain the
crude deprotected intermediate as a white amorphous solid,
which was further treated in the next step.
37 °C, the number of revertant colonies was counted. The five
strains (Salmonella typhimurium TA98, TA100, TA1535, TA1537,
and Escherichia coli WP2uvrA) utilized were obtained from the
National Institute of Technology and Evaluation. If the test
Using a similar procedure as for 5T, the Boc and DMTr
article increased the number of revertant colonies to a level of
groups were removed with trifluoroacetic acid to obtain com-
2-fold or more of the corresponding negative control value, it
pound 5A (1.07 g, yield 63%) as a white solid.
was judged to be mutagenic.
1H NMR (D2O) δ: 8.49 (1H, s), 8.46 (1H, s), 6.34 (1H, s), 5.05
(1H, d, J = 1.0 Hz), 4.69(1H, s), 4.10–4.02 (2H, m), 3.81 (1H, d, J
= 9.8 Hz), 3.73 (1H, d, J = 9.8 Hz); 13C NMR (D2O) δ: 158.74,
Conflicts of interest
153.35, 150.54, 148.23, 144.55, 121.86, 91.84, 87.59, 73.05,
66.44, 55.18; HRMS (ESI) calcd for C12H17N8O3 [M + H]+:
There are no conflicts to declare.
321.1418, found 321.1413; elemental analysis calcd for
C12H16N8O3·2 HCl·1.3 H2O: C, 34.59; H, 4.98; N, 26.89; Cl,
17.02, found: C, 34.55; H, 4.88; N, 26.63; Cl, 17.26; HPLC
purity: 99.9%.
Acknowledgements
We thank Chieko Okagaki, Eri Oboki and Kaori Murakoshi
(Mitsubishi Tanabe Pharma Corporation) for the technical
Synthesis of 5G
Using a similar procedure as for 5A, compound 5G (206 mg,
yield 65%) was prepared as a white solid from 3G-Boc (3.24 g,
2.9 mmol).
support of synthesis and analysis of compounds. A part of this
work was supported by the Platform Project for Supporting
Drug Discovery and Life Science Research (Basis for
Supporting Innovative Drug Discovery and Life Science
Research (BINDS)) from AMED under Grant Number
JP19am0101084.
1H NMR (D2O) δ: 8.56 (1H, s), 6.13 (1H, s), 4.99 (1H, s), 4.04
(1H, s), 3.80 (2H, brs), 3.80–3.70 (2H, m); 13C NMR (D2O) δ:
159.55, 158.78, 157.92, 152.63, 138.45, 114.70, 92.01, 87.49,
72.74, 66.07, 60.11, 55.19; HRMS (ESI) calcd for C12H17N8O4
[M + H]+: 337.1367, found 337.1361; elemental analysis calcd
for C12H16N8O4·1.55 HCl·1.6 H2O: C, 34.18; H, 4.96; N, 26.58;
Cl, 13.03, found: C, 34.31; H, 4.79; N, 26.41; Cl, 12.99; HPLC
purity: 99.7%.
Notes and references
1 T. C. Roberts, R. Langer and M. J. A. Wood, Nat. Rev. Drug
Discovery, 2020, 19, 673–694.
Synthesis of 5mC
2 (a) S. T. Crooke, Antisense Drug Technology: Principles,
Strategies, and Applications, CRC Press, 2nd edn, 2007,
ISBN: 0849387965; (b) V. K. Sharma, R. K. Sharma and
S. K. Singh, MedChemComm, 2014, 5, 1454–1471;
(c) D. R. Scoles, E. V. Minikel and S. M. Pulst, Neurol.:
Genet., 2019, 5, e323.
Using a similar procedure as for 5A, compound 5mC (1.08 g,
yield 98%) was prepared as a white solid from 3mC-Boc (1.96 g,
2.4 mmol).
1H NMR (D2O) δ: 7.85 (1H, s), 5.78 (1H, s), 4.72 (1H, s), 4.31
(1H, s), 4.08–4.01 (2H, m), 3.76–3.64 (2H, m), 2.07 (3H, s); 13C
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
3 A. R. Shrestha, Y. Kotobuki, Y. Hari and S. Obika, Chem.
Commun., 2014, 50, 575–577.
4 (a) H. Hashimoto, M. G. Nelson and C. Switzer, J. Am.
Chem. Soc., 1993, 115, 7128–7134; (b) R. O. Dempcy,
O. Almarsson and T. C. Bruice, Proc. Natl. Acad. Sci. USA,
1994, 91, 7864–7868; (c) A. Blaskó, R. O. Dempcy,
Ceoc group is known to be easily removed by aqueous
piperidine. However, when GuNA was protected with Ceoc,
substitution of the 2-cyanoethanol by piperidine took place
and stable urea was formed. We could not find any other
appropriate deprotection conditions for the Ceoc protective
group.
E. E. Minyat and T. C. Bruice, J. Am. Chem. Soc., 1996, 118, 11 Guanidiylating reagents with Teoc: R. P. Singh, J. Das,
7892–7899; (d) T. P. Prakash, A. Püschl, E. Lesnik,
V. Mohan, V. Tereshko, M. Egli and M. Manoharan, Org.
M. Yousufuddin, D. Gout and C. J. Lovely, Org. Lett., 2017,
19, 4110–4113.
Lett., 2004, 6, 1971–1974; (e) J. Barman, D. Gurav, 12 Guanidiylating reagents with Fmoc: S.-H. Lu, R. Phang and
O. P. Oommen and O. P. Varghese, RSC Adv., 2015, 5, J.-M. Fang, Org. Lett., 2016, 18, 1724–1727.
12257–12260; (f) B. Schmidtgall, A. Kuepper, M. Meng, 13 Guanidiylating reagents with an iB protective group were
T. N. Grossmann and C. Ducho, Chem. – Eur. J., 2018, 24, synthesized in the same manner.
1544–1553; (g) M. Meng and C. Ducho, Beilstein J. Org. 14 N. Horie, S. Kumagai, Y. Kotobuki, T. Yamaguchi and
Chem., 2018, 14, 1293–1308.
S. Obika, Org. Biomol. Chem., 2018, 16, 6531–
6536.
5 J. Ravn, C. Rosenbohm, S. M. Christensen and T. Koch,
Nucleosides, Nucleotides Nucleic Acids, 2006, 25, 843–847.
6 S. K. Singh, R. Kumar and J. Wengel, J. Org. Chem., 1998,
63, 6078–6079.
7 S. K. Singh, R. Kumar and J. Wengel, J. Org. Chem., 1998,
63, 10035–10039.
15 In the 31P-NMR spectra, each amidite showed two to four
signals at around 149 ppm. These signals are considered
come from rotamer (or tautomer) formation of the guani-
dine moiety as well as the chirality at the phosphorus
atom.
8 H. Sawamoto, Y. Arai, S. Yamakoshi, S. Obika and 16 Macugen® Prescribing Information, Gilead Science, Inc., San
E. Kawanishi, Org. Lett., 2018, 20, 1928–1931. Dimas, CA, 2011.
9 (a) R. S. Basic, C. Aspects, S. J. W. Liss, T. M. Science, 17 C. L. Berman, S. A. Barros, S. M. Galloway, P. Kasper,
A. M. Krieg, A. Yi, S. Matson, T. J. Waldschmldtt,
F. B. Oleson, C. C. Priestley, K. S. Sweder, M. J. Schlosser
G. A. Bishop and R. Teasdale, Nature, 1995, 374, 546–549;
and Z. Sobol, Nucleic Acid Ther., 2016, 26, 73–85.
(b) U. Ohto, T. Shibata, H. Tanji, H. Ishida, E. Krayukhina, 18 D. M. Maron and B. N. Ames, Mutat. Res., 1983, 113,
S. Uchiyama, K. Miyake and T. Shimizu, Nature, 2015, 520,
702–705.
10 2-Cyanoethoxycarbonyl (Ceoc) group has been reported as
a suitable protective group for guanidine moiety.4d,e The
173.
19 D. Gatehouse, S. Haworth, T. Cebula, E. Gocke, L. Kier,
T. Matsushima, C. Melcion, T. Nohmi, T. Ohta, S. Venitt
and E. Zeiger, Mutat. Res., 1994, 312, 217.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020