Synthesis of Trifluoromethylated Purine Ribonucleotides and Their Evaluation as 19F NMR Probes

Article abstract of DOI:10.1021/acs.joc.9b03198

Protected guanosine and adenosine ribonucleosides and guanine nucleotides are readily functionalized with CF₃ substituents within the nucleobase. Protected guanosine is trifluoromethylated at the C8 position under radical-generating conditions in up to 95% yield and guanosine 5′-oligophosphates in up to 35% yield. In the case of adenosine, the selectivity of trifluoromethylation depends heavily on the functional group protection strategy and leads to a set of CF₃-modified nucleosides with different substitution patterns (C8, C2, or both) in up to 37% yield. Further transformations based on phosphorimidazolide chemistry afford various CF₃-substituted mono- and dinucleoside oligophosphates in good yields. The utility of the trifluoromethylated nucleotides as probes for ¹⁹F NMR-based real-time enzymatic reaction monitoring is demonstrated with three different human nucleotide hydrolases (Fhit, DcpS, and cNIIIB). Substrate and product(s) resonances were sufficiently separated to enable effective tracking of each enzymatic activity of interest.

Full text of DOI:10.1021/acs.joc.9b03198

Subscriber access provided by FONDREN LIBRARY, RICE UNIVERSITY
Article
Synthesis of trifluoromethylated purine ribonucleotides
and their evaluation as 19F NMR probes
Miko#aj Chromi#ski, Marek Rafa# Baranowski, Sebastian
Chmielinski, Joanna Kowalska, and Jacek Jemielity
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.9b03198 • Publication Date (Web): 29 Jan 2020
Downloaded from pubs.acs.org on February 2, 2020
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 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Synthesis of trifluoromethylated purine ribonucleotides and their evaluation as 19F NMR
probes
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
Mikolaj Chrominski,1 Marek R. Baranowski,2 Sebastian Chmielinski,1 Joanna Kowalska,2 Jacek
Jemielity1*
1Centre of New Technologies, University of Warsaw, S. Banacha 2c, 02-097 Warsaw, Poland.
2Division of Biophysics, Institute of Experimental Physics, Faculty of Physics, University of Warsaw,
Pasteura 5, 02-093 Warsaw, Poland.
Abstract
Protected guanosine and adenosine ribonucleosides and guanine nucleotides are readily
functionalized with CF
3
substituents within the nucleobase. Protected guanosine is trifluoromethylated
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 39
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
at the C-8 position under radical-generating conditions in up to 95% yield, and guanosine 5’-
oligophosphates in up to 35% yield. In the case of adenosine, the selectivity of trifluoromethylation
depends heavily on the functional group protection strategy and leads to a set of CF
nucleosides with different substitution patterns (C-8, C-2, or both) in up to 37% yield. Further
transformations based on phosphorimidazolide chemistry afford various CF -substituted mono- and
3
modified
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
3
dinucleoside oligophosphates in good yields. The utility of the trifluoromethylated nucleotides as
probes for 19F NMR-based real-time enzymatic reaction monitoring is demonstrated with three
different human nucleotide hydrolases (Fhit, DcpS, and cNIIIB). Substrate and product(s) resonances
were sufficiently separated to enable effective tracking of each enzymatic activity of interest.
Introduction
Inserting a trifluoromethyl group (CF
for fine-tuning the properties of drug candidates1-3 and designing molecular probes for 19F nuclear
magnetic resonance (NMR) experiments.4, 5 The placement of the CF
3
) into an organic molecule is a common and effective approach
3
substituent in an organic
compound can affect the neighboring functional groups and alter its overall polarity, acid-base
properties, reactivity, and many other properties. Consequently, CF functionalization has been
3
explored in medicinal chemistry as a strategy for modulating the biological activities of drug candidates
by influencing metabolic stability, conformational equilibrium, lipophilicity, pharmacodynamics, and
pharmacokinetics.1-3
19
The F nucleus has several properties that are beneficial for NMR spectroscopy, including spin of 1/2,
1
one of the highest magnetic sensitivities (83% that of H), and high abundance (100%). Moreover, the
wide chemical shift range (up to 400 ppm) and the absence of fluorine in natural compounds (low
physiological content) simplify 19F NMR spectral analysis even for complex biomolecular mixtures. An
additional advantage of using CF group as a biomolecular NMR tag (instead of a single fluorine atom,
3
for example) is the presence of three equivalent F atoms, which increases sensitivity. As such,
fluorinated organicmolecules and biopolymers have found application in NMR-based ligand screening
ACS Paragon Plus Environment

Page 3 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
assays such as FAXS (fluorine chemical shift anisotropy and exchange for screening)6, 7 and FABS
(
fluorine atoms for biochemical screening),7, 8 enzymatic assays, protein and nucleic acid structure
9
studies, and others. Consequently, there is a high demand for synthetic transformations that assure
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
the efficient and robust preparation of CF -containing building blocks, preferably through late stage,
3
site-selective direct trifluoromethylation. Recent years have seen extensive developments in direct C-
H trifluoromethylation methodologies, including electrophilic/nucleophilic reactions,10-14 photoredox-
based reactions,15 metal-mediated reactions,16 and radical reactions with various “CF
” sources.11,
3
17
Fluorinated nucleosides, nucleotides, and oligonucleotides18 are of particular interest as anticancer
and antiviral compounds or probes for studying nucleic acids structure, interactions, and biological
transformations.19-31 However, the synthesis of fluorinated nucleic acid components often poses
complex challenges, especially in the case purine derivatives.
Consequently, few methods that access trifluoromethylated purine nucleosides have been reported;
hence the properties of these compounds are underexplored. One of the early strategies involves the
32
use of CF
3
-containing building blocks for the de novo synthesis of the six-membered purine ring. This
33
34
approach has been applied by Langer and Pankiewicz for the preparation of CF
3
-substituted purine
nucleosides as enzymatic inhibitors. The trifuoromethylation of halogenated purine ribosides was
introduced by Kobayashi (Scheme 1),35 and requires the preparation of a CF
3
-copper complex as a CF
3
source followed by reaction with an O-protected nucleoside bearing a halogen substituent at the C8-
position. Although this process is multistep, operationally demanding, and afforded moderate yields,
its modified variants have become methods of choice for preparing purine ribosides
trifluoromethylated at positions C8,36, 37 C2,38-40 and C6.41
Montesarchio reported the direct trifluoromethylation of canonical nucleoside derivatives under mild
conditions (Scheme 1)42 in moderate yields by taking advantage of CF
Na as a CF -radical
3
SO
2
3
43
precursor. The substrate scope included deoxyguanosine, deoxyadenosine, and inosine.
Scheme 1. Synthetic approaches toward trifluoromethylated purines.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 39
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
Baran reported (CF
3
SO
2
)
2
Zn as a versatile reagent for the C-H trifluoromethylation of heterocycles,44
including drugs bearing the purine motif.45 Inspired by this work, herein we aimed to develop
CF SO Zn-based protocols for the synthesis of protected and unprotected purine ribonucleosides
(
3
)
2 2
and ribonucleotides. As a result we report the synthesis of trifluoromethylated adenine and guanine
nucleosides, nucleotides, and dinucleotides and evaluate their potential as 19F NMR probes for
enzymatic reactions.
ACS Paragon Plus Environment

Page 5 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Results and Discussion
In a pilot experiment, we subjected guanosine (1, Table1, entry 1) to the conditions described by
44
Baran. The reaction was selective, although the isolated yield of 8-trifluoromethylguanosine (2) was
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
low (5%), even at elevated temperature (Table 1, entry 2). The effect of solvent mixtures (including
two phase system) to the reaction was next evaluated.Application of two phase DCM/water system
resulted in no product formation (Table 1, entry 3), while DMSO/water and DMSO/10% AcOH (which
was recently reported by Parish and Krska to be optimal solvent for trifluoromethylation of complex
compounds, such as peptides 46) mixtures raised the yield to 20% and 25%, respectively (Table 1,
entries 4 and 5). The highest conversion was achieved in 10% AcOH/DMSO with t-BuOOH added at 0
°C (Table 1, entry 6); 2 was isolated in 40% yield. While the reaction was selective for the expected
product, the starting material was not fully consumed.
Table 1. Optimisation of trifluoromethylation of guanosinea
d
Entry
Solvent
DMSO
Time [h]
72
Temperature [°C]
Yield [%]
1
2
3
4
5
6
rt
60b
rt
5
7
DMSO
24
DCM/water, 1/1
DMSO/water, 1/1
DMSO/10% AcOH, 1/1
DMSO/10% AcOH, 1/1
72
0
72
rt
20
25
40
72
rt
72
0 to rtc
a
General conditions: guanosine (0.2 mmol), solvent (2 ml), (CF
70% solution in water, 1 mmol).
3
SO ) Zn dihydrate (0.6 mmol), t-BuOOH
2
2
(
b
The heating was started after addition of whole t-BuOOH solution.
The solution of guanosine and (CF SO Zn was cooled to 0 °C prior addition of t-BuOOH solution. This
c
3
)
2 2
temperature was maintained until whole t-BuOOH was added.
d
Isolated yield.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 39
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
We next examined the application of similar conditions to the trifluoromethylation of guanosine 5’-
oligophosphates (3-5, Table 2). As observed for guanosine, higher yields were obtained in AcOH/DMSO
at low temperatures (Table 2, entries 1-4), and increasing the AcOH concentration resulted in
significant drop of yields (Table 2, entries 4-6). The optimized conditions afforded 8-trifluoromethyl-
GMP (6) in 35% isolated yield (Table 2, entry 4), while GDP 4 and GTP 5 afforded the corresponding
products 7 and 8 in 25% and 15% isolated yields, respectively (Table 2, entries 8 and 9). Prolonged
reaction times led to the decomposition of products 7 and 8 through pyrophosphate bond hydrolysis.
It is worth to mention that the reaction outcome was insensitive towards the counter ion of the
nucleotide (Table 2, entry 4 vs entry 7). Surprisingly, adenosine, adenosine monophosphate, and
diadenosine 5’,5’-triphosphate gave only traces of the trifluoromethylated compounds under the
conditions examined.
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
Table 2. Optimisation of trifluoromethylation of guanosine 5’-oligophosphatesa
d
Entry Substrate
Solvent
Temperature [°C]
Product
Yield [%]
traces
< 5
1
2
3
3
3
3
3
3
3
4
DMSO
rt
6
6
6
6
6
6
6
7
DMSO/water, 1/1
DMSO/10% AcOH, 1/1
DMSO/10% AcOH, 1/1
DMSO/20% AcOH, 1/1
DMSO/30% AcOH, 1/1
DMSO/10% AcOH, 1/1
DMSO/10% AcOH, 1/1
rt
3
rt
15
4
0 to rt
0 to rt
0 to rt
0 to rt
0 to rt
35
5
20
6
10
7b
8c
35
25
ACS Paragon Plus Environment

Page 7 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
9c
5
DMSO/10% AcOH, 1/1
0 to rt
8
15
a
General conditions: triethyl ammonium salt of respective nucleotide (0.1 mmol), solvent (1 ml),
(CF
3
SO ) Zn dihydrate (0.3 mmol), t-BuOOH (70% solution in water,0.5 mmol).
2
2
b
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
sodium salt of 3 was used.
c
reaction was stopped after 24 h due to the detection of decomposition products in the reaction
mixture (HPLC analysis).
d
Isolated yield.
Although the conditions developed for guanosine and its derivatives appear acceptable, incomplete
conversion led to tedious product isolation. Moreover, the adenosine problem remained unsolved. To
improve conversions and to simplify isolation procedures, we examined the reactions of protected
nucleosides, since this strategy has been reported effective for purine 2’-deoxynucleosides42.
2
',3',5'-Tri-O-acetylguanosine (9, Scheme 2) turned out to be an excellent starting material for this
transformation, giving 10 in 95% isolated yield. The product was easily isolated by extraction followed
by simple silica-gel chromatography to remove trace impurities. Treatment of 10 with MeNH /EtOH
2
solution afforded pure 2 in nearly quantitative yield by solvent evaporation; 2 was then transformed
into monophosphate 6 under Yoshikawa conditions47 in the presence of lutidine (to avoid
depurination). This reaction provided 6 in 90% yield after ion-exchange chromatography. Since all
reactions in this sequence were selective and high yielding, we decided to repeat it without isolating
the intermediates by chromatography, which resulted in 85% overall yield of 6 from 9 over three steps
(
Scheme 2), which is practical from a preparative perspective.
Monophosphate 6 was converted into its P-imidazolide derivative 6-Im by reacting it with imidazole in
the presence of 2,2’-dithiodipyridine/triphenylphosphine (Scheme 2);48 6-Im decomposed upon
storage, but when used directly after preparation, it afforded good yields in further transformations.
For example, when 6-Im was reacted with triethylammonium phosphate or pyrophosphate in the
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 39
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
presence of ZnCl
2
, diphosphate 7 or triphosphate 8 were produced in satisfactory yields, respectively
(Scheme 2).
Scheme 2. The synthesis of 8-trifluoromethylguanosine and subsequent transformations.
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
Protecting the OH groups of guanosine significantly improved the yield and simplified product
isolation; consequently, we applied the same approach for adenosine. Indeed, when 2',3',5'-tri-O-
acetyladenosine (11, Scheme 3) was subjected to similar conditions, the 8-trifluoromethylated product
12 was obtained in 37% yield. Further manipulation of the reaction conditions did not improve this
result.
6
A complex mixture of products was obtained when N ,2',3',5'-tri-O-tetraacetyladenosine 14 was
subjected to similar conditions (Scheme 3). Chromatographic separation and analysis revealed the
presence of three trifluoromethylated products: the C8- and C2-trifluoromethyladenosine derivatives
1
5 (9%) and 16 (10%), and the C2,C8-disubstituted adenosine derivative 17 (7%). To the best of our
knowledge, 17 is the first reported example of an adenosine derivative bearing two trifluoromethyl
substituents on the purine ring. Yields were not improved nor was the product distribution significantly
altered by changing the reaction conditions (temperature, solvent). Treating compounds 12, and 15–
ACS Paragon Plus Environment

Page 9 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
7 with 33% MeNH
2
/EtOH afforded nucleosides 13, 18, and 19, which were phosphorylated to their
respective monophosphates 20–22 in good yields (Scheme 3).
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
Scheme 3. The synthesis of trifluoromethyladenosine 5’-monophosphates.
Monophosphate 20 was converted into its P-imidazolide derivative 20-Im (Scheme 4), which enabled
the efficient elongation of the phosphate chain to give 8-CF adenosine diphosphate 23 and
3
triphosphate 24 in good yields.
Scheme 4. The synthesis of 8-trifloromethyladenosine oligophosphates.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 39
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
Finally, we prepared a series of trifluoromethylated dinucleoside 5',5'-triphosphates using some of the
synthesized mononucleotides (Scheme 5). This was achieved by coupling together two nucleotide
building blocks, one of which is activated as a phosphorimidazolide, in the presence of excess ZnCl to
2
form a new pyrophosphate bond. The reaction of imidazolide 20-Im with diphosphate 23 led to a
complex reaction mixture containing the desired dinucleotide 25, the monophosphate 20 (from the
hydrolysis of 20-Im), and the coupling product of 20-Im and 20. Nevertheless, we isolated the desired
dinucleotide 25 in 20% yield. The reaction of 20-Im with a sub-equimolar amount of triethylammonium
phosphate proved to be more efficient; the in situ formed diphosphate 23 readily reacted with excess
2
0-Im present in the reaction mixture to yield 25 in 63%. The reaction of imidazole-activated adenosine
monophosphate ADP-Im with monophosphate 6 led to dinucleotide 26 in 40% yield, which could not
be improved by reacting 6-Im with adenosine diphosphate. On the other hand, 6 reacted smoothly
with GDP-Im to give dinucleotide 27 composed of two guanosines, one of which bears an 8-CF
3
group.
The presence of the electron-withdrawing 8-CF group in the guanosine structure dramatically
3
decreased the nucleophilicity of the neighboring (N7) nitrogen. The drop in reactivity was evident
when 27 was treated with iodomethane affording exclusively mRNA 5' cap analogue 28, which was
site-selectively methylated at the guanosine rather than 8-CF -guanosine.
3
ACS Paragon Plus Environment

Page 11 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Scheme 5. The synthesis of trifluoromethylated dinucleotides.
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
To evaluate the usefulness of trifluoromethylated purine nucleotides as probes for enzymatic activity
monitoring by 19F NMR, we subjected select compounds to three nucleotide-specific
phosphohydrolases, hFhit, hDcpS, and hcNIIIB. These enzymes are of interest due to their function in
regulation of endogenous nucleotide metabolism and links to disease development, and as such,
assays have been developed that allow monitoring their activity and the discovery of inhibitors. These
include assays involving radioactivity,49, 50, fluorogenic probes,31 malachite green (MG) assay, (2) and
fluorescent FRET probes.52, 53(3) However, these assays have some limitations such as discontinuity,
susceptibility to interference from UV-VIS absorbing and emitting inhibitors, or high structural
complexity of the probes. Thus, methods that enable robust and straightforward real-time monitoring
of activity these enzymes are still desired.
51
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 39
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
We first used compounds 25–27 to monitor the activity of the human fragile histidine triad (hFhit)
pyrophosphatase, which unsymmetrically cleaves diadenosine 5’,5’-triphosphate (Ap A) and other
3
54
purine dinucleotides. hFhit is considered to be a tumor suppressor, and its function has been linked
to substrate binding. Hence, efforts have been made to identify hFhit inhibitors,52, 53 but not by 19
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
F
NMR spectroscopy. We verified that hFhit accepts bis(trifluoromethylated) Ap A analog 25 as a
3
19
substrate. Compound 25 exhibits a single narrow signal at -61.38 ppm in its F NMR spectrum, which
corresponds to six equivalent fluorine atoms (Figure 1A). A decrease in intensity of this resonance was
observed after adding 20 nM hFhit to a 100 μM buffered solution of 25, and two new slightly upfield-
shifted signals appeared, consistent with substrate hydrolysis to give 20 (8-CF
3
-AMP, δ
F
-61.39) and 23
19
(
8-CF -ADP, δ -61.40; Figure S1A). The signals of 20 and 23 were assigned on the basis of the F NMR
3
F
spectra of synthetic references (Figure S1B-D). Despite the small differences in chemical shifts from
the substrate, it was convenient to monitor the reaction progress by 19F NMR and observe the
inhibitory effect of a previously identified compound53 (Figure 1A). Similar experiments were
performed for unsymmetrical 8-trifluoromethylguanosine-containing dinucleotides 26 and 27 (100 μM
19
each in the presence of 20 nM or 25 nM hFhit, respectively; Figure S2). F NMR spectroscopy revealed
the formation of two trifluoromethylated products from each compound, namely 8-CF -GDP (7; δ
1.28) and 8-CF -GMP (6; δ -61.27), albeit in different proportions (Figures S3). Under these
3
F
-
6
3
F
comparable conditions, 25 and 26 are preferred substrates of hFhit, rather than 27, consistent with
the known preference of the enzyme for adenine-containing dinucleotides.
We next investigated decapping scavenger (hDcpS) and cytosolic nucleotidase IIIB (hcNIIIB), two
enzymes involved in the cellular metabolism of N7-methylguanine nucleotides. hDcpS degrades cap
7
) released during 3’-to-5’ mRNA degradation55 and has been identified as a
moieties (m GpppN
n
therapeutic target for spinal muscular atrophy and acute myeloid leukemia, which created demand for
inhibitors.49, 50, 56, 57 To verify that trifluoromethylated nucleotides can be used to study hDcpS activity,
mRNA cap analog 28 (100 µM;) was incubated with 80 nM enzyme in the absence and presence of
RG3039, a potent inhibitor. 19F NMR analysis revealed that the substrate (δ
-61.22) is site-selectively
F
ACS Paragon Plus Environment

Page 13 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
cleaved to release 7 (8-CF
methylguanosine, which controls the regioselectivity; inhibition by RG3039 was also clearly visible
Figure 1B).
3
-GDP; δ -61.25), consistent with the high specificity of DcpS for 7-
F
(
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
19
Figure 1. Monitoring enzymatic activity by F NMR spectroscopy. (A) 25 with hFhit, (B) 28 with hDcpS,
and (C) 6 with hcNIIIB. For experimental details see Supporting Information.
7
58
hcNIIIB dephosphorylates m GMP to 7-methylguanosine and its inhibitors are potential modulators
51
of mononucleotide metabolism and downstream RNA degradation pathways. hcNIIIB also hydrolyses
electron-poor pyrimidine nucleotides, whereas GMP and AMP are very poor substrates.59 Since
trifluoromethylation decreases the electron density in the purine, we tested 8-CF -GMP (6) as an
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 39
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
artificial substrate for hcNIIIB. Indeed, the substrate peak (δ
F
-61.27) was observed to disappear when
-61.17) emerged, which was
6
at 100 μM was incubated with 120 nM hcNIIIB, and a product signal (δ
F
independently confirmed to be 8-trifluoromethylguanosine (2). The reaction was almost completely
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
stopped by hcNIIIB-specific inhibitor (Figure 1C).
Conclusion
In summary, we optimized the conditions for the synthesis of trifluoromethylated purine nucleotides
and nucleosides using (CF
3
SO
2
)
2
Zn as a source of CF radicals. The synthe-sized compounds include
3
trifluoromethylguanosine and trifluormethyladenosine, their 5’-mono, di-, and triphos-phates, as well
as several dinucleoside 5’,5’-triphosphates. The synthetized trifluoromethylated (di)nucleotides were
successfully used as molecular probes to monitor the activ-ities of three enzymes (hFhit and hDcpS
pyrophosphatases and hcNIIIB phosphatase) by 19F NMR spectroscopy. The introduction of CF
3
moieties into the purines in dinucleotide analogs does not prevent specific recognition by either hFhit
7
or hDcpS. Interestingly, 8-CF
3
-GMP (6) acted as a m GMP mimic, as manifested by its efficient
dephosphory-lation by cNIIIB. Substrate and product(s) resonances were sufficiently separated to
enable effective monitoring of the enzymatic activity of interest, which opens possibilities for the
19
development of F NMR-based inhibitor-discovery and evaluation assays. We envisage that the higher
syn-thetic availability of trifluoromethylated guanine- and adenine-derived building blocks offered by
our work will also pave the way for their use in more-complex biomolecular systems, such as
oligonucleotides and nucleic acids, thereby facilitating studies on nucleic acid structure and function.
Experimental Section
General information. All commercial reagents and solvents were used as received without additional
purification. Guanosine and adenosine were purchased from Carbosynth. (CF
3
SO ) Zn dihydrate and
2
2
tBuOOH (70% in water) were purchased from TCI. Anhydrous solvents were purchased from Sigma-
Aldrich. Thin Layer Chromatography (TLC) analysis was carried out on pre-coated Silica Gel 60 Å on
aluminum foil with fluorescence indicator (Sigma-Aldrich) and visualised under UV lamp (254 nm).
ACS Paragon Plus Environment

Page 15 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Preparative chromatography. Preparative chromatography (SiO and RP C18) was performed using
2
Reveleris X2 flash chromatography system (BUCHI) with FlashPure cartridges (4 g, 12 g, 24 g, 40 g).
Conditioning methods, loading and flow rates were set according to producers guidelines. UV
detection was performed at 3 wavelengths (254 nm, 265 nm and 280 nm) simultaneously.
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
Ion-exchange chromatography. The synthesized nucleotides were purified by ion-exchange
3-
chromatography on DEAE Sephadex A-25 (HCO form) column. After loading the column with reaction
mixture and washing it with water, the products were eluted using different linear gradients of
triethylammonium bicarbonate (TEAB) in deionized water: 0–0.7 M for nucleoside monophosphates,
0
–1.0 M for nucleoside diphosphates and dinucleotides or 0–1.2 M for nucleoside triphosphates.
Fractions containing desired product were collected together after RP HPLC and spectrophotometric
at 260 nm) analysis. Evaporation under reduced pressure with repeated additions of 96% and then
9.8% ethanol resulted in isolation of nucleotide analogues as triethylammonium salts.
(
9
Analytical and preparative HPLC. Analytical HPLC was performed on Agilent Tech. Series 1200 using
Supelcosil LC-18-T HPLC column (4.6 x 250 mm, flow rate 1.3 mL/min) with linear gradients of methanol
in 0.05 M ammonium acetate buffer and UV-detection at 254 nm. Analytical HPLC programs included
in supporting informations. Semi-preparative HPLC was performed on the same apparatus equipped
with Discovery RP Amide C-16 HPLC column (25 cm x 21.2 mm, 5μm, flow rate 5.0 mL/min) with linear
gradients of MeCN in 0.05 M ammonium acetate buffer (pH 5.9) and UV-detection at 260 nm.
Spectroscopic analysis of the synthesized compounds. The structure and purity of compounds
product were confirmed high resolution mass spectrometry using electrospray ionization (HRMS ESI),
and NMR spectroscopy. Purity of water soluble compounds was additionally confirmed by RP HPLC.
Mass spectra were recorded on Thermo Scientific LTQ OrbitrapVelos (high resolution spectra) and AB
Sciex API 3200 (low resolution spectra) spectrometers. NMR spectra were recorded on a Varian INOVA
4
00 MHz or 500 MHz spectrometer equipped with a high stability temperature unit using 5 mm 4NUC
probe at 25 °C. The chemical shifts were reported in ppm with residual solvent peak as internal
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 39
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
standard. The 31P NMR chemical shifts were reported in ppm and referenced to 20% phosphoric acid
in D
for spectra recorded in CDCl
recorded in D O, -121.50 ppm) as external standard. Signal assignments of compounds 2, 13, 16 and
7 were based on COSY, HSQC and HMBC spectra analysis
Optimisation of trifluoromethylation of guanosine, general procedure. Guanosine (56 mg, 0.2 mmol)
was dissolved/suspended in the solvent/solvent mixture (2 ml) followed by addition of (CF SO Zn
2
O as an external standard. The 19F chemical shifts were reported in ppm and referenced to CFCl
3
(
3
and DMSO-d6 , 0.65, -0.24 ppm respectively) or NaF (for spectra
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
1
3
)
2 2
dihydrate (219 mg, 0.6 mmol). To this mixture t-BuOOH (70% solution in water, 130 l, 1 mmol) in 10
aliquots (13 l each) in 20 min. intervals was added upon vigorous stirring. The reaction mixture was
stirred at room temperature. The progress of the reaction was monitored by TLC analysis (10% MeOH
in DCM) until no further progress could be detected. After the indicated time the reaction mixture was
diluted with water (approx. 20 ml) and extracted with DCM (3 x 10 ml). The combined organic fractions
were washed with water (approx. 20 ml), brine (approx. 20 ml) and dried over Na
2
SO . The drying agent
4
was filtered off, washed with DCM and the filtrate concentrated in vacuo. To the residue 10 ml of
MeOH and 0.5 g of silica was added. The slurry was concentrated in vacuo and the dry residue was
loaded on the preconditioned silica gel column (4 g). The product was eluted with the mixture of MeOH
in DCM (0 to 10% linear gradient). The fractions containing the desired product were combined,
concentrated in vacuo, co-evaporated with diethyl ether (approx. 5 ml) and dried overnight under high
vacuum giving 2 as off-white solid. Isolated yields and additional information are summarised in Table
1
5
. 1H NMR (400 MHz, DMSO-d
6
)  11.00 (bs, 1H, NH), 6.71 (bs, 2H, -NH ), 5.63 (d, J = 6.2 Hz, 1H, C1’),
2
.49 (d, J = 6.22 Hz, 1H, C2’-OH), 5.11 (d, J = 4.9 Hz, 1H, C3’-OH), 5.05 (d, J = 5.9 Hz, 11.7 Hz, 1H, C2’),
4.91 (t, J = 6.0 Hz, 1H, C5’-OH), 4.16 (m, 1H, C3’), 3.89 (m, 1H, C4’), 3.71-3.63 (m, 1H, C5’), 3.57-3.49
13
(
m, 1H, C5’); C {H} NMR (100 MHz, DMSO-d
6
)  156.5 (C6), 154.4 (C2), 152.5 (C4), 133.3 (q, J = 39.0
Hz, C8), 118.6 (q, J = 269.9 Hz, CF
3
), 116.5 (C5), 89.4 (C1’), 86.4 (C4’), 70.7 (C2’), 70.6 (C3’), 61.9 (C5’);
19F NMR (376 MHz, DMSO-d
) -59.87; HRMS (+) ESI m/z: [M+H] calcd for C11
+
O 352.0863
+
6
13
H F
3
N
5
5
found 352.0861,.
ACS Paragon Plus Environment

Page 17 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Optimisation of trifluoromethylation of guanosine phosphates, general procedure. A triethyl
ammonium salt of respective nucleotide (0.1 mmol) was dissolved in the indicated solvent (1 ml)
followed by addition of (CF
3
SO ) Zn dihydrate (110 mg, 0.3 mmol). To this mixture t-BuOOH (70%
2
2
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
solution in water, 65 l, 0.5 mmol) in 10 aliquots (6.5 l each) in 20 min intervals was added under
vigorous stirring. The progress of the reaction was monitored by RP HPLC analysis until no further
progress could be detected (usually 72 h) or decomposition of the starting material/products was
detected. The reaction mixture was diluted with EDTA solution (100 mg in 10 ml of water) and
neutralised with 10% NaHCO
3
. The resulting mixture was loaded on DEAE Sephadex A-25 column
3-
(
HCO form, 10 g), the column was washed with water (50 ml) and then eluted using TEAB in deionized
water (400 ml, linear gradient). The fractions containing the mixture of the desired product and starting
material (UV and RP HPLC analysis) were combined, concentrated in vacuo, co-evaporated with 96%
EtOH (approx. 50 ml), co-evaporated with MeCN (approx. 50 ml) and the residue was dried overnight
under high vacuum. The product was separated from starting material and possible by products using
preparative RP HPLC chromatography with linear gradient of MeCN in 0.05 M ammonium acetate
buffer (pH = 5.9). The fractions containing pure product were concentrated in vacuo, co-evaporated
with 96% EtOH (approx. 50 ml) and co-evaporated with MeCN (approx. 50 ml). The residue was
dissolved in MQ water and freeze-dried. After repeated freeze-drying, product was isolated as
ammonium salt. Isolated yields and additional information are summarised in Table 2.
1
8
=
1
-trifluoromethylguanosine 5’-monophosphate (6). White foam. H NMR (400 MHz, D
2
O) = 5.89 (d, J
6.1 Hz, 1H), 5.33 (t, J = 5.8 Hz, 1H), 4.59 (dd, J = 3.6 Hz, 5.7 Hz, 1H), 4.31-4.23 (m, 1H), 4.22-4.09 (m,
H); 13C {H} NMR (125 MHz, D
2
O)  161.0, 156.7, 155.5, 138 (q, J = 40.1 Hz), 120.6 (q, J = 270.3 Hz),
118.1, 91.9, 86.7, 73.7, 72.8, 67.20; 19F NMR (376 MHz, D
O)= -61.23; 31P NMR (162 MHz, D
O) =
2
2
-
P- 430.0381; found 430.0380,.
1
.75; HRMS (-) ESI m/z: [M-H] calcd for C11
H
12
3
F N
5
O
8
1
8
-trifluoromethylguanosine 5’-diphosphate (7). White foam. H NMR (400 MHz, D
2
O) = 5.90 (d, J =
6
.2 Hz, 1H), 5.38 (t, J = 5.9 Hz, 1H), 4.67 (dd, J = 3.1 Hz, 5.5 Hz, 1H), 4.35-4.27 (m, 2H), 4.26-4.17 (m,
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 39
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
1
H); 19F NMR (376 MHz, D
2
O)  -61.21; 31P NMR (162 MHz, D
2
O) = -9.85 (d, J = 20.7 Hz, 1P), -10.29
-
-
(dt, J = 5.9 Hz, 20.7 Hz, 1P); HRMS (-) ESI m/z: [M-H] calcd for C11
H F
13 3
5
N O
11
P
2
510.0044; found:
5
10.0041,.
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
8
-trifluoromethylguanosine 5’-triphosphate (8). White foam. H NMR (400 MHz, D
2
O) = 5.90 (d, J =
6.2 Hz, 1H), 5.39 (t, J = 5.9 Hz, 1H), 4.69 (dd, J = 2.3 Hz, 5.3 Hz, 1H), 4.38-4.22 (m, 3H); 19F NMR (376
31
MHz, D
2
O)= -61.20; P NMR (162 MHz, D
2
O) = -9.82 (d, J = 19.3 Hz, 1P), -10.35 (d, J = 19.7 Hz, 1P),
-
-
-22.10 (t, J = 19.5 Hz, 1P); HRMS (-) ESI m/z: [M-H] calcd for C11
14
H F
3
N
5
O
14
P 589.9708; found
3
5
89.9708.
',3',5'-tri-O-acetyl-8-trifluoromethylguanosine (10). 2',3',5'-tri-O-acetyl-guanosine (9, 818 mg, 2
SO Zn dihydrate (2200 mg, 6 mmol)
2
mmol) was dissolved in DMSO (12 ml) followed by addition of (CF
3
)
2 2
upon vigorous stirring at room temperature. When the clear solution was formed (15-20 min), t-
BuOOH (70% solution in water, 1.3 ml, 10 mmol) in 10 aliquots (130 l each) in 20 min. intervals. During
addition of t-BuOOH the reaction mixture started to became yellow. The reaction mixture was stirred
for 24h after which TLC analysis (3% MeOH in DCM) indicated full consumption of the starting material.
The reaction mixture was poured into 200 ml of water and extracted with DCM (3 x 50 ml). The
combined organic layers were washed with water (3 x approx. 100 ml), brine (approx. 100 ml) and
dried over Na
2
SO . The drying agent was filtered off, washed with DCM and the filtrate was
4
concentrated in vacuo. The oily residue was dissolved in small amount of DCM and loaded on the
preconditioned silica gel column (12 g). The product was eluted with the mixture of MeOH in DCM (5%,
v/v). The fractions containing the product were combined, concentrated in vacuo, co-evaporated with
diethyl ether (approx. 10 ml) and dried overnight under high vacuum giving 10 as off-white solid (906
1
mg, 95%). H NMR (400 MHz, CDCl
3
)  6.36 (bs, 1H), 5.95 (d, J = 4.6 Hz, 1H), 5.94-5.89 (m, 1H), 4.56-
4.50 (m, 1H), 4.47-4.36 (m, 2H), 2.15 (s, 3H), 2.11 (s, 3H), 2.04 (s, 3H); 19F NMR (376 MHz, DMSO-d
)
6
+
+
 - 61.19; HRMS (+) ESI m/z: [M+H] calcd for C17
H
19
F N
3
5
O 478.1180; found 478.1187.
8
ACS Paragon Plus Environment

Page 19 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Deprotection of 10. 10 (906 mg, 1.9 mmol) was placed in 50 ml round bottom flask equipped with
rubber septum and flushed with stream of argon. Then MeNH (33% in EtOH, 10 ml) was added under
2
gentle flow of argon and the resulting mixture was stirred for 4 h at room temperature after which TLC
analysis (5% MeOH in DCM) indicated full consumption of the starting material. The reaction mixture
was concentrated in vacuo. To the residue approx. 20 ml of MeOH and 2 g of silica was added. The
slurry was concentrated in vacuo and the dry residue was loaded on the preconditioned silica gel
column (10 g). The product was eluted with the mixture of MeOH in DCM (10%, v/v). The fractions
containing the desired product were combined, concentrated in vacuo, co-evaporated with diethyl
ether (approx. 10 ml) and dried overnight under high vacuum giving 2 as yellowish solid (634 mg, 95%).
The analytical data matched those obtained for 2 synthesized by direct trifluoromethylation of
guanosine.
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
8
-trifluoromethylguanosine 5’-monophosphate (6). 8-trifluoromethylguanosine (2, 351 mg, 1 mmol)
was dissolved in anhydrous (MeO) PO (10 ml) under gentle flow of argon. The resulting solution was
cooled below 0°C (ice/brine bath) followed by addition of 2,6-lutidine (350 l, 3 mmol) and dropwise
addition of freshly distilled POCl (280l, 3 mmol). During the reaction white precipitate was formed.
3
3
The reaction mixture was stirred below 0°C for 4 h after which RP HPLC analysis indicated full
consumption of the starting material. The reaction mixture was poured to cold, deionised water
(approx. 100 ml) and neutralised with 10% NaHCO . The resulting mixture was loaded on DEAE
3
3-
Sephadex A-25 column (HCO form, 100 g), the column was washed with thoroughly with water and
then eluted using TEAB in deionized water (3600 ml, 0 to 0.7 M linear gradient). The fractions
containing pure product (UV and RP HPLC analysis) were combined, concentrated in vacuo, co-
evaporated with 96% EtOH (approx. 50 ml), co-evaporated with MeCN (approx. 50 ml). The residue
was dissolved in MQ water, loaded on RP C18 column (20g) and eluted with 20% MeCN in 0.05 M
ammonium acetate buffer (pH 5.9). The fractions containing the desired product were combined,
concentrated in vacuo co-evaporated with 96% EtOH (approx. 50ml), co-evaporated with MeCN. The
residue was dissolved in MQ water and freeze-dried. After repeated freeze-drying, product was
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 39
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
isolated as ammonium salt (419 mg, 90%). The analytical data matched those obtained for 6
synthesized by direct trifluoromethylation of guanosine monophosphate.
2
',3',5'-tri-O-acetyl-8-trifluoromethyl-adenosine (12). 2',3',5'-tri-O-acetyl-adenosine (11, 393 mg, 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
mmol) was dissolved in DMSO (6 ml) followed by addition of (CF SO Zn dihydrate (1100 mg, 3 mmol)
3
)
2 2
upon vigorous stirring at room temperature. When the clear solution was formed (15-20 min), t-
BuOOH (70% solution in water, 0.65 ml, 5 mmol) in 10 aliquots (65 l each) in 20 min intervals. During
addition of t-BuOOH the reaction mixture started to became yellow. The reaction mixture was stirred
for 24 h after which TLC analysis (3% MeOH in DCM) indicated no further progress of the reaction. The
reaction mixture was poured into 100 ml of water and extracted with DCM (3 x 30 ml). The combined
organic layers were washed with water (3 x approx. 100 ml), brine (approx. 100 ml) and dried over
Na
2
SO . The drying agent was filtered off, washed with DCM and the filtrate was concentrated in vacuo.
4
To the oily residue approx. 20 ml of DCM and 2 g of silica was added. The slurry was concentrated in
vacuo and the dry residue was loaded on the preconditioned silica gel column (24 g). The product was
eluted with the mixture of MeOH in DCM (0 to 5% linear gradient). The fractions containing the product
were combined, concentrated in vacuo, co-evaporated with diethyl ether (approx. 10 ml) and dried
1
overnight under high vacuum giving 12 as off-white solid (171 mg, 37%). H NMR (400 MHz, CDCl
3
) 
8
.42 (s, 1H), 6.41 (dd, J = 4.7 Hz, 6.1 Hz, 1H), 6.09 (d, J = 4.7 Hz, 1H), 5.90 (m, 3H), 4.55 (m, 1H), 4.44-
4.36 (m, 2H), 2.16 (s, 3H), 2.09 (s, 3H), 2.06 (s, 3H); 19F NMR (376 MHz, CDCl
3
)  - 61.39; HRMS (+)
+
+
ESI m/z: [M+H] calcd for C17
H
19
3
F N
5
O
7
462.1231; found 462.1225.
6
6
Trifluoromethylation of N ,2',3',5'-tri-O-tetraacetyladenosine. N ,2',3',5'-tri-O-tetraacetyladenosine
(
14, 870 mg, 2 mmol) was dissolved in DMSO (12 ml) followed by addition of (CF
3
SO ) Zn dihydrate
2
2
(2200 mg, 6 mmol) upon vigorous stirring at room temperature. When the clear solution was formed
(15-20 min), t-BuOOH (70% solution in water, 1.3 ml, 10 mmol) in 10 aliquots (135 l each) in 20 min.
intervals. During addition of t-BuOOH the reaction mixture started to became yellow. The reaction
mixture was stirred for 24 h after which TLC analysis (3% MeOH in DCM) indicated no further progress
ACS Paragon Plus Environment

Page 21 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
of the reaction. The reaction mixture was poured into 100 ml of water and extracted with DCM (3 x 30
ml). The combined organic layers were washed with water (3 x approx. 100 ml), brine (approx. 100 ml)
and dried over Na
2
SO . The drying agent was filtered off, washed with DCM and the filtrate was
4
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
concentrated in vacuo. To the oily residue approx. 20 ml of DCM and 2 g of silica was added. The slurry
was concentrated in vacuo and the dry residue was loaded on the preconditioned silica gel column (40
g). The column was eluted with the mixture of MeOH in DCM by stepwise changing of the MeOH
concentration (0 to 3%). The products and remaining starting material were washed from the column
in the following order: 17, 15, 16, 14 (see figure S4). The fractions containing each product were
combined, concentrated in vacuo, co-evaporated with diethyl ether (approx. 10 ml) and dried
overnight under high vacuum.
6
1
N ,2',3',5'-tri-O-tetraacetyl-8-trifluoromethyl-adenosine (15). White solid (91 mg, 9%). H NMR (400
MHz, CDCl )  8.78 (s, 1H), 8.73 (bs, 1H), 6.40 (dd, J = 4.9 Hz, 6.1 Hz, 1H), 6.11 (d, J = 4.9 Hz, 1H), 5.86
3
(dd, J = 5.2 Hz, 6.1 Hz, 1H), 4.45 (dd, J = 3.5 Hz, 11.4 Hz, 1H), 4.47-4.36 (m 2H), 2.65 (s, 3H), 2.18 (s, 3H),
2.08 (s, 3H), 2.06 (s, 3H); 13C {H} NMR (100 MHz, CDCl
3
)  = 170.6, 170.4, 169.6, 169.4, 154.6, 151.7,
1
2
C
51.0, 140.7 (q, J = 40.7 Hz) 120.2, 118.2 (q, J = 272.4 Hz), 88.1 (q, J = 2.4 Hz), 80.9, 72.0, 70.5, 62.86,
6.0, 20.8, 20.6, 20.4; 19F NMR (376 MHz, CDCl
)  -61.63; HRMS (+) ESI m/z: [M+H] calcd for
+
3
+
19
H
21
F
3
N
5
O
8
504.1337; found 504.1336.
6
1
N ,2',3',5'-tri-O-tetraacetyl-2-trifluoromethyl-adenosine (16). White solid (101 mg, 10%). H NMR (400
3
MHz, CDCl )  9.50 (bs, 1H), 8.48 (s, 1H), 6.26 (d, J = 5.3 Hz, 1H), 5.87 (dd, J = 5.5 Hz, 5.9 Hz, 1H), 5.68
(dd, J = 4.5 Hz, 5.6 Hz, 1H), 4.53-4.47 (m, 1H), 4.46-4.38 (m, 2H), 2.76 (s, 3H), 2.19 (s, 3H), 2.11 (s, 3H),
2
.09 (s, 3H); 13C {H} NMR (100 MHz, CDCl
) 171.9, 170.5, 169.7, 169.6, 150.9, 150.01 (q, J = 37.4
3
Hz), 149.9, 144.0, 122.7, 119.7 (q, J = 275.1 Hz), 87.1, 80.9, 73.6, 70.8, 63.2, 26.4, 20.8, 20.7, 20.4; 19
F
+
+
NMR (376 MHz, CDCl
04.1333.
3
)  -69.37; HRMS (+) ESI m/z: [M+H] calcd for C19
21
H F
N
3 5
O
8
504.1337; found
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 39
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
6
1
N ,2',3',5'-tri-O-tetraacetyl-2,8-ditrifluoromethyl-adenosine (17). Yellowish solid (80 mg, 7%). H NMR
400 MHz, CDCl )  8.55 (bs, 1H), 6.21 (dd, J = 4.6 Hz, 6.2 Hz, 1H), 6.11 (d, J = 4.6 Hz, 1H), 5.83 (dd, J
5.8 Hz, 6.2 Hz, 1H), 4.55 (dd, J = 4.0 Hz, 11.8 Hz, 1H), 4.48-4.42 (m, 1H), 4.41-4.35 (m, 1H), 2.77 (s,
(
3
=
19
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
3H), 2.18 (s, 3H), 2.08 (s, 3H), 2.04 (s, 3H); F NMR (376 MHz, CDCl )  - 61.79, -69. 71; HRMS (+) ESI
3
+
+
m/z: [M+H] calcd for C20
20 6
H F N
5
O 572.1211; found 572.1200.
8
1
4 was recovered in 33% yield.
General procedure for deprotection of 12, 15, 16 and 17, the synthesis of 13, 18 and 19. Protected
nucleoside was placed in round bottom flask equipped with rubber septum and flushed with stream
of argon. Then MeNH (33% in EtOH, 100 l per each 10 mg of the starting material) was added under
2
gentle flow of argon ant the resulting mixture was stirred for 4 h at room temperature after which TLC
analysis (5% MeOH in DCM) indicated full consumption of the starting material. The reaction mixture
was concentrated in vacuo. To the residue a small amount of MeOH and silica was added. The slurry
was concentrated in vacuo and the dry residue was loaded on the preconditioned silica gel column.
The product was eluted with the mixture of MeOH in DCM (5%, v/v). The fractions containing the
desired product were combined, concentrated in vacuo, co-evaporated with diethyl ether (approx. 10
ml) and dried overnight under high vacuum.
1
8
-trifluoromethyl-adenosine (13). White solid (64 mg, starting from 92 mg (0.20 mmol) of 12, 95%),. H
NMR (400 MHz, DMSO-d
6
) 8.26 (s, 1H, C2), 8.12-7.79 (m, 2H, -NH
2
), 5.79 (d, J = 6.8 Hz, 1H, C1’),
5
.54 (dd, J = 3.8 Hz, 8.8 Hz, 1H, C5’-OH), 5.50 (d, J = 6.2 Hz, 1H, C2’-OH), 5.27 (d, J = 4.4 Hz, 1H, C3’-OH),
13
5.11 (m, 1H, C2’), 4.22 (m, 2H, C3’), 4.02 (m, 1H, C4’), 3.75-3.67 (m, 1H, C5’), 3.60-3.51 (m, 1H, C5’); C
{
H} NMR (100 MHz, DMSO-d
6
) 157.33 (C6), 154.6 (C2), 150.0 (C4), 136.8 (q, J = 39.1 Hz, C8), 118.7
(
q, J = 271.4 Hz, CF
3
), 118.0 (C5), 89.7 (C1’), 87.2 (C4’), 71.5 (C2’), 70.9 (C3’), 62.0 (C5’); 19F NMR (376
+
+
MHz, DMSO-d
6
)  -59.95; HRMS (+) ESI m/z: [M+H] calcd for C11H1
F N
3 3 5
O
4
336.0914; found
336.0911.
ACS Paragon Plus Environment

Page 23 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
2
-trifluoromethyl-adenosine (18). White solid (32 mg, starting from 50 mg (0.10 mmol) of 16, 95%). H
NMR (400 MHz, DMSO-d ) 8.57 (s, 1H, C8), 7.97 (bs, 1H, -NH ), 5.91 (d, J = 6.1 Hz, 1H, C1’), 5.49 (d,
J = 6.2 Hz, 1H, C2’-OH), 4.92 (d, J = 4.9 Hz, 1H, C3’-OH), 4.99 (dd, J = 5.5 Hz, 5.7 Hz, 1H, C5’-OH), 4.60
6
2
13
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
(m, 1H, C2’), 4.15 (m, 1H, C3’), 3.95 (m, 1H, C4’), 3.71-3.62 (m, 1H, C5’), 3.59-3.51 (m, 1H, C5’); C {H}
NMR (100 MHz, DMSO-d
6
)= 156.3 (C6), 149.2 (q, J = 34.7 Hz, C2), 148.9 (C4), 141.7 (C8), 120.0 (q, J =
2
75.0 Hz, CF
DMSO-d
336.0910.
3
), 119.85 (C5), 87.3 (C1’), 85.9 (C4’), 73.6 (C2’), 70.5 (C3’), 61.4 (C5’); 19F NMR (376 MHz,
+
+
6
)  = -68.03 ppm. HRMS (+) ESI m/z: [M+H] calcd for C11H1
3 3
F N
5
O
4
336.0914; found
2,8-ditrifluoromethyl-adenosine (19) Off-white solid (38 mg, starting from 57 mg (0.10 mmol) of 17,
1
9
=
5
5%). H NMR (400 MHz, DMSO-d
6
) = 8.64-8.38 (m, 2H, -NH ), 5.82 (d, J = 6.2 Hz, 1H, C1’), 5.53 (d, J
2
5.9 Hz, 1H, C2’-OH), 5.32 (d, J = 4.8 Hz, 1H, C3’-OH), 5.09 (m, 1H, C2’), 4.77 (dd, J = 5.6 Hz, 6.6 Hz, 1H,
13
’-OH), 4.25 (m, 1H, C3’), 3.96 (m, 1H, C4’), 3.77-3.67 (m, 1H, C5’), 3.59-3.49 (m, 1H, C5’); C {H} NMR
)  = 157.5 (C6), 151.1 (q, J = 35.7 Hz, C2), 149.8 (C4), 138.3 (q, J = 39.3 Hz, C8),
19.7 (q, J = 275.7 Hz, C2-CF ), 118.7 (C5), 118.3 (q, J = 271.0 Hz, C8-CF ), 89.9 (C1’), 86.7 (C4’), 71.4
)  = -60.23, -68.66; HRMS (+) ESI m/z: [M+H]+
(100 MHz, DMSO-d
6
1
3
3
19
(C2’), 70.6 (C3’), 61.6 (C5’); F NMR (376 MHz, DMSO-d
6
+
calcd for C12
H F
12 6
5
N O
4
404.0788; found 404.0784.
General procedure for phosphorylation of 13, 18 and 19, the synthesis of 20, 21 and 22. Nucleoside
(1 equiv.) was placed in round bottom flask equipped with rubber septum and flushed with stream of
argon. (MeO)
below 0°C (ice/brine bath) followed by addition of 2,6 lutidine (3 equiv.) and dropwise addition of
freshly distilled POCl (3 equiv.). The reaction mixture was stirred below 0 °C until RP HPLC analysis
indicated full consumption of the starting material (usually 3-4 h). The reaction mixture was poured to
3
PO (0.1 M) was added under gentle flow of argon and the resulting solution was cooled
3
cold, deionised water (10 times the volume of the solvent used) and neutralised with 10% NaHCO .
3
3-
The resulting mixture was loaded on DEAE Sephadex A-25 column (HCO form, 10 g), the column was
washed with thoroughly with water and then eluted using TEAB in deionized water (400 ml, 0 to 0.7 M
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 39
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
linear gradient). The fractions containing pure product (UV and RP HPLC analysis) were combined,
concentrated in vacuo, co-evaporated with 96% EtOH (approx. 50 ml), co-evaporated with MeCN
(approx. 50 ml). The residue was dissolved in MQ water and freeze-dried. The products were purified
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
using preparative RP HPLC chromatography with linear gradient of MeCN in 0.05 M ammonium acetate
buffer (pH = 5.9). The fractions containing pure product were concentrated in vacuo, co-evaporated
with 96% EtOH and co-evaporated with MeCN (approx. 50 ml). The residue was dissolved in MQ water
and freeze-dried. After repeated freeze-drying, products were obtained as ammonium salt.
8
-trifluoromethyl-adenosine 5’-monophosphate (20). White foam (58.6 mg, starting from 50.3 mg (0.15
1
mmol) of 13, 87%),. H NMR (400 MHz, D
2
O) 8.25 (s, 1H), 6.02 (d, J = 5.8 Hz, 1H), 5.28 (dd, J = 5.9
13
Hz, 6.1 Hz, 1H), 4.59 (dd, J = 4.7 Hz, 6.1 Hz, 1H), 4.26 (m, 1H), 4.16-4.01 (m, 2H); C {H} NMR (125 MHz,
O) 158.6, 157.2, 152.6, 140.8 (q, J = 40.3 Hz), 120.3 (q, J = 271.1 Hz), 120.0, 91.5, 86.4, 73.5, 72.1,
D
2
6
6.4; 19F NMR (376 MHz, D
2
O)= -61.32; 31P NMR (162 MHz, D O) = 0.95; HRMS (-) ESI m/z: [M-H]-
2
calcd for C11H F
12 3
5
N O
7
P- 414.0432; found 414.0432.
2
-trifluoromethyl-adenosine 5’-monophosphate (21). White foam (23.6 mg, starting from 23.5 mg (0.07
1
mmol) of 18, 75%). H NMR (400 MHz, D
2
O)  = 8.54 (s, 1H), 6.16 (d, J = 5.3 Hz, 1H), 4,73 (m, 1H), 4.51
13
(
m, 1H), 4.38 (m, 1H), 4.14 (m, 2H); C {H} NMR (125 MHz, D
2
O) 158.4, 152.9 (q, J = 36.0 Hz) , 151.5,
1
44.1, 122.2 (q, J = 274.6 Hz), 122.0, 90.1, 86.5, 77.2, 72.9, 67.0; 19F NMR (376 MHz, D
2
O)= -68.98;
31P NMR (162 MHz, D
14.0432.
O) = 1.15; HRMS (-) ESI m/z: [M-H] calcd for C11
-
P- 414.0432; found
2
12
H F
3
N
5
O
7
4
2,8-ditrifluoromethyl-adenosine 5’-monophosphate (22) White foam (7.2 mg, starting from 8.1 mg
1
(
0.02 mmol) of 19, 70%). H NMR (400 MHz, D
2
O)  = 6.12 (d, J = 4.9 Hz, 1H), 5.27 (dd, J = 5.3 Hz, 5.9
O)
159.7, 154.5 (q, J = 36.4 Hz), 152.4, 142.7 (q, J = 40.8 Hz), 123.1 (q, J = 274.8 Hz), 121.3, 120.5 (q, J
Hz, 1H), 4.73 (dd, J = 5.1 Hz, 6.0 Hz, 1H), 4.33 (m, 1H), 4.27-4.09 (m, 2H); 13C {H} NMR (125 MHz, D
2
19
31
=
271.2 Hz),92.5, 86.9, 74.8, 72.9, 67.5; F NMR (376 MHz, D
2
O)= - 61.53, -69.61; P NMR (162 MHz,
-
P- 482.0306; found 482.0305.
D
2
O) = 1.44; HRMS (-) ESI m/z: [M-H] calcd for C12
H
11
6 5
F N O
7
ACS Paragon Plus Environment

Page 25 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
General procedure for the synthesis of 6-Im and 20-Im. Monophosphate 6 or 20 (1 equiv.) was
dissolved in anhydrous DMF (0.05 M) followed by addition of imidazole (20 equiv.), 2,2’-
dithiodipyridine (5 equiv.) and trimethylamine (5 equiv.). The resulting solution was stirred at room
temperature for 15 min and then triphenylphosphine (5 equiv.) was added in one portion. The solution
turned immediately deep yellow. The reaction mixture was stirred overnight at room temperature.
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
Addition of a solution of anhydrous LiClO (5 equiv.) in anhydrous acetone (10 times volumes of DMF
4
used) resulted in precipitation of the product as lithium salt. The suspension was cooled at 4 °C for
approx. 2 h and the precipitate was centrifuged, washed repeatedly with cold acetone until the
supernatant was colourless. The resulting solid was additionally washed with cold diethyl ether and
dried overnight under high vacuum. Yields + 95%. CAUTION: A substantial level of decomposition of 6-
Im and 20-Im was observed while storage, even at - 18°C. To assure good yields of further reactions
they should be used immediately after preparation.
General procedure for the synthesis of 7, 8, 23 and 24. Freshly prepared P-imidazolide of the
respective nucleotide (1 equiv.) was dissolved in anhydrous DMF (0.05 M). Triethyl ammonium
phosphate or triethylammonium pyrophosphate (8 equiv.) was added followed by addition of
anhydrous ZnCl (8 equiv.) The reaction mixture was stirred at room temperature until HPLC analysis
2
showed full consumption of the starting material (usually 24-30 h). The reaction mixture was diluted
with EDTA solution (8.1 equiv. in water, 10 times volume of DMF used) and neutralised with 10%
3-
NaHCO . The resulting mixture was loaded on DEAE Sephadex A-25 column (HCO form, 10 g), the
3
column was washed with water (approx. 50 ml) and then eluted using TEAB in deionized water (400
ml, linear gradient). The fractions containing the mixture of the desired product and starting material
(UV and RP HPLC analysis) were combined, concentrated in vacuo, co-evaporated with 96% EtOH
(approx. 50 ml), co-evaporated with MeCN (approx. 50 ml) and the residue was dried overnight under
high vacuum. The product was purified using RP HPLC chromatography with linear gradient of MeCN
in 0.05 M ammonium acetate buffer (pH = 5.9). The fractions containing pure product were
concentrated in vacuo, co-evaporated with 96% EtOH (approx. 50 ml) and co-evaporated with MeCN
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 39
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
(
approx. 50 ml). The residue was dissolved in MQ water and freeze-dried. After repeated freeze-drying,
product was obtained as ammonium salt.
was isolated in 80% yield (9.4 mg starting from 10.0 mg (0.021 mmol) of 6-Im). 8 was isolated in 40%
7
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
yield (5.5 mg starting from 10.0 mg (0.021 mmol) of 6-Im). For 7 and 8 the analytical data matched
those obtained from the 7 and 8 synthesized by direct trifluoromethylation of guanosine diphosphate
and triphosphate respectively.
8-trifluoromethyl-adenosine 5’-diphosphate (23). White foam (9.2 mg starting from 10.0 mg (0.021
1
mmol) of 20-Im, 80%). H NMR (400 MHz, D
2
O)  = 8.36 (s, 1H), 6.05 (d, J = 5.6 Hz, 1H), 5.28 (dd, J = 5.7
19
Hz, 6.2 Hz, 1H), 4.65 (dd, J = 4.3 Hz, 5.8 Hz, 1H), 4.35-4.18 (m, 3H); F NMR (376 MHz, D
2
O)= - 61.43;
31P NMR (162 MHz, D
2
O) = - 9.91 (d, J = 20.1 Hz, 1p), -10.34 (d, J = 20.3 Hz, 1P); HRMS (-) ESI m/z: [M-
-
10P -
494.0095; found 494.0095.
H] calcd for C11
13
H F
3 5
N O
2
8-trifluoromethyl-adenosine 5’-triphosphate (24). White foam (7.7 mg starting from 10.0 mg (0.021
1
mmol) of 20-Im, 60%). H NMR (400 MHz, D
2
O)  = 8.37 (s, 1H), 6.08 (d, J = 6.0 Hz, 1H), 5.39 (dd, J = 6.0
O)= - 61.31; 31P NMR (162
O) = -9.37- -9.99 (m, 1P), -10.36 (d, J = 19.8 Hz, 1P), -22.2 (t, J = 19.8 Hz, 1P); HRMS (-) ESI
Hz, 6.7 Hz, 1H), 4.68 (m, 1H), 4.40-4.19 (m, 3H); 19F NMR (376 MHz, D
MHz, D
m/z: [M-H] calcd for C11
2
2
-
13P -
573.9759; found: 573.9761.
H
14
3 5
F N O
3
The synthesis of dinucleotide 25. Freshly prepared 20-Im (10.0 mg, 0.021 mmol) was dissolved in
anhydrous DMSO (0.5 ml) followed by addition of triethyl ammonium phosphate (6.0 mg, 0.015 mmol).
To this solution anhydrous ZnCl was added (32.0 mg, 0.235 mmol). After approx. 2h HPLC analysis
2
indicated formation of substantial amounts of 23 and the presence of 20 (formed by the hydrolysis of
the starting material). Additional portion of 20-Im (10.0 mg, 0.021 mmol) was added and the mixture
was stirred overnight at room temperature. The reaction was stopped by addition of EDTA solution
(80 mg, 0.238 mmol in 5 ml of water) and neutralised with 10% NaHCO . The resulting mixture was
3
3-
loaded on DEAE Sephadex A-25 column (HCO form, 10 g), the column was washed with water (approx.
50 ml) and then eluted using TEAB in deionized water (400 ml, 0 to 1 M linear gradient). The fractions
ACS Paragon Plus Environment

Page 27 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
containing the mixture of the desired product and traces of by-products (UV and RP HPLC analysis)
were combined, concentrated in vacuo, co-evaporated with 96% EtOH (approx. 50 ml), co-evaporated
with MeCN (approx. 50 ml) and the residue was dried overnight under high vacuum. The product was
purified using RP HPLC chromatography with linear gradient of MeCN in 0.05 M ammonium acetate
buffer (pH = 5.9). The fractions containing pure product were concentrated in vacuo, co-evaporated
with 96% EtOH (approx. 50 ml) and co-evaporated with MeCN (approx. 50 ml). The residue was
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
dissolved in MQ water and freeze-dried. After repeated freeze-drying, product was obtained as
1
ammonium salt (white foam, 8.9 mg, 63%). H NMR (400 MHz, D
2
O)  = 8.05 (s, 2H), 5.77 (d, J = 6.0 Hz,
2
4
H), 5.06 (dd, J = 5.8 Hz, 6.1 Hz, 2H), 4.47 (dd, J = 3.6 Hz, 5.7 Hz, 2H), 4.25-4.15 (m, 2H), 4.14-4.01 (m,
19
31
H); F NMR (376 MHz, D
2
O) = -61.28; P NMR (162 MHz, D
2
O) = -10.46 (d, J = 17.7 Hz, 2P), -22.01
-
-
(
t, J = 17.8 Hz, 1P); HRMS (-) ESI m/z: [M-H] calcd for C22
H
24
6
F N
10
O
16
P
3
891.0494; found 891.0502.
The synthesis of dinucleotide 26. ADP-Im (sodium salt, 27.0 mg, 0.051 mmol) and 6 (ammonium salt,
7.0 mg, 0.036 mmol) were dissolved in anhydrous DMSO (2 ml) followed by addition of anhydrous
ZnCl (86.0 mg, 0.632 mmol). The reaction mixture was stirred at room temperature for 72h at which
time HPLC analysis no further progress of the reaction. The reaction mixture was diluted with EDTA
solution (235 mg, 0.642 mmol in 20 ml of water) and neutralised with 10% NaHCO . The resulting
1
2
3
3-
mixture was loaded on DEAE Sephadex A-25 column (HCO form, 10 g), the column was washed with
water (approx. 50 ml) and then eluted using TEAB in deionized water (400 ml, 0 to 1 M linear gradient).
The fractions containing the mixture of the desired product and traces of starting material (UV and RP
HPLC analysis) were combined, concentrated in vacuo, co-evaporated with 96% EtOH (approx. 50 ml),
co-evaporated with MeCN (approx. 50 ml) and the residue was dried overnight under high vacuum.
The product was purified using RP HPLC chromatography with linear gradient of MeCN in 0.05 M
ammonium acetate buffer (pH = 5.9). The fractions containing pure product were concentrated in
vacuo, co-evaporated with 96% EtOH (approx. 50 ml) and co-evaporated with MeCN (approx. 50 ml).
The residue was dissolved in MQ water and freeze-dried. After repeated freeze-drying, product was
1
obtained as ammonium salt (white foam, 12.8 mg, 40%). H NMR (400 MHz, D
2
O)  = 8.46 (s, 1H), 8.27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 39
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
(
s, 1H), 6.00 (d, J = 5.5 Hz, 1H), 5.77 (d, J = 6.4 Hz, 1H), 5.30 (dd, J = 5.8 Hz, 6.2 Hz, 1H), 4.63 (dd, J = 5.4
H, 5.7 Hz, 1H), 4.59 (dd, J = 2.6 Hz, 5.4Hz, 1H), 4.50 (m, 1H), 4.42-4.16 (m, 6H); 19F NMR (376 MHz,
D
2
O)= - 61.14; 31P NMR (162 MHz, D
2
O) = - 10.46 (t, J = 17.0 Hz, 2P), -22.16 (t, J = 18.9 Hz, 1P);
-
-
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
HRMS (-) ESI m/z: [M-H] calcd for C21
25
H F
3 10
N O
17
P
3
839.0570; found 839.0571.
The synthesis of dinucleotide 27. GDP-Im (sodium salt, 34.0 mg, 0.051 mmol) and 6 (ammonium salt,
17.0 mg, 0.036 mmol) were dissolved in anhydrous DMSO (2 ml) followed by addition of anhydrous
ZnCl
time HPLC analysis no further progress of the reaction. The reaction mixture was diluted with EDTA
solution (235 mg, 0.642 mmol in 20 ml of water) and neutralised with 10% NaHCO . The resulting
2
(86.0 mg, 0.632 mmol). The reaction mixture was stirred at room temperature for 48h at which
3
3-
mixture was loaded on DEAE Sephadex A-25 column (HCO form, 10 g), the column was washed with
water (approx. 50 ml) and then eluted using TEAB in deionized water (400 ml, 0 to 1 M linear gradient).
The fractions containing the mixture of the desired product and traces of starting material (UV and RP
HPLC analysis) were combined, concentrated in vacuo, co-evaporated with 96% EtOH (approx. 50 ml),
co-evaporated with MeCN (approx. 50 ml) and the residue was dried overnight under high vacuum.
The product was purified using RP HPLC chromatography with linear gradient of MeCN in 0.05 M
ammonium acetate buffer (pH = 5.9). The fractions containing pure product were concentrated in
vacuo, co-evaporated with 96% EtOH (approx. 50 ml) and co-evaporated with MeCN (approx. 50 ml).
The residue was dissolved in MQ water and freeze-dried. After repeated freeze-drying, product was
1
obtained as ammonium salt (white foam, 27.1 mg, 83%). H NMR (400 MHz, D
2
O)  = 8.11 (s, 1H), 5.82
(d, J = 6.5 Hz, 1H), 5.79 (d, J = 5.8 Hz, 1H), 5.36 (dd, J = 5.8 Hz, 6.6 Hz, 1H), 4.66 (dd, J = 5.5 Hz, 6.0 Hz,
1
H), 4.59 (m, 1H), 4.49 (m, 1H), 4.39-4.17 (m, 6H); 19F NMR (376 MHz, D O)= - 61.06; 31P NMR (162
2
MHz, D
2
O)= -10.41 (t, J = 6.8 Hz, 1P), -10.50 (t, J = 6.5 Hz, 1P), -23.14 (t, J = 19.0 Hz, 1P); HRMS (-) ESI
-
m/z: [M-H] calcd for C21
H
25
3 10
F N O
18
P 855.0519; found 855.0524.
3
The synthesis of dinucleotide 28. 27 (9.0 mg, 0.01 mmol) was dissolved in anhydrous DMSO (0.5 ml)
followed by addition of MeI (40 l, excess). The reaction was stirred at room temperature until HPLC
ACS Paragon Plus Environment

Page 29 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
analysis indicated full consumption of the starting material (approx. 6 h). The reaction mixture was
diluted with water (5 ml) and washed with diethyl ether (3 x 5 ml). The aqueous phase was freeze-
dried and the residue was purified using RP HPLC chromatography with linear gradient of MeCN in 0.05
M ammonium acetate buffer (pH = 5.9). The fractions containing pure product were concentrated in
vacuo, co-evaporated with 96% EtOH (approx. 50 ml) and co-evaporated with MeCN (approx. 50 ml).
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
The residue was dissolved in MQ water and freeze-dried. After repeated freeze-drying, product was
1
obtained as ammonium salt (white foam, 7.2 mg, 80%). H NMR (400 MHz, D
2
O) =5.79-5.71 (m, 2H),
5.16 (m, 1H), 4.55 (m, 1H), 4.47 (m, 1H), 4.44-4.32 (m, 4H), 4.31-4.16 (m, 3H), 4.06 (bs, 3H); 19F NMR
(376 MHz, D
2
O)= -61.17; 31P NMR (162 MHz, D
2
O) = -10.36 - -10.69 (m, 2P), -22.15 (t, J = 19.4 Hz,
-
-
1
P); HRMS (-) ESI m/z: [M-H] calcd for C22
H
27
3
F N
10
O
18
P
3
869.0675; found: 869.0681.
Stock solutions of studied (di)nucleotides. Compounds 6, 20, 23, 25, 26, 27 and 28 were dissolved in
pure water (200-300 µL), 13 was dissolved in DMSO (200 µL) and concentrations were estimated
spectrophotometrically by measurement of absorbance at 260 nm in 0.1 M phosphate buffer pH 6.0
(6, 28) or 7.0 (23, 25, 26, 27). To calculate the exact concentrations the following molar extinction
-1
-1
coefficients [M cm ] were employed: ε = 11400 (6, 13, 20), ε = 15020 (23), ε = 27036 (25, 26), ε =
2600 (27), ε = 21132 (28).
Samples preparation for enzymatic studies. For studies with human Fhit compounds 26 or 27 were
diluted in buffer containing 50 mM MES KOH pH 6.50, 1 mM MgCl and 10% D O to a final
concentration of 100 µM. For studies with human DcpS compound 28 was diluted in buffer containing
0 mM Tris HCl pH 7.60, 0.2 M KCl, 0.5 mM EDTA and 10% D O to final concentration of 100 µM. For
studies with human cNIIIB compound 6 was diluted in buffer containing 20 mM HEPES KOH pH 7.50,
0 mM KCl, 5 mM MgCl , 10% D O to final concentration of 100 µM. RG3039 was purchased from
2
2
2
5
2
5
2
2
KareBay Biochem. Human Fhit inhibitor (IN-A, 7,8-Dihydro-7,7-dimethyl-10-(4-chlorophenyl)-5H-
indeno[1,2-b]quinoline-9,11(6H,10H)-dione) was synthesized as previously described.60 All inhibitors
were dissolved in DMSO and the concentrations were established by mass (RG3039, IN-A) or
ACS Paragon Plus Environment