Synthesis of c-di-GMP analogs with thiourea, urea, carbodiimide, and guanidinium linkages

Article abstract of DOI:10.1021/ol403154w

The first syntheses of neutral thiourea, urea, and carbodiimide analogs, along with two guanidinium analogs, of the bacterial signaling molecule cyclic diguanosine monophosphate (c-di-GMP) are reported. The key intermediate, obtained in nine steps, is a 3

Full text of DOI:10.1021/ol403154w

Letter
pubs.acs.org/OrgLett
Synthesis of c‑di-GMP Analogs with Thiourea, Urea, Carbodiimide,
and Guanidinium Linkages
Barbara L. Gaffney and Roger A. Jones*
Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, 610 Taylor Road, Piscataway, New
Jersey 08854, United States
S
* Supporting Information
ABSTRACT: The first syntheses of neutral thiourea, urea, and carbodiimide
analogs, along with two guanidinium analogs, of the bacterial signaling
molecule cyclic diguanosine monophosphate (c-di-GMP) are reported. The
key intermediate, obtained in nine steps, is a 3′-amino-5′-azido-3′,5′-dideoxy
derivative. The 5′-azide serves as a masked amine from which the amine is
obtained by Staudinger reduction, while the 3′-amine is converted to an
isothiocyanate that, while stable to chromatography, and Staudinger
conditions, nevertheless reacts well with the 5′-amine.
Scheme 1. Synthesis of Key Intermediate 4
he bacterial signaling molecule cyclic diguanosine mono-
Tphosphate (c-di-GMP) is responsible for regulating
bacterial responses to a variety of environmental factors,
including aggregation into the biofilm state.1−6 Binding of c-di-
GMP as a monomer and as a self-intercalated dimer to the PilZ
domain proteins has been demonstrated.1,2,7,8
Activation of two different classes of riboswitches in
noncoding regulatory mRNA domains also has been identified
upon binding c-di-GMP.9−12 Finally, c-di-GMP, among other
cyclic dinucleotides, plays a role in triggering an innate immune
response13,14 through a transmembrane protein named STING
in the innate immune sensing pathway, where a specific
receptor for cyclic dinucleotides has been identified.15
A number of synthetic routes to c-di-GMP and its
thiophosphate analogs have been reported.16−19 Two analogs
with a nonphosphate backbone have been prepared, one a
methylphosphonate,20 the other a carbamate,21 but each lacks a
2′-hydroxyl group. An analog with a 2′-fluoro in place of the 2′-
hydroxyl, with a phosphate backbone, was reported most
recently.22 The goal of the work reported below was to prepare
c-di-GMP analogs with urea or urea related backbone linkages
that should be stable to the bacterial phosphodiesterases that
regulate c-di-GMP. The syntheses start with the introduction of
nitrogen atoms to the guanosine 3′ and 5′ positions. The first
steps are to prepare the 5′-azido-5′-deoxy derivative 3, as
shown in Scheme 1.
azide and that 3 was readily isolated simply by the addition of
methanol to the reaction mixture. The N2-dmf group was used
in this synthesis, as it has been shown to be essential for the
reaction of guanosine with α-acetoxyisobutyryl bromide.25 The
reaction of 3 with this reagent proceeded analogously to that
reported for 1, with no degradation of the azido group under
the acidic reaction conditions. In addition to the desired
product, 4, a small amount of the 2′-Br isomer was produced, in
the ratio of 92:8, by LC-MS. These isomers were not separable
by silica chromatography, but 4 was readily crystallized from
methylene chloride, which efficiently removed the 2′-Br isomer.
No chromatography was required for the preparation of
compounds 1−4, so that these reactions were conveniently
The N2-dimethylformamidine (dmf) derivative of guanosine,
1, was prepared by standard methods as described in detail in
the Supporting Information. Preparation of 2 and 3 followed
procedures reported for guanosine by Martin23 and by Dean,24
respectively. The major differences in this case were that
heating was not required for the reaction of 2 with sodium
Received: November 1, 2013
Published: December 6, 2013
© 2013 American Chemical Society
158
dx.doi.org/10.1021/ol403154w | Org. Lett. 2014, 16, 158−161

Organic Letters
Letter
carried out starting with 20 g of guanosine to give 4 in an
overall yield of 38%.
reaction with tert-butyldimethylsilyl chloride, as shown in
Scheme 3. Addition of the TBS group makes 10 again
The conversion of 4 to the 3′-amino-5′-azido derivative 9,
shown in Scheme 2, proceeded analogously to the preparation
Scheme 3. Synthesis of the Linear Dimer 14
Scheme 2. Synthesis of 3′-Amino-5′-azido Derivative 9
of 3′-amino-3′-deoxyguanosine reported by Zhang, although by
using extensively altered conditions, and some different
reagents, the reaction times were significantly reduced.26
Catalytic DMAP in methanol with a few equivalents of TEA
effected clean removal of the acetyl group from 4. The reaction
of 5 with benzylisocyanate in acetonitrile then proceeded in 2 h
to give 6. After investigating numerous reagents for cyclization
to 7, tBuONa in THF was found to give complete conversion
in 45 min. Saponification of 7 to 8 by addition of 10 N NaOH
to a methanol solution of 7 proceeded in 2 h. It is somewhat
surprising that the N2-dmf group survived these strongly basic
conditions with only minimal loss. After neutralization of the
reaction mixture, 8 was isolated by extraction. The steps from 4
to 8 were carried out in one flask, without isolation of
intermediates, and 8 did not need purification before
conversion to 9.
Because of the 5′-azide it was not possible to use reduction to
debenzylate the 3′-amino group in 8, and instead oxidation
using diisopropylazodicarboxylate (DIAD) was employed.27
This is a slow reaction that required overnight to give the
corresponding imine (not shown). Hydrolysis to 9 was effected
using 1 N HCl, within 10 min, again with minimal loss of the
N2-dmf group. After neutralization of the reaction mixture with
NaHCO3, 9 was isolated by extraction, in this case remaining in
the aqueous phase, while excess reagent was removed in the
organic phase. The purification of 9 was carried out by reversed
phase chromatography using 10 mM aqueous ammonium
bicarbonate and acetonitrile, to give 9 in a yield of 30% from 4.
Although the N2-dmf group survives limited time treatment
with NaOH or HCl, it is slowly hydrolyzed by the ammonium
bicarbonate eluant, so that solutions of 9 should not be allowed
to stand for long periods of time after purification.
amenable to silica chromatography, and all of the subsequent
intermediates were purified on silica using gradients of
methanol (with 0.5% TEA for those with a free amino group
or the acid labile monomethoxytrityl group) and methylene
chloride. The strategy for synthesis of the linear and cyclic
dimers was to elaborate the 3′-amino group into an
isothiocyanate and to couple this to a 5′-amino group obtained
by Staudinger reduction of the 5′-azide. The 3′-isothiocyanate
is stable to silica chromatography, so that intermediates 11 and
14 are easily handled, but it does react well with the 5′-amino
group. Thus the 5′-azide functions as a stable masked amino
group that can be converted to the amine without harming the
3′-isothiocyanate of 14.
Formation of the 3′-isothiocyanate derivative 11 was carried
out by reaction of 10 with carbon disulfide followed by reaction
of the resulting dithiocarbamate (not shown) with tosyl
chloride or benzenesulfonyl chloride.28 This was done as a
two-step procedure using a 10-fold excess of CS2 in the first
step that was readily removed on a rotary evaporator before
reaction with the sulfonyl chloride. The 5′-amino nucleoside 12
was obtained by Staudinger reduction after protection of the 3′-
amino group of 10 by reaction with monomethoxytrityl
chloride. Condensation of 11 and 12 in THF at room
temperature gave clean conversion to the linear dimer 13
within 17 h in 93% yield. The monomethoxytrityl group was
removed using dichloroacetic acid (DCA), and the amino
group converted to an isothiocyanate to give 14 by the same
two-step procedure used for preparation of 11.
The derivatization of 9 for synthesis of the cyclic dimers
required protection of the 2′-hydroxyl, conveniently done by
Cyclization of the linear dimer 14 to the cyclic dimer 15 was
effected by a two-step sequence starting with reaction of 14
159
dx.doi.org/10.1021/ol403154w | Org. Lett. 2014, 16, 158−161

Organic Letters
Letter
with triphenyl phosphine to give the azine (not shown, but
sufficiently stable to be clearly visibly by LC-MS), followed by
dilution of the reaction mixture with THF/water/TEA and
heating at 60 °C for 2 h (Scheme 4). Although LC-MS showed
Scheme 5. Syntheses of Carbodiimide 18 and Guanidines
19a and 19b
Scheme 4. Syntheses of Thiourea 16 and Urea 17
The carbodiimide 18 proved to be sufficiently stable to be
handled and purified using the same conditions used for 17 and
19a/b. All of these compounds have poor solubility in water,
but are soluble in 0.1 N NaOH. Purification of each was done
by RP chromatography using 0.1 N NaOH and methanol.
Neutralization of the product fractions using CO2 gas gave each
compound as a white solid easily isolated by filtration. The
preparations of 17, 18, and 19a/b were carried out on small
scales only and were not optimized.
Recent reports of activation of the innate immune system
with the 2′/3′ isomers of cyclic GMP-AMP (cGAMP)33−35
provide a new impetus for preparation of cyclic dinucleotides
(CDNs) and their analogs. The compounds reported here are
the first examples of a new class of CDN analogs that possess a
urea, or urea related, backbone.
that the protected cyclic dimer 15 was the only significant
product, there were a number of small impurities visible,
possibly oligomers from an intermolecular reaction, even under
the dilute conditions of the cyclization. It was again possible to
purify 15 by silica chromatography, using a steep gradient of
methanol in methylene chloride. The isolated yield for 15 was
only 40%, presumably because of competing intermolecular
reactions or degradation.
ASSOCIATED CONTENT
■
S
* Supporting Information
1
Synthetic procedures, HPLC, H, 13C NMR, and UV spectra.
The deprotection of 15 to the cyclic thiourea 16 was effected
using 2 N NaOH in methanol/water (1:1). Under these
conditions the TBS groups were removed in minutes, at room
temperature, while the N2-dmf groups required heating at 60
°C for 2 h to effect removal, consistent with the surprising
stability noted earlier. Neutralization of the reaction mixture
with either 1 N HCl or acetic acid caused precipitation of 16,
which was isolated by filtration in quantitative yield. Of the
many potential routes for conversion of thioureas to ureas,29
reaction of 16 with DMSO and catalytic iodine, at 80 °C, was
employed.30 This is a simple, if slow, procedure that does not
involve metals or unusual conditions and gave clean conversion
to 17, in 37% yield.
The reaction of 16 with iodine, this time in DMF at room
temperature with triethylamine, was also effective for
preparation of the carbodiimide 18 (Scheme 5).31 Although
this reaction is reported to require aryl thioureas,31 it worked
well for preparation of 18. The reaction of carbodiimides with
amines for synthesis of guanidines is well-known,32 and
aqueous methylamine and aqueous ammonia gave 19a and
19b, respectively, although slowly and in modest yields.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: r.jones@rutgers.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was initated with support from NIH Grant
GM79760.
■
REFERENCES
■
(1) Hengge, R. Nat. Rev. Microbiol. 2009, 7, 263−273.
(2) Krasteva, P. V.; Giglio, K. M.; Sondermann, H. Protein Sci. 2012,
21, 929−948.
(3) Mills, E.; Pultz, I. S.; Kulasekara, H. D.; Miller, S. I. Cell. Microbiol.
2011, 13, 1122−1129.
(4) Povolotsky, T. L.; Hengge, R. J. Biotechnol. 2012, 160, 10−16.
160
dx.doi.org/10.1021/ol403154w | Org. Lett. 2014, 16, 158−161

Organic Letters
Letter
́
(5) Quin, M. B.; Berrisford, J. M.; Newman, J. A.; Basle, A.; Lewis, R.
J.; Marles-Wright, J. Structure 2012, 20, 350−363.
(6) Sondermann, H.; Shikuma, N. J.; Yildiz, F. H. Curr. Opin.
Microbiol. 2012, 15, 140−146.
(7) Schirmer, T.; Jenal, U. Nat. Rev. Micro. 2009, 7, 724−735.
(8) Ko, J.; Ryu, K.-S.; Kim, H.; Shin, J.-S.; Lee, J.-O.; Cheong, C.;
Choi, B.-S. J. Mol. Biol. 2010, 398, 97−110.
(9) Shanahan, C. A.; Gaffney, B. L.; Jones, R. A.; Strobel, S. A. J. Am.
Chem. Soc. 2011, 133, 15578−15592.
(10) Smith, K. D.; Lipchock, S. V.; Ames, T. D.; Wang, J.; Breaker, R.
R.; Strobel, S. A. Nat. Struct. Mol. Biol. 2009, 16, 1218−1223.
(11) Smith, K. D.; Shanahan, C. A.; Moore, E. L.; Simon, A. C.;
Strobel, S. A. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 7757−7762.
(12) Sudarsan, N.; Lee, E. R.; Weinberg, Z.; Moy, R. H.; Kim, J. N.;
Link, K. H.; Breaker, R. R. Science (Wash.) 2008, 321, 411−413.
(13) Karaolis, D. K. R.; Means, T. K.; Yang, D.; Takahashi, M.;
Yoshimura, T.; Muraille, E.; Philpott, D.; Schroeder, J. T.; Hyodo, M.;
Hayakawa, Y.; Talbot, B. G.; Brouillette, E.; Malouin, F. J. Immunol.
2007, 178, 2171−2181.
(14) Woodward, J. J.; Iavarone, A. T.; Portnoy, D. A. Science 2010,
328, 1703−1705.
(15) Burdette, D. L.; Monroe, K. M.; Sotelo-Troha, K.; Iwig, J. S.;
Eckert, B.; Hyodo, M.; Hayakawa, Y.; Vance, R. E. Nature 2011, 478,
515−518.
(16) Gaffney, B. L.; Veliath, E.; Zhao, J.; Jones, R. A. Org. Lett. 2010,
12, 3269−3271.
(17) Kiburu, I.; Shurer, A.; Yan, L.; Sintim, H. O. Mol. BioSyst. 2008,
4, 518−520.
(18) Yan, H.; Wang, X.; KuoLee, R.; Chen, W. Biorg. Med. Chem.
Lett. 2008, 18, 5631−5634.
(19) Hyodo, M.; Sato, Y.; Hayakawa, Y. Tetrahedron 2006, 62, 3089−
3094.
(20) Shanahan, C. A.; Gaffney, B. L.; Jones, R. A.; Strobel, S. A.
Biochemistry 2013, 52, 365−377.
(21) Kline, T.; Jackson, S. R.; Deng, W.; Verlinde, C. L. M. J.; Miller,
S. I. Nucleosides Nucleotides Nucl. Acids 2008, 27, 1282−1300.
(22) Zhou, J.; Watt, S.; Wang, J.; Nakayama, S.; Syre, D. A.; Lam, Y.;
Lee, V. T.; Sintim, H. O. Biorg. Med. Chem. 2013, 21, 4396−4404.
(23) McGee, D. P. C.; Martin, J. C. Can. J. Chem. 1986, 64, 1885−
1889.
(24) Dean, D. K. Synth. Commun. 2002, 32, 1517−1521.
(25) He, G.-X.; Bischofberger, N. Tetrahedron Lett. 1995, 36, 6991−
6994.
(26) Zhang, L.; Cui, Z.; Zhang, B. Helv. Chim. Acta 2003, 86, 703−
710.
̌
(27) Kroutil, J.; Trnka, T.; Cerny, M. Synthesis 2004, 446−450.
́
(28) Wong, R.; Dolman, S. J. J. Org. Chem. 2007, 72, 3639−3971.
(29) Sahu, S.; Sahoo, P. R.; Patel, S.; Mishra, B. K. J. Sulfur Chem.
2011, 32, 171−197.
(30) Mikołajczyk, M.; Łuczak, J. Synthesis 1975, 114−115.
(31) Ali, A. R.; Ghosh, H.; Patel, B. K. Tetrahedron Lett. 2010, 51,
1019−1021.
(32) Katritzky, A. R.; Rogovoy, B. V. ARKIVOC 2005, 2005, 49−87.
(33) Gao, P.; Ascano, M.; Wu, Y.; Barchet, W.; Gaffney, B. L.;
Zillinger, T.; Serganov, A. A.; Liu, Y.; Jones, R. A.; Hartmann, G.;
Tuschl, T.; Patel, D. J. Cell 2013, 153, 1094−1107.
(34) Wu, J.; Sun, L.; Chen, X.; Du, F.; Shi, H.; Chen, C.; Chen, Z. J.
Science 2013, 339, 826−830.
(35) Ablasser, A.; Goldeck, M.; Cavlar, T.; Deimling, T.; Witte, G.;
Rohl, I.; Hopfner, K.-P.; Ludwig, J.; Hornung, V. Nature 2013, 498,
̈
380−384.
161
dx.doi.org/10.1021/ol403154w | Org. Lett. 2014, 16, 158−161