Synthesis of cyclic di-nucleotidic acids as potential inhibitors targeting diguanylate cyclase

Article abstract of DOI:10.1016/j.bmc.2010.07.068

Five analogs of cyclic di-nucleotidic acid including c-di-GMP were synthesized and evaluated for their biological activities on Slr1143, a diguanylate cyclase of Synechocystis sp. Slr1143 was overexpressed from the recombinant plasmid which contained the gene of interest and subsequently purified by affinity chromatography. A new HPLC method capable of separating the compound and product peaks with good resolution was optimized and applied to the analysis of the compounds. Results obtained show that cyclic di-inosinylic acid 1b demonstrates a stronger inhibition on Slr1143 than c-di-GMP and is a potential inhibitor for biofilm formation.

Full text of DOI:10.1016/j.bmc.2010.07.068

Bioorganic & Medicinal Chemistry 18 (2010) 6657–6665
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis of cyclic di-nucleotidic acids as potential inhibitors targeting
diguanylate cyclase
Shi Min Ching a, Wan Jun Tan a, Kim Lee Chua b, Yulin Lam a,
*
a Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore
b Department of Biochemistry, National University of Singapore, 8 Medical Drive, Singapore 117597, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
Five analogs of cyclic di-nucleotidic acid including c-di-GMP were synthesized and evaluated for their
biological activities on Slr1143, a diguanylate cyclase ofSynechocystis sp. Slr1143 was overexpressed from
the recombinant plasmid which contained the gene of interest and subsequently purified by affinity chro-
matography. A new HPLC method capable of separating the compound and product peaks with good res-
olution was optimized and applied to the analysis of the compounds. Results obtained show that cyclic
di-inosinylic acid 1b demonstrates a stronger inhibition on Slr1143 than c-di-GMP and is a potential
inhibitor for biofilm formation.
Received 1 June 2010
Revised 28 July 2010
Accepted 29 July 2010
Available online 4 August 2010
Keywords:
Cyclic di-nucleotidic acid
Inhibitor
Ó 2010 Elsevier Ltd. All rights reserved.
Diguanylate cyclase
1. Introduction
Current therapies to reduce the rate of biofilm infections in-
clude prophylactic use of antibiotics and microbiocides through
Bacteria biofilm1,2 has long been a cause of concern for medical
specialists and researchers as many chronic bacterial infections are
a result of its formation. Some of the pathogenic bacteria involved
are Vibrio cholerae,3,4 Yersinia pestis, Pseudomonas aeruginosa,5
Staphylococcus aureus,6 etc. For instance, S. aureus is an important
human and animal pathogen that is found on the skin and mucosal
surfaces of humans, specifically in the anterior nares. It is the pri-
mary and most common cause of surgical infections and nosoco-
mial bacteremia due to biofilm-based infections,6,7 which can
increase hospital stays by durations of up to 2–3 days, incurring
billions of added cost per year. In addition, biofilms are resistant
to antimicrobial and antibiotics for a myriad of reasons. The altered
living conditions of the biofilm (low pH, low pO2, high pCO2, low
hydration level, etc) may result in low metabolic activity and hence
low antimicrobial activity.1,8 Furthermore, the antimicrobial
agents may be trapped as waste or chelated by inactivating en-
zymes. Horizontal gene transfer within bacteria in biofilms also al-
lows bacteria to gain resistance.1 Quorum sensing signaling
systems allow synchronization of the target gene expression with-
in the biofilm, allowing the bacteria to evade the effects of antimi-
methods such as device coatings, device immersion, and surgical
site irrigation.12 Quorum quenching enzymes or inhibitors,13 as
well as antimicrobials to destroy persister cells, have been devel-
oped over the years. However despite these efforts, biofilms have
continued to gain resistance to antimicrobial and antibiotics,1,8–10
hence alternative therapeutic methods need to be explored.
Cyclic purine ribonucleotides such as cyclic adenosine mono-
phosphate (cAMP) and cyclic guanosine monophosphate (cGMP)
are well-studied examples of second messengers in cellular signal-
ing and function. Regulation of the intracellular levels of these cyc-
lic purine ribonucleotide molecules is responsible for the bacterial
response to external stimuli such as change in temperature, light,
pH, oxygen levels, and nutrients. Although cGMP is commonly in-
volved in cellular signaling in eukaryotic cells, prokaryotes do not
seem to use it as a signaling molecule. Instead they utilize an alter-
native, cyclic guanosine monophosphate (c-di-GMP), also known
as cyclic (30–50)diguanylic acid in cellular signaling. Synthesis of
c-di-GMP involves the conversion of two guanosine triphosphate
(GTP) molecules by diguanylate cyclases (DGCs) whereas the deg-
radation of c-di-GMP is achieved via hydrolytic cleavage of the cyc-
lic compound into guanosine monophosphate (GMP) by
phosphodiesterases (PDEs) through the GGDEF domain of DGCs
and the EAL domains in PDEs, respectively.14 Recent works have
established that c-di-GMP is a ubiquitous signaling molecule in
bacteria but not in eukaryotic cells or archaea.15 It has the ability
to regulate motility and virulence gene expression of bacteria
and, at high intracellular concentrations of c-di-GMP, biofilm for-
mation and exopolysaccharide movement.16 In addition, increases
crobial agents.9,10 In addition,
a fraction of bacteria may
differentiate into persisters cells which have extremely low meta-
bolic rate and are non-growing. These cells are resistant and may
regenerate the biofilm once the therapy has ended.11
* Corresponding author. Tel.: +65 6516 2688; fax: +65 6779 1691.
E-mail address: chmlamyl@nus.edu.sg (Y. Lam).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.07.068

6658
S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
in the extracellular concentration of c-di-GMP was discovered to
inhibit the growth of biofilm in methicillin resistant S. aureus, hu-
man and bovine intramammary mastitis isolates of S. aureus and
human epithelial cells HeLa.6 Subsequent research also proved that
c-di-GMP is an effective immunomodulator and vaccine adjuvant
against pneumococcal infection.17
Since c-di-GMP has shown to possess numerous biological
potentials and is also able to control biofilm formation via regula-
tion of c-di-GMP levels, we were interested to study the effects of
other cyclic dinucleotides as potential antimicrobial agents. Hence
we herein report the synthesis of five cyclic dinucleotides with the
same backbone structure as c-di-GMP and their inhibitory activi-
ties against Slr1143, a diguanylate cyclase of Synechocystis sp.
and 50-OH groups. The product after extraction was of high purity
(as observed from TLC) and was directly protected at the 50-OH
with a dilute concentration (so as to avoid the protection of the
30-OH) of DMTrCl under anhydrous condition to give 5. Unlike 4a
and 3b, treatment of 3c with HF–pyridine complex gave a detrity-
lated product which was insoluble in DCM. Hence workup was
simplified as only filtration was needed and the solid was washed
thoroughly with saturated NaHCO3 to neutralize the acid and re-
move the residual pyridine.
2-Cyanoethyl N,N-diisopropylchlorophosphoramidite 11 was
initially synthesized (Scheme 2) and coupled with the free 30-OH
on 5 as according to Hyodo et al.19 However 11 proved to be diffi-
cult to handle as it is highly moisture-sensitive. Furthermore,
extractive workup performed after the phosphitylation reaction
utilizing excess 11 caused the monochlorophosphoramidite to be
hydrolyzed to monohydroxyphosphoramidite which was difficult
to remove by column chromatography as it tails seriously and
eventually eluted together with 6. To circumvent this problem,
we explored using 2-cyanoethyl bis-N,N-diisopropylphosphordi-
amidite 12.21 Coupling of 5 with 12 was achieved in the presence
of 1H-tetrazole in anhydrous acetonitrile. 1H-Tetrazole besides
acting as a base to extract the proton from the 30-OH also converts
diisopropylamine to diisopropylammonium tetrazolide salt, which
precipitated out of the reaction mixture, thus simplifying the puri-
fication process. Although both 11 and 12 gave similar yields of 6
(ꢀ80%), 12 is not moisture-sensitive and thus purification of the
reaction mixture was much easier.
Next, the diisopropylamine functionality on 6 was displaced by al-
lyl alcohol with the aid of imidazolium perchlorate (IMP). Subsequent
oxidationofthediribonucleosidephosphitetodiribonucleosidephos-
phate was achieved with butanone peroxide (BPO) and finally the
DMTr protecting group was cleaved with dichloroacetic acid to pro-
vide 7 in high yields. With 7 in hand, the compound was coupled to
analogs of 6 in the presence of promoter IMP and molecular sieves
as moisture scavenger.22 This was followed by oxidation and detrity-
lation to give 8 which was then treated with NaI in refluxing acetone
to provide the alkoxide with allyl iodide as the side product. The alk-
oxideintermediatewashighlyhygroscopicandcouldonlybesuccess-
fully isolated when it was precipitated from cold anhydrous acetone.
Treatment of alkoxide with N-methylimidazole (N-MeIm) as the
nucleophilic catalyst and 2,4,6-triisopropylbenzenesulfonyl chloride
(TPSCl) as the condensing agent according to the procedure of Hyodo
et al.19 facilitated the cyclization of 8. However the yields obtained
were generally low and the cyclization reaction did notgo tofullcom-
pletion even after 60 h. In addition, long reaction time also resulted in
the formation of many side products (as shown by TLC analysis). Thus
to circumventtheseproblems, we proceededto modifythe procedure
by adding moisture scavenger (molecular sieves 3 Å) to the reaction
mixture and using 5–7 equiv of TPSCl and N-MeIm. The reaction
was completed in 36 h with minimal side products and the cyclized
product 9 could be isolated more easily. As the reaction is diluted,
the molecules are not in proximity, hence intramolecular reaction is
promoted and minimal dimers of 8 were obtained. The major com-
pound obtained is the cyclized product 9a–d. However, the low yield
of 9a–d may be due to the presence of uncyclized products and disin-
tegration of the dimers to form monomers.
2. Results and discussion
2.1. Synthesis of c-di-GMP and its analogues
There are two general approaches towards the synthesis of c-di-
GMP (Fig. 1). The first approach synthesizes the cyclic backbone
first with the base introduced downstream of the synthesis, as
demonstrated by Giese and co-workers.18 Although this method al-
lows base derivatives of c-di-GMP to be made from a common syn-
thetic intermediate, the analogues prepared can only have
identical bases on both riboses. The second approach utilizes a
nucleoside as the starting material. Hence besides obtaining ana-
logues with identical nucleosides, those with different combina-
tions of nucleosides can also be prepared. This allows a more
diverse library of cyclic dinucleotides to be formed. Thus for our
synthesis, we have chosen to utilize the second approach and the
synthetic route used (Scheme 1) was adapted and optimized from
the procedure reported earlier by Hyodo et al.19
20-O-TBDMS-protected nucleoside 3 was prepared by regiose-
lectively protecting the 30- and 50-hydroxyl groups20 of 2 before
protecting the 20-OH with TBDMSCl. To prepare 3a, guanosine 2a
was first suspended in DMF at 0 °C. The di-tert-butylsilanediyl
ditriflate then added preferentially reacted with the 50-OH then
with the 30-OH to form the protected diol that was soluble in
DMF. This allowed more guanosine to dissolve, resulting in a clear
solution 45 min later. Imidazole was then added to neutralize the
triflic acid formed and to serve as a nucleophilic catalyst for the
protection of the 20-OH by TBDMSCl to form a white precipitate
3a. The free amine on the purine ring of 3a was subsequently pro-
tected with a dimethylformamidine group to provide 4a. An imine
protecting group was chosen because it is susceptible to strong
bases and could be cleaved at the end of the synthesis together
with the b-cyanoethyl protecting group. For uridine 2b, it was sol-
uble in DMF but not inosine 2c. However interestingly, 3b and 3c
did not precipitate out of the reaction mixture and thus had to
be purified by extraction and column chromatography.
To allow the 30-OH to be phosphorylated, the silyl diol protect-
ing group was cleaved with HF–pyridine complex to free the 30-
O
O
OH
O
PG
P
OH
OH
O
OH
OH
O
O
Treatment of 9 with a 1:1 mixture of concentrated aqueous
ammonia and methanol followed by triethylammonium fluoride
resulted in the deprotection of the dimethylformamidine, b-cyano-
ethyl, and silyl groups. The resulting crude product was purified by
reverse-phase HPLC to provide 1a–e in 10–15% overall yields.
O
O
OH
O
O
PG
P
OH
O
O
O
O
OH
OH
B2
PG
P
B
O
HO
O
O
B1
O
O
OH
O
O
PG
B
P
O
HO
O
O
O
O
2.2. Overexpression and purification of Slr1143
P
PG
OH
PG O O PG
OH
Gomelsky and co-workers have recently isolated PCR-amplified
bacterial DNA fragments coding for the GGDEF domain-containing
Figure 1. Retrosynthetic analysis of two major approaches for synthesizing c-di-
GMP.

S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
6659
Gdmf
B
B
O
HO
O
(Me)2NCH(OMe)2
MeOH, 50 ºC, 5 h
O
O
O
1. t-Bu2Si(OTf)2, DMF, 0 ºC, 45 min
3a
Si
Si
2. Imidazole, 0 ºC to rt, 30 min
3. TBDMSCl, 60 ºC, 2h
OH OH
O
OTBDMS
OTBDMS
O
2a: B = G
2b: B = U
2c: B = I
3a-3c
4a
DMTrO
B
1.
MS 3A,
OH
CH3CN, 30 min
O
N
B
HO
O
1. HF-pyridine, 0 ºC,
CH2Cl2, 2 h
O
DMTrO
B
3b
(NCCH2CH2O)P(N(i-C3H7)2)2
O
P
OTBDMS
2. IMP, 45 min
O
3c
4a
O
O
NC
OTBDMS
P
O
2. DMTrCl, pyridine,
rt, 12 h
NC
3. BPO, 10 min
1H-Tetrazole, ACN, rt
OH OTBDMS
O
4. CHCl2COOH, 0 ºC, 5 min
7a-7c
5a-5c
6a-6c
1. 6, MS 3A,
CH3CN, 30 min
CN
O
O
O
HO
O
B1
O
OH
1. Conc. aq. NH3-MeOH
50 ºC, 12 h
1. NaI, acetone,
reflux, 2-3 h
O
P
P
TBDMSO
OH
O
O
O
B2
2. IMP, 45 min
B2
O
O
O
O
OTBDMS
P
O
NC
O
3. BPO, 10 min
4. CHCl2COOH,
2. TPSCl, MS 3A,
O
2. (C2H5)3N.3HF, rt, 12 h
B2
O
O
B1
O
O
B1
O
OTBDMS
OH
P
P
M-MeIm, 36-48 h,
rt
O
O
HO
O
0 ºC, 5 min
O
NC
O
OTBDMS
P
NC
O
O
9a: B1 = B2 = Gdmf 9b: B1 = B2 = I
9c: B1 = B2 = U
8a: B1 = B2 = Gdmf 8b: B1 = B2 = I
8c: B1 = B2 = U
1a: B1 = B2 = G
1c: B1 = B2 = U
1b: B1 = B2 = I
1d: B1 = G, B2 = U
9d: B1 = Gdmf, B2 = U
8d: B1 = Gdmf, B2 = U
1e: B1 = I, B2 = U
9e: B1 = I, B2 = U
8e: B1 = I, B2 = U
Scheme 1. Synthesis of cyclic di-nucleotidic acid 1 (IMP = imidazolium perchlorate).
(50 mM Tris–HCl (pH 7.6), 10 mM MgCl2, 0.5 mM EDTA, and
50 mM NaCl) for 2 min. The reaction mixture was aliquoted out
every 30 s and quenched by heating at 95 °C for 3 min. The exper-
iment was repeated twice and a graph of product (c-di-GMP) peak
area was plotted against time to obtain the rate of reaction. Subse-
quently, the average rate of reaction was plotted against the GTP
concentration. From the results obtained (Fig. 2), it was observed
that the rate of reaction initially increases before undergoing a de-
cline. This indicates that Slr1143 exhibits allosteric product inhibi-
tion. At low concentrations of GTP, the amount of c-di-GMP formed
was too low to inhibit the enzyme allosterically. As the GTP con-
centration increases, the frequency of collision between Slr1143
and the GTP increases, thus increasing in the rate of reaction. How-
ever the rate of reaction reaches a maximum and thereafter, the
amount of c-di-GMP synthesized will be high enough to inhibit
the activity of Slr1143, resulting in a reduction of the rate of reac-
tion. From Figure 2, the optimum substrate concentration at which
the enzyme activity exhibits the maximum rate was found to be
Cl
O
CN
CH3CN, 0 ºC, 2 h
P
CN
PCl3
+
HO
Cl
10
Cl
P
N
H
NC
O
N
(2 equiv)
ether, rt, 12 h
11 (78%)
10
N
P
N
H
NC
O
N
(5 equiv)
ether, rt, 12 h
12 (79%)
Scheme 2. Synthesis of 11 and 12.
proteins and cloned them into vector pMAL-c2x (New England Bio-
labs) in strain Escherichia coli DH5a 15
To evaluate the inhibitory
.
activities of compounds 1a–e against diguanylate cyclase, we ob-
tained the recombinant bacterial cells from him and overexpressed
Slr1143, a 344 amino acid diguanylate cyclase from oxygenic pho-
totroph Synechocystis sp. (cyanobacteria), from it.
The cells were lysed by adding lysozyme (1 mg/mL of buffer)
and incubating the cell mixture at 37 °C for 30 min. To further dis-
rupt the cells to facilitate the release of the Slr1143 into the
supernatant, the cells were sonicated for 10 min, 10 s on, 15 s off,
on ice. Thereafter the crude cell extract was centrifuged and the
supernatant obtained was subjected to affinity chromatography
to obtain the pure Slr1143 protein.
60000
R2=0.9527
40000
20000
0
2.3. Screening of compounds 1a–e activities against Slr1143
0
100
200
300
400
Substrate concentration/µM
Prior to the enzymatic assay, a Michaelis–Menten graph was
plotted to determine if Slr1143 was inactivated by an allosteric
inhibitor like other DGCs.23 To obtain this graph, different concen-
trations of substrate GTP were introduced to a mixture of pre-incu-
Figure 2. Activity of Slr1143 at 30 °C in the presence of different substrate (GTP)
concentration. Final enzyme concentration: 1 M. Enzymatic assay buffer: 50 mM
l
Tris–HCl (pH 7.6), 10 mM MgCl2, 0.5 mM EDTA, and 50 mM NaCl. HPLC eluent pair:
20 mM triethylammonium bicarbonate buffer, pH 7.0, methanol. Reaction time:
2 min.
bated Slr1143 (Final concentration:
1 lM) and assay buffer

6660
S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
ꢀ100
l
M. Hence for our enzymatic assay, 100
lM of the respective
compound 1a–e and 100
lM of GTP were used.
For the enzymatic assay, Slr1143 was incubated with the sub-
strate, GTP, and the respective compound 1a–e for 2 min at 37 °C.
Thereafter, the reactions were quenched by heating at 95 °C and
the mixtures were filtered using a 0.2
lm HPLC filter and separated
by reverse-phase HPLC.
Phosphate buffers are commonly used for the separation of GTP
and c-di-GMP peaks with HPLC.14c,15,24 However when we applied
it to our HPLC analysis, we encountered problems of noisy baseline
and column clogging. We attempted to use standard buffer sys-
tems of CH3CN/H2O and 0.1% TFA in CH3CN/0.1% TFA in H2O but
in these running buffers, the GTP peak was not observed at
254 nm. Hence we tried triethylammonium bicarbonate25 as it is
a liquid at room temperature and would thus circumvent the prob-
lem of potential column clogging. A smoothened baseline was ob-
served and after optimizing the analysis conditions, a good
resolution of the peaks was obtained and the running time was re-
duced from 50 min to 30 min (Fig. 3). A negative peak was also ob-
served in all HPLC analysis at 15 min and this is attributed to the
elution of EDTA in the enzymatic buffer. This optimized analysis
condition was used in the screening of compounds 1a–e. The prod-
uct peak area was determined by the LC solution Ver 1.2 software.
To determine the inhibitory activity of the compounds, c-di-GMP
was assayed relative to the control. Results obtained (Fig. 4) shows
that compound 1b is a much stronger inhibitor of Slr1143 than c-
Figure 4. Effect of compounds 1a–e on the enzymatic activity of Slr1143. 100
substrate, GTP was added to a mixture of 1 M of enzyme in assay buffer and
100 M 1a–e. The reaction was quenched by heating after 2 min and analyzed by
lM of
l
l
HPLC. A reaction mixture without compound was used as a control.
di-GMP 1a, providing an inhibition of 60% at 100
lM with an IC50
value of 68.9 M (Fig. 5). Interestingly, compound 1c, where both
l
bases are replaced with uracil, and compound 1d, where only
one of the bases is replaced with uracil, were found to activate
Slr1143. Presently we have not determined how compounds 1c
and 1d activate Slr1143 although earlier studies have identified
determinants involved in diguanylate cyclase activation.26
2.4. Inhibitory activity of 1b
Figure 5. Effect of the concentration of compound 1b (cyclic di-inosinylic acid) on
the activity of Slr1143. 100
enzyme in assay buffer and 10, 25, 50, 75 and 100
without compound was used as a control. The reaction was quenched by heating
after 2 min and analyzed by HPLC.
l
M of substrate, GTP was added to a mixture of 1
lM of
It is well reported that bacterial diguanylate cyclase has a feed-
back control system via allosteric non competitive product inhibi-
tion to regulate the cellular concentration of c-di-GMP.23,27
Dimeric base-intercalated c-di-GMPs are known to act as inter-
subunit cross linker, linking a primary inhibition site (Ip) with a
secondary inhibition site (Is) of an adjacent dimer, locking the en-
zyme in the inactive state. Thus the two active half sites which
contain the monomer could not come within proximity to form
the dimer, c-di-GMP. This feedback control, which is also known
as inhibition by domain immobilization, avoids excessive GTP con-
sumption, helps to set an upper limit for c-di-GMP accumulation
and might buffer against stochastic variations in cellular c-di-
GMP content.23,27
l
M of 1b. A reaction mixture
In this study, we acknowledge the varied effects analogs of c-di-
GMP may have on the bioactivity of the enzyme diguanylate cy-
clase. Compound 1b is structurally similar to c-di-GMP, except that
it has a hypoxanthine base instead of a guanine, has shown to have
an inhibitory activity, stronger than that of c-di-GMP. This could be
attributed to the lack of amino group in 1b which could result in a
tighter dimeric base-intercalation and better fit within the alloste-
ric sites than 1a. As dissociation constant decreases, more enzyme
mV(x10)
SPD Ch2:254nm
%
B.Conc.(Method)
75.0
50.0
25.0
0.0
2.5
0.0
0.0
2.5
5.0
7.5
10.0
12.5
15.0
17.5
min
Elution Time/min
Figure 3. HPLC chromatogram obtained after optimizing the conditions to obtain a good separation. HPLC eluent pair: 20 mM triethylammonium bicarbonate buffer, pH 7.0
and methanol. Absorbance is measured at 254 nm. c-di-GMP has a longer retention time (12 min) and is eluted out later than GTP (9 min).

S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
6661
remain immobilized and inactivated, thus providing a stronger
inhibitory activity.
(30 mL). The layers were separated and the aqueous layer was ex-
tracted with ether (3 ꢁ 20 mL). The combined organic extract was
washed with brine (50 mL), dried over anhydrous Na2SO4, concen-
trated in vacuo and purified by column chromatography (1:30
EtOAc/CH2Cl2 to 1:20 EtOAc/CH2Cl2 for 3b and 1:50 MeOH/CH2Cl2
to 1:40 MeOH/CH2Cl2 for 3c).
3. Conclusion
In summary, five cyclic di-nucleotidic acids were synthesized.
The compounds were evaluated for their inhibitory activities
against Slr1143 which was overexpressed in E. coli, isolated and
purified. Results obtained show that cyclic di-inosinylic acid 1b
provides a greater inhibition on Slr1143 than c-di-GMP. Going for-
ward, the inhibitory activity of 1b could be further analyzed by
assaying on diguanylate cyclase of other bacteria species to deter-
mine its potential as an inhibitor of biofilm formation.
4.2.1.1. 20-O-(tert-Butyldimethylsilyl)-30,50-O-(di-tert-butylsilane
diyl)guanosine 3a.
1H NMR (300 MHz, DMSO-d6) d 0.07 (s,
3H), 0.09 (s, 3H), 0.86 (s, 9H), 1.01 (s, 9H), 1.06 (s, 9H), 3.93–4.01
(m, 2H), 4.26–4.35 (m, 2H), 4.57 (d, J = 5.1 Hz, 1H), 5.72 (s, 1H),
þ
6.35 (s, br, 2H), 7.91 (s, 1H), 10.65 (s, 1H); C24H44O5N5Si2
(M+H+) calcd m/z 538.2876, found m/z 538.2895. Yield = 72%.
4.2.1.2. 20-O-(tert-Butyldimethylsilyl)-30,50-O-(di-tert-butylsilane
diyl)uridine 3b.
0.21 (s, 3H), 0.96 (s, 9H), 1.06 (s, 18H), 4.10–4.16 (m, 3H), 4.46 (d,
1H NMR (300 MHz, (CD3)2CO) d 0.17 (s, 3H),
4. Experimental
4.1. General
J = 4.2 Hz, 1H), 4.52 (d, J = 4.2 Hz, 1H), 5.62 (d, J = 8.1 Hz, 1H), 5.77
þ
(s, 1H), 7.60 (d, J = 8.1 Hz, 1H), 10.27 (s, br, 1H); C23H42O6N2NaSi2
All solvents and reagents were purchased from commercial
sources and used without further purification. Powdery molecular
sieves (MS) 3 Å were used without further treatment. Imidazolium
perchlorate22 and 2-cyanoethyl bis-N,N-diisopropylphosphorami-
dite21 were prepared according to reported methods. Reactions
were monitored by thin layer chromatography (TLC) using pre-
coated plates (Merck Silica Gel 60, F254) and visualized with UV
light or by charring with ninhydrin or phosphomolybdic acid. Flash
column chromatography was performed with silica (Merck, 230–
400 mesh). 1H NMR and 31P NMR spectra were recorded at 298 K
on either a Bruker ACF300, Bruker AMX500 or Bruker DPX-300
NMR spectrometer calibrated using residual undeuterated solvent
as an internal reference. The following abbreviations were used to
explain the multiplicities: s (singlet), d (doublet), t (triplet), q
(quartet), m (multiplet). The number of protons (n) for a given res-
onance was indicated as nH. ESI mass spectra were acquired using
Finnigan LCQ or Finnigan TSQ7000 spectrometer. HPLC analysis
(M+Na+) calcd m/z 521.2474, found m/z 521.2490. Yield = 67%.
4.2.1.3. 20-O-(tert-Butyldimethylsilyl)-30,50-O-(di-tert-butylsilane
diyl)inosine 3c.
1H NMR (300 MHz, CDCl3) d 0.06 (s, 3H), 0.08
(s, 3H), 0.85 (s, 9H), 0.98 (s, 9H), 1.06 (s, 9H), 4.00–4.08 (m, 2H),
4.35–4.36 (m, 1H), 4.50–4.55 (m, 1H), 4.58–4.59 (m, 1H), 5.93 (s,
1H), 8.04 (s, 1H), 8.28 (s, 1H), 12.43 (s, br, 1H);
þ
C24H42O5N4NaSi2
(M+Na+) calcd m/z 545.2586, found m/z
545.2612. Yield = 80%.
4.2.2. Synthesis of N2-(dimethylaminomethylene)-20-O-(tert-
butyldimethylsilyl)-30,50-O-(di-tert-butylsilanediyl) guanosine
4a
N,N-Dimethylformamide dimethyl acetal (6.96 mL, 52 mmol)
was added to a suspension of 3a (13 mmol) in methanol (80 mL).
The reaction mixture was stirred at 50 °C for 5 h. After the solvent
was removed in vacuo, hexane was added to precipitate out a
white solid under cooling. The solid was collected by filtration,
washed with cold hexane and dried in vacuo to afford pure 4a.
1H NMR (300 MHz, DMSO-d6) d 0.08 (s, 3H), 0.11 (s, 3H), 0.87 (s,
9H), 1.01 (s, 9H), 1.05 (s, 9H), 3.04 (s, 3H), 3.12 (s, 3H), 3.93–4.08
(m, 2H), 4.31–4.40 (m, 2H), 4.59 (d, J = 4.9 Hz, 1H), 5.88 (s, 1H),
was carried out using a Phenomenex Luna 3l-C18 column [4.6
(diameter) ꢁ 50 (height) mm]. Semi-preparative HPLC was
achieved using a COSMOSIL 5C18-AR-300 column [10 (diame-
ter) ꢁ 250 (height) mm].
All methods for the preparation of media and solutions can be
obtained from the New England Biolabs pMAL™ Protein Fusion
and Purification System Instruction Manual. Plasmid coding for
Slr1143 was kindly provided by Gomelsky.15 Enzymatic assay buf-
fer was prepared according to the procedure reported by Gomelsky
and co-workers.15 Slr1143 was overexpressed and isolated. The
protein concentration was determined to be 2.069 mg/mL by Brad-
ford assay. HPLC analysis was carried out using a Phenomenex
7.99 (s, 1H), 8.48 (s, 1H), 11.40 (s, 1H); C27H49O6N2Si2 (M+H+)
þ
calcd m/z 593.3298, found m/z 593.3311. Yield = 96%.
4.2.3. General procedure for the synthesis of 5
To a solution of 3b, 3c or 4a (10 mmol) in CH2Cl2 (40 mL) at 0 °C
was added dropwise over 15 min a chilled solution of hydrogen
fluoride–pyridine complex (3.96 g, 40 mmol) in pyridine (5 mL).
The reaction mixture was stirred at 0 °C for 2 h. Thereafter, the
reaction mixture was washed with saturated NaHCO3 solution
(2 ꢁ 30 mL) and extracted with CH2Cl2 (3 ꢁ 30 mL). The combined
organic extract was washed with saturated NaCl (40 mL), dried
over anhydrous Na2SO4, concentrated and dried in vacuo to give
the detritylated product of 3b and 4a as white solids. For detritylat-
ed 3c, it precipitated out of the reaction mixture. Saturated NaH-
CO3 solution was added to the stirred reaction mixture till
alkaline. The white solid of detritylated 3c was collected by filtra-
tion and washed with water followed by cold CH2Cl2. The resulting
dried solid of detritylated 3b, 3c and 4a were then, respectively,
dissolved in anhydrous pyridine (100 mL). Dimethoxytrityl chlo-
ride (6.78 g, 20 mmol) was added and the reaction mixture stirred
at room temperature for 12 h. The reaction was quenched by addi-
tion of MeOH (3 mL). Concentration of the reaction mixture gave a
viscous liquid which was washed with saturated NaHCO3 and the
aqueous layer was then extracted with CH2Cl2 (3 ꢁ 30 mL). The
combined organic extract was washed with saturated NaCl, dried
Luna 3l-C18 column [4.6 (diameter) ꢁ 50 (height) mm].
4.2. Synthesis of cyclic di-nucleotidic acid 1
4.2.1. General procedure for the synthesis of 3
To
a stirred suspension of the respective nucleoside 2
(15 mmol) in DMF (30 mL) at 0 °C was added di-tert-butylsilandiyl
ditriflate (6.6 mL, 18 mmol) dropwise over 15 min. After stirring at
the same temperature for 30 min, imidazole (5.1 g, 75 mmol) was
added and the resulting mixture was stirred at 0 °C for another
5 min and then at room temperature for 45 min. Thereafter tert-
butyldimethylchlorosilane (3.4 g, 22.5 mmol) was added and the
reaction mixture was stirred at 60 °C for 2 h. For 3a, the compound
precipitated from the reaction mixture and was filtered, washed
with cold methanol and dried in vacuo. For 3b and 3c, the reaction
mixture remained clear at the end of the reaction. Thus the DMF
was partially removed in vacuo and the remaining reaction mix-
ture was partially partitioned between water (30 mL) and ether

6662
S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
over anhydrous Na2SO4 and concentrated in vacuo. The resulting
residue was purified by flash column chromatography using a elu-
tion gradient from 1:1 EtOAc/hexane to 1:20 MeOH/CH2Cl2 to af-
ford pure 5a–c.
10.27 (s, br, 1H); 31P NMR (121.5 MHz, (CD3)2CO) d 150.80,
151.22, 151.43; HRMS (ESI+) C45H61O9N4NaPSi+ (M+Na+) calcd m/
z 883.3838, found m/z 883.3850. Yield = 84%.
0
4.2.4.3. 20-O-(tert-Butyldimethylsilyl)-50-O (p,p -dimethoxytrityl)
4.2.3.1.
methyl silyl)-50-O-(p,p 0-dimethoxytrityl)guanosine 5a.
N2-(Dimethylaminomethylene)-20-O-(tert-butyldi-
1H
inosine 30-[(2-cyanoethyl) N,N-diisopropylamino phosphorami-
dite] 6c.
1H NMR (300 MHz, (CD3)2CO) d ꢂ0.10 (s, 3H), 0.05,
NMR (300 MHz, CD3OD) d ꢂ0.06 (s, 3H), 0.03 (s, 3H), 0.82 (s, 9H),
2.95 (s, 3H), 3.02 (s, 3H), 3.32–3.46 (m, 2H), 3.70 (s, 6H), 4.17–
4.18 (m, 1H), 4.36–4.39 (m, 1H), 4.71 (t, J = 5.0 Hz, 1H), 6.00 (d,
J = 5.0 Hz, 1H), 6.79 (d, J = 8.9 Hz, 4H), 7.16–7.44 (m, 9H), 8.02 (s,
1H), 8.49 (s, 1H); HRMS (ESI+) C40H51O7N6Si+ (M+H+) calcd m/z
755.3583, found m/z 755.3601. Yield = 89%.
0.06 (2s, 3H), 0.83 (s, 9H), 1.14–1.34 (m, 12H), 2.55–2.63 (m, 2H),
2.79–2.83 (m, 1H), 3.43–3.48 (m, 1H), 3.55–3.62 (m, 2H), 3.66–
3.72 (m, 2H), 3.79 (s, 6H), 4.01–4.08 (m, 1H), 4.50 (m, 2H), 5.10
(m, 1H), 6.07–6.09 (m, 1H), 6.88–6.92 (m, 4H), 7.24–7.57 (m,
9H), 8.10–8.11 (m, 1H), 8.16–8.19 (m, 1H); 31P NMR (121.5 MHz,
(CD3)2CO)
d
149.35, 150.90; HRMS (ESI+) C46H61O8N6NaPSi+
(M+Na+) calcd m/z 907.3950, found m/z 907.3962. Yield = 87%.
0
4.2.3.2. 20-O-(tert-Butyldimethylsilyl)-50-O-(p,p -dimethoxytri-
tyl) uridine 5b.
1H NMR (300 MHz, (CD3OD) d 0.14 (s, 6H),
4.2.5. General procedure for preparation of 7
0.92 (s, 9H), 3.46–3.47 (m, 2H), 4.04 (s, 6H), 4.11 (m, 6H), 4.32–
4.33 (m, 2H), 5.25 (d, J = 8.1 Hz, 2H), 5.89 (s, 1H), 6.84 (d,
J = 8.7 Hz, 4H), 7.20–7.42 (m, 9H), 8.00 (d, J = 8.1 Hz, 1H); HRMS
(ESI+) C36H44O8N2NaSi+ (M+Na+) calcd m/z 683.2759, found m/z
683.2766. Yield = 90%.
To a solution of the respective compound 6 (2.5 mmol) in anhy-
drous MeCN (9 mL) was added powdery MS 3 Å (115 mg) and allyl
alcohol (0.20 mL, 3 mmol). The resulting mixture was stirred at
room temperature for 30 min. Imidazolium perchlorate (0.85 g,
5 mmol) was then added and stirring was continued for an addi-
tional 45 min. Thereafter, a 31% solution of 2-butanone peroxide/
dimethyl phthalate in toluene (1.2 mL) was added and the reaction
mixture was stirred for 10 min. After which the MS 3 Å was re-
moved by filtration through a small pad of Celite 545 and the fil-
trate was diluted with EtOAc (20 mL) and then washed with
saturated NaHCO3. The aqueous layer was extracted with EtOAc
(3 ꢁ 20 mL) and the combined organic extract was washed with
saturated NaCl, dried over anhydrous Na2SO4 and concentrated in
vacuo to obtain a viscous liquid. This liquid was then dissolved
in CH2Cl2 (20 mL), cooled to 0 °C and dichloroacetic acid
(4.12 mL, 50 mmol) was then added dropwise to the reaction mix-
ture. Thereafter, the reaction mixture was stirred for 5 min and
then saturated NaHCO3 was added till the pH was alkaline. The
aqueous layer was extracted with CH2Cl2 (3 ꢁ 20 mL) and the com-
bined organic extract was washed with saturated NaCl (30 mL),
dried over anhydrous Na2SO4 and concentrated in vacuo. The
resulting residue was purified by flash column chromatography
using an elution gradient of 1:30 MeOH/CH2Cl2 to 1:10 MeOH/
CH2Cl2 to afford 7a–c, respectively.
0
4.2.3.3. 20-O-(tert-Butyldimethylsilyl)-50-O-(p,p -dimethoxytri-
tyl) inosine 5c.
1H NMR (300 MHz, (CD3)2CO) d 0.54 (s, 3H),
0.64 (s, 3H), 1.44 (s, 9H), 3.39–3.42 (m, 2H), 4.02–4.04 (m, 2H),
4.37 (s, 6H), 4.82–4.85 (m, 1H), 5.00–5.03 (m, 1H), 5.52–5.55 (m,
1H), 6.64–6.66 (m, 1H), 7.38–7.48 (m, 4H), 7.79–8.11 (m, 9H),
8.63 (s, 1H), 8.73 (s, 1H); HRMS (ESI+) C37H44O7N4NaSi+ (M+Na+)
calcd m/z 707.2871, found m/z 707.2884. Yield = 89%.
4.2.4. General procedure for synthesis of 6
To a solution or suspension of the respective compound 5
(5 mmol) in anhydrous MeCN (30 mL) was added 1H-tetrazole
(0.35 g, 5 mmol). Thereafter, NCCH2CH2OP(N(i-C3H7)2)2 12 (3.0 g,
10 mmol) was added dropwise at room temperature and the reac-
tion mixture was stirred for 12 h. The white solid that formed was
removed by filtration and washed with EtOAc. The filtrate was con-
centrated, diluted with EtOAc (30 mL) and washed with saturated
NaHCO3. The aqueous layer was extracted with EtOAc (3 ꢁ 30 mL)
and the combined organic extract was then washed with saturated
NaCl, dried over anhydrous Na2SO4 and concentrated in vacuo. The
resulting residue was purified by flash column chromatography
using 1:2 EtOAc/hexane to afford 6b. Elution using a gradient of
3:2 EtOAc/hexane to 2:1 EtOAc/CH2Cl2 afforded purified 6a and
6c, respectively.
4.2.5.1.
methyl silyl)guanosine
7a.
N2-(Dimethylaminomethylene)-20-O-(tert-butyldi-
30-(allyl-2-cyanoethyl phosphate)
1H NMR (300 MHz, CD3OD) d ꢂ0.18, ꢂ0.17 (2s, 3H),
ꢂ0.01, 0.00 (2s, 3H), 0.78, 0.79 (2s, 9H), 2.92–2.94 (m, 2H), 3.11
(s, 3H), 3.20 (s, 3H), 3.85–3.88 (m, 2H), 4.31–4.34 (m, 2H), 4.42–
4.44 (m, 1H), 4.66–4.72 (m, 2H), 4.98–5.02 (m, 2H), 5.31–5.48
(m, 1H), 5.43–5.50 (m, 1H), 5.98 (d, J = 6.42 Hz, 1H), 6.05–6.11
(m, 1H), 8.14 (s, 1H), 8.59 (s, 1H); 31P NMR (121.5 MHz, CD3OD),
4.2.4.1.
N2-(Dimethylaminomethylene)-20-O-(tert-butyldi-
methyl silyl)-50-O-(p,p 0-dimethoxytrityl)guanosine 30-[(2-
cyano-ethyl)N,N-diisopropylaminophosphoramidite]
6a.
1H NMR (300 MHz, CD3OD) d ꢂ0.13, ꢂ0.09 (2s, 3H), 0.03,
d
ꢂ1.02, ꢂ0.99; HRMS (ESI+) C25H41O8N7PSi+ (M+H+) calcd m/z
0.04 (2s, 3H), 0.80, 0.83 (2s, 9H), 1.03–1.23 (m, 12H), 2.40–2.44 (m,
1H), 2.72–2.78 (m, 1H), 2.94, 2.96, 3.08 (3s, 6H), 3.47–3.68 (m, 4H),
3.77, 3.78 (2s, 6H), 3.84–4.01 (m, 1H), 4.29–4.40 (m, 1H), 4.46–4.51
(m, 1H), 4.88–4.97 (m, 1H), 5.99–6.05 (m, 1H), 6.81–6.87 (m, 4H),
7.21–7.48 (m, 9H), 8.02, 8.04 (2s, 1H), 8.48, 8.51 (2s, 1H); 31P NMR
(121.5 MHz, CD3OD) d 150.21, 151.72; HRMS (ESI+) C49H68O8N8PSi+
(M+H+) calcd m/z 955.4662, found m/z 955.4658. Yield = 79%.
626.2518, found m/z 626.2507. Yield = 85%.
4.2.5.2. 20-O-(tert-Butyldimethylsilyl)uridine 30-(allyl 2-cyano-
ethyl phosphate) 7b.
1H NMR (300 MHz, CD3OD) d 0.09, 0.10,
0.13, 0.14 (4s, 6H), 0.91 (s, 9H), 2.93 (t, J = 5.9 Hz, 2H), 3.82–3.85
(m, 2H), 4.28–4.37 (m, 3H), 4.54–4.58 (m, 1H), 4.65–4.71 (m,
2H), 5.31–5.35 (m, 1H), 5.42–5.49 (m, 1H), 5.77 (d, J = 8.1 Hz,
1H), 5.99–6.09 (m, 2H), 8.05 (d, J = 8.1 Hz, 1H); 31P NMR
(121.5 MHz, CD3OD) d ꢂ1.83; HRMS (ESI+) C21H34O9N3NaPSi+
(M+Na+) calcd m/z 554.1694, found m/z 554.1706. Yield = 78%.
0
4.2.4.2. 20-O-(tert-Butyldimethylsilyl)-50-O-(p,p -dimethoxytri-
tyl) uridine 30-[(2-cyanoethyl) N,N-diisopropylamino phospho-
ramidite] 6b.
1H NMR (300 MHz, (CD3)2CO) d 0.27, 0.29 (2s,
6H), 1.03, 1.04 (2s, 9H), 1.16–1.31 (m, 12H), 2.69 (t, J = 6.0 Hz,
1H), 2.84–2.88 (m, 1H), 3.57–3.82 (m, 5H), 3.87 (s, 6H), 3.94–
4.02 (m, 1H), 4.38–4.46 (m, 1H), 4.50–4.57 (m, 1H), 4.62–4.68
(m, 1H), 5.41 (dd, J = 6.6, 8.1 Hz, 1H), 6.04–6.11 (m, 1H), 6.99–
7.03 (m, 4H), 7.35–7.62 (m, 9H), 8.02 (dd, J = 8.2, 16.8 Hz, 1H)
4.2.5.3. 20-O-(tert-Butyldimethylsilyl)inosine 30-(allyl 2-cyano-
ethyl phosphate) 7c.
1H NMR (500 MHz, CDCl3) d 0.22, 0.25
(2s, 6H), 0.93, 0.96 (2s, 9H), 2.78–2.87 (m, 1H), 2.89–2.94 (m,
1H), 3.84–3.91 (m, 1H), 4.14–4.19 (m, 1H), 4.33–4.34 (m, 2H),
4.42–4.46 (m, 2H), 4.65–4.67 (m, 2H), 5.28–5.44 (m, 4H), 5.94–

S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
6663
6.10 (m, 2H), 8.51 (s, 1H), 9.87–9.96 (m, 1H), 13.59 (s, br, 1H); 31P
NMR (202.5 MHz, CDCl3) d ꢂ1.23, ꢂ1.08; HRMS (ESI+) C22H34O8N5-
NaPSi+ (M+Na+) calcd m/z 578.1807, found m/z 578.1820.
Yield = 80%.
43.5 Hz), 1.99 (s, 1H), 2.90–2.98 (m, 4H), 3.86 (m, 2H), 4.31–4.71
(m, 11H), 5.03 (d, 2H, J = 4.5 Hz), 5.33 (d, 1H, J = 10.4 Hz), 5.45 (d,
1H, J = 17.1 Hz), 6.00–5.78 (m, 2H), 6.01–6.09 (m, 2H), 7.68–7.73
(m, 1H), 8.12 (s,1H), 8.40 (s, 1H); 31P NMR (202.5 MHz, CD3OD) d
8.04, 8.68; HRMS (ESI+) C40H62O16N8NaP2Si2 (M+H+) calcd m/z
þ
4.2.6. General procedure for preparation of 8
1051.3190, found m/z 1051.3220. Yield = 53%.
A mixture of the respective compound 6 (0.9 mmol) and the
respective compound 7 (0.9 mmol) were dissolved in anhydrous
MeCN (7 mL). MS 3 Å (80 mg) was then added and the reaction
mixture was stirred at room temperature for 30 min. Thereafter,
imidazolium perchlorate (0.30 g, 1.8 mmol) was added and stirring
was continued for an additional 45 min. To the resulting mixture
4.2.7. General procedure for preparation of 9
To a solution of the respective 8 (0.24 mmol) in anhydrous ace-
tone (7.2 mL) was added sodium iodide (0.36 g, 2.4 mmol). The
resulting mixture was stirred under reflux for 2–3 h, concentrated
and chilled anhydrous acetone was then added. The white precip-
itate that formed was filtered, washed with chilled anhydrous ace-
tone and subsequently dissolved in MeOH. Thereafter, the MeOH
was removed in vacuo to obtain a colorless crystalline solid which
was then suspended in anhydrous THF (40 mL). MS 3 Å (50 mg), N-
was added
a 31% solution of 2-butanone peroxide/dimethyl
phthalate in toluene (1.2 mL) and the reaction mixture was stirred
for 10 min. After which, the MS 3 Å was removed by filtration
through a small pad of Celite 545 and the filtrate was concentrated
to obtain a viscous liquid. This liquid was then dissolved in CH2Cl2,
cooled to 0 °C and dichloroacetic acid (1.48 mL, 18 mmol) was
added dropwise to the reaction mixture. Thereafter, the reaction
mixture was stirred for 5 min and then saturated NaHCO3 was
added till the pH was alkaline. The aqueous layer was extracted
with CH2Cl2 (3 ꢁ 20 mL) and the combined organic extract was
washed with saturated NaCl, dried over anhydrous Na2SO4 and
concentrated in vacuo. The resulting residue was purified by flash
column chromatography using an elution gradient of 1:30 MeOH/
CH2Cl2 to 1:5 MeOH/CH2Cl2 to afford compound 8.
methylimidazole
(0.96 mL,
12 mmol)
and
2,4,6-tri-
isopropylbenzenesulfonyl chloride (3.63 g, 12 mmol) were added
and the resulting mixture was stirred at room temperature for
36–48 h. Water (18 mL) was then added and stirring was contin-
ued for an additional 1 h. The reaction mixture was concentrated
in vacuo and the residual material was dissolved in EtOAc, washed
with saturated NaCl and extracted with CH2Cl2 (3 ꢁ 20 mL). The
combined organic extract was dried over anhydrous Na2SO4, con-
centrated in vacuo and the resulting residue was purified by flash
column chromatography using an elution gradient of 1:30 MeOH/
CH2Cl2 to 1:10 MeOH/CH2Cl2 to afford the respective 9.
4.2.6.1. Guanylyl (30–50)guanosine 30-phosphate 8a.
1H NMR
(300 MHz, CDCl3) dꢂ0.23–0.09 (m, 12H), 0.75–0.85 (m, 18H), 2.80–
2.82 (m, 4H), 3.09, 3.12, 3.19, 3.23 (4s, 9H), 3.72–3.82 (m, 2H),
4.31–4.63 (m, 10H), 4.89–5.04 (m, 4H), 5.29–5.45 (m, 2H), 5.78–
6.02 (m, 3H), 7.80–7.91 (m, 2H), 8.40 (s, 1H), 8.61 (s, 1H); 31P
NMR (121.5 MHz, CDCl3), d ꢂ1.68, ꢂ1.41, ꢂ1.13, ꢂ0.88; HRMS
4.2.7.1. Fully protected cyclic (30–50)diguanylic acid 9a.
1H
NMR (300 MHz, CD3OD) d ꢂ0.17–0.06 (m, 12H), 0.79–0.83 (m,
18H), 2.91–2.95 (m, 4H), 3.13 (s, 6H), 3.21 (s, 6H), 4.08–4.13 (m,
2H), 4.29–4.38 (m, 4H), 4.66–4.73 (m, 4H), 5.31 (m, 1H), 5.34 (m,
1H), 5.42 (m, 1H), 5.47 (m, 1H), 5.96 (d, J = 6.6 Hz, 2H), 8.62 (s,
2H), 8.90 (s, 2H); 31P NMR (202.5 MHz, CD3OD) d ꢂ1.72, ꢂ1.69;
(ESI+) C44H68O14N14NaP2Si2 (M+Na+) calcd m/z 1157.3945, found
þ
HRMS (ESI+) C44H68O14NaP2Si2 (M+Na+) calcd m/z 1157.3945,
þ
m/z 1157.3958. Yield = 57%.
found m/z 1157.3958. Yield = 40%.
4.2.6.2. Inosinylyl(30–50)inosine 30-phosphate 8b.
1H NMR
(300 MHz, MeOD) d ꢂ0.21–0.09 (m, 12H), 0.74–0.97 (m, 18H),
2.89–2.95 (m, 4H), 3.77–3.78 (m, 2H), 4.33–4.37 (m, 4H), 4.96–
4.99 (m, 2H), 5.17 (m, 1H), 5.33–5.36 (m, 1H), 5.44–5.50 (m, 1H),
6.04–6.09 (m, 3H), 7.61–7.67 (m, 2H), 7.71–7.75 (m, 2H), 8.15 (d,
J = 5.8 Hz, 1H) 8.31 (s, 1H), 8.42 (s, 1H); 31P NMR (202.5 MHz,
4.2.7.2. Fully protected cyclic (30–50)diinosinic acid 9b.
1H
NMR (500 MHz, MeOD) d ꢂ0.14 (s, 6H), 0.11 (s, 6H), 0.78 (s,
18H), 2.98–3.00 (m, 4H), 4.24–4.28 (m, 2H), 4.58 (dd, J = 4.4,
10.7 Hz, 4H), 4.74–4.78 (m, 4H), 5.33 (dd, J = 5.1, 8.2 Hz, 2H),
5.41–5.45 (m, 2H), 6.04 (d, J = 7.6 Hz, 2H), 8.12 (s, 2H), 8.26 (s,
CDCl3) d
C
ꢂ1.64, ꢂ1.51, ꢂ1.13, ꢂ1.03, ꢂ0.95; HRMS (ESI+)
2H); 31P NMR (202.5 MHz, CD3OD)
d
0.88; HRMS (ESI+)
41H62O15N10NaP2Si2 (M+Na+) calcd m/z 1075.3302, found m/z
C
38H56O14N10NaP2Si2 (M+Na+) calcd m/z 1017.2884, found m/z
þ
þ
1075.3303. Yield = 60%.
1017.2842. Yield = 45%.
4.2.6.3. Uridylyl(30–50)uridine 30-phosphate 8c.
1H NMR
4.2.7.3. Fully protected cyclic (30–50)diuridylic acid 9c.
1H
(300 MHz, (CD3)CO) d 0.12–0.16 (m, 12H), 0.91 (s, 18H), 2.95–
3.01 (m, 6H), 3.88 (s, 2H), 4.27–4.68 (m, 12H), 4.94–4.96 (m, 2H),
5.33 (d, 1H, J = 10.4 Hz), 5.44 (d, 1H, J = 15.8 Hz), 5.66–5.72 (2H,
m), 5.91–5.98 (m, 3H), 7.74 (d, 1H, J = 8.2 Hz), 8.00 (d, 1H,
J = 5.4 Hz); 31P NMR (121.5 MHz, (CD3)CO) d ꢂ0.99, ꢂ1.20 (m);
NMR (300 MHz, MeOD) d 0.13 (s, 6H), 0.19 (s, 6H), 0.91 (s, 9H),
0.94 (s, 9H), 2.95 (t, J = 5.9 Hz, 4H), 4.24–4.68 (m, 10H), 5.70–5.76
(d, 2H), 5.82 (d, J = 6.1 Hz, 2H), 7.68 (d, J = 8.2 Hz, 2H); 31P NMR
(121.5 MHz, CD3OD) d ꢂ0.66; HRMS (ESI+) C36H56N6O16NaP2Si2
þ
(M+Na+) calcd m/z 969.2659, found m/z 969.2688. Yield = 30%.
HRMS (ESI+) C39H62N6NaO17P2Si2 (M+Na+) calcd m/z 1027.3078,
þ
found m/z 1027.3092 Yield = 79%.
4.2.7.4. Fully protected cyclic
9d.
(30–50)guanylic/uridylic acid
1H NMR (300 MHz, MeOD) d ꢂ0.10, 0.10, 0.20 (3s, 12H),
4.2.6.4. Guanylyl(30–50)uridine 30-phosphate 8d.
1H NMR
0.79, 0.90 (2s, 18H), 2.91–3.00 (m, 4H), 3.20 (s, 6H), 4.10–4.68
(m, 12H), 5.13–5.19 (m, 1H), 5.32–5.38 (m, 1H), 5.76 (d, 1H,
J = 8.1 Hz), 5.86–5.94 (m, 2H), 7.66 (d, J = 8.1 Hz, 1H), 7.95 (s, 1H),
8.69 (s, 1H); 31P NMR (121.5 MHz, CD3OD) d 0.34, 0.68; HRMS
(500 MHz, CDCl3) d ꢂ0.25–0.12 (m, 12 H), 0.78–0.89 (m, 18H),
2.82 (m, 4H), 3.11 (s, 3H), 3.19 (s, 3H), 3.72–3.91 (m, 4H), 4.29–
4.63 (m, 10H), 4.93–5.07 (m, 3H), 5.32–5.43 (m, 2H), 5.68–5.78
(m, 2H), 5.95 (m, 2H), 7.47 (m, 1H), 7.86 (s, 1H), 8.43 (s, 1H),
9.15 (m, 1H); 31P NMR (202.5 MHz, CDCl3) d ꢂ1.96, ꢂ1.80, ꢂ1.31,
(ESI+) C40H62N10O15NaP2Si2 (M+Na+) calcd m/z 1063.3, found m/z
þ
1063.3. Yield = 39%.
ꢂ0.99, ꢂ0.88; HRMS (ESI+) C43H69O16N10P2Si2 (M+H+) calcd m/z
þ
1099.3901, found m/z 1099.3899. Yield = 49%.
4.2.7.5. Fully protected cyclic (30–50)inosylyl/uridylic acid
9e.
1H NMR (300 MHz, MeOD) d ꢂ0.14–0.19 (m, 12H), 0.78–
4.2.6.5. Inosylyl(30–50)uridine 30-phosphate 8e.
(300 MHz, MeOD) d ꢂ0.18–0.16 (m, 12H), 0.84 (dd, 18H, J = 2.4,
1H NMR
1.68 (m, 18H), 2.95–3.00 (m, 4H), 4.21–4.76 (m, 11H), 4.90–5.37
(m, 3H), 5.74 (d, J = 8.3 Hz, 1H), 5.85 (d, J = 7.3 Hz, 1H), 6.03 (d,

6664
S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
1H, J = 7.0 Hz), 7.66 (d, J = 8.0 Hz, 1H), 8.11 (s, 1H), 8.25 (s, 1H); 31
P
600 nm. Three milliliters of 0.1 M isopropyl b-D-1-thiogalactopyra-
NMR (202.5 MHz, (CD3)2CO) d
0.88–3.71 (m); HRMS (ESI+)
noside (IPTG) was then added to the subculture to a final concen-
tration of 0.3 mM and the culture media was further incubated for
2 h to induce the expression of protein. After which, the cells were
centrifuged at 4 °C for 20 min at 4000g and the supernatant was
discarded. Twenty five milliliters of column buffer was added to
resuspend the cells before Halt Protease Inhibitor Cocktail (Pierce)
was added and the mixture was kept at ꢂ20 °C overnight. Lyso-
zyme was added at 1 mg/mL of buffer and incubated for 30 min.
To further disrupt the cells to facilitate the release of protein into
the supernatant, the cells were sonicated for 10 min, 10 s on, 15 s
off, on ice. The crude mixture was then centrifuged at 9000g for
30 min. The supernatant was subsequently incubated with 1 mL
of amylose beads (New England Biolabs), which were washed
beforehand according to the manufacturer’s instructions, for 1 h
at 4 °C. The targeted protein would bind to the amylase beads.
þ
C
37H56O15N8NaP2Si2 calcd m/z 993.2771, found m/z 993.2804.
Yield = 30%.
4.2.8. General procedure for preparation of 1
To a solution of the respective 9 (0.05 mmol) in MeOH (8 mL)
was added concentrated aqueous ammonium hydroxide (8 mL)
and the resulting mixture was stirred at 50 °C for 12 h. After which
the reaction mixture was concentrated and dried in vacuo to obtain
a residual material which was dissolved in triethylammonium
fluoride (1.0 mL) and stirred at room temperature for 12 h. After
which, a 1 M ammonium acetate buffer solution (10 mL) was
added and the reaction mixture was stirred vigorously at 40 °C to
precipitate a pale yellow solid. After the removal of the precipitate,
the aqueous solution was subjected to semi-preparative HPLC
using a COSMOSIL 5C18-AR-300 column [20 (diameter) ꢁ 250
(height) mm]. Elution was carried out under the following condi-
tions to obtain 1a–e, respectively: [A = water with 1% TFA,
B = 20:80 mixture of water and MeCN with 1% TFA] gradient: 0–
3 min: 100% A, 3–35 min: (linear gradient) 100% A to 85% A/15%
B, 35–45 min: 100% B, 45–55 min: 100% A; detection at 254 nm;
flow rate 3 mL/min. Relevant fractions were collected, dried and
subsequently washed with MeCN. Centrifugation afforded a white
solid which was dissolved in 1 M aqueous ammonium acetate.
Concentration in vacuo afforded the respective diammonium salt
of 1 with average overall yield of 10–15%.
4.3.2. Purification of Slr1143 diguanylate cyclase
The amylose beads were poured into an affinity column and the
flow through was collected. As the targeted protein would bind to
the amylase beads, washing with 10 column volumes of column
buffer would remove all unwanted protein. The fusion protein
was subsequently eluted with 10 mM of maltose in column buffer.
Small fractions of 0.6 mL of eluant were collected and SDS–PAGE
was carried out on all the fractions to determine which fractions
contained the fusion protein. Glycerol was added at a concentra-
tion of 20% before a Bradford assay was performed to determine
the concentration of the protein. Subsequently, the protein was ali-
4.2.8.1. Cyclic (30–50)diguanylic acid 1a.
1H NMR (500 MHz,
quoted into smaller fractions of 20
l
l for storage at ꢂ78 °C.
D2O), d 4.01–4.04 (m, 2H), 4.32–4.40 (m, 4H), 4.83 (s, 2H), 5.04
(m, 2H), 5.81 (s, 2H), 7.95 (s, 2H); 31P NMR (202.5 MHz, D2O) d
ꢂ1.05; HRMS (ESIꢂ) C20H23O14N10P2
(M ꢂ Hꢂ) calcd m/z
ꢂ
4.3.3. Analysis of the activity of the Slr1143 diguanylate cyclase
using different GTP concentrations
689.0865, found m/z 689.0849. Yield = 84%.
Enzyme (2 ll, 1 lM) (kept on ice) in enzymatic assay buffer
4.2.8.2. Cyclic (30–50)diinosylic acid 1b.
1H NMR (500 MHz,
(50 mM Tris–HCl (pH 7.6), 10 mM MgCl2, 0.5 mM EDTA, and
50 mM NaCl) was incubated at 30 °C for 5 min. Following which,
GTP (kept on ice) was added to the desired concentration (total
volume = 200 ll) and the reaction mixture was incubated at
30 °C. Fifty microliters of the reaction mixture was pipetted out
at 0.5, 1, 1.5 and 2 min and quenched by heating to 95 °C. Each
sample was filtered with a 0.2 lm HPLC filter before 10 ll was ana-
lyzed by HPLC. The experiment was then repeated for other GTP
concentrations.
D2O) d 3.92–3.98 (m, 2H), 4.12 (m, 2H), 4.51–4.60 (m, 4H), 6.09
(s, 1H), 6.21 (s, 1H), 8.10 (s, 1H), 8.30 (s, 1H), 8.63 (s, 2H); 31P
NMR (202.5 MHz, D2O) d ꢂ1.04; HRMS (ESI+) C20H23O14N8P2
þ
(M+H+) calcd m/z 661.0804, found m/z 661.0828. Yield = 85%.
4.2.8.3. Cyclic (30–50)diuridylic acid 1c.
1H NMR (300 MHz,
MeOD) d 3.97–3.94 (m, 2H), 4.40–4.28 (m, 10H), 4.56–4.48 (m,
4H), 5.5–5.70 (m, 4H), 7.90 (d, 2H, J = 7.7 Hz); 31P NMR
(121.5 MHz, D2O) d ꢂ1.08; HRMS (ESIꢂ) C18H20O16N4P2 (M ꢂ Hꢂ)
ꢂ
calcd m/z 611.0433, found m/z 611.0452. Yield = 90%.
4.3.4. Effect of compounds 1a–e on the activity of Slr1143
diguanylate cyclase
Five microliters of a 1 mM concentration of the respective com-
pound 1 in enzymatic assay buffer was added to a mixture of
4.2.8.4. Cyclic (30–50)guanylic/uridylic acid 1d.
(500 MHz, D2O) d 4.08 (m, 2H), 4.41–4.52 (m, 4H), 5.34 (m, 2H),
1H NMR
5.60 (m, 1H), 5.93 (m, 1H), 7.49 (s, 1H), 7.92 (m, 2H), 8.45 (m,
Slr1143 (1
(38 l; enzymatic assay buffer: 50 mM Tris–HCl (pH 7.6), 10 mM
MgCl2, 0.5 mM EDTA, and 50 mM NaCl) to a final concentration
of 100 M in 50 l and the mixture was incubated at 30 °C for
5 min. Following that, 5 l of 1 mM GTP (kept on ice) was added
to a final concentration of 100 M and the reaction was quenched
at 2 min by heating at 95 °C. Each sample was filtered with a
0.2 m HPLC filter before 10 l was analyzed by HPLC. For elution
lM, 2 ll) (kept on ice) in enzymatic assay buffer
1H), 8.70 (s, 1H); 31P NMR (202.5 MHz, D2O) d ꢂ0.89; MS (ESI+)
l
þ
C19H23N7O15NaP2
(M+Na+) calcd m/z 673.0547, found m/z
673.0568. Yield = 82%.
l
l
l
4.2.8.5. Cyclic (30–50) inosylic/uridylic acid 1e.
1H NMR
l
(500 MHz, D2O) d 4.05 (m, 2H), 4.51–4.36 (m, 7H), 5.42 (s, 1H),
6.16 (s, 1H), 7.89 (s, 1H), 8.16 (s, 1H); 31P NMR (202.5 MHz, D2O)
ꢂ0.97 (s); HRMS (ESIꢂ) C19H21O15N6P2 calcd m/z 635.0540, found
m/z 635.0535. Yield = 92%.
l
l
condition for the HPLC analysis was: Eluent A: 20 mM triethylam-
monium bicarbonate buffer; Eluent B: methanol; 0–2 min: A:
100%, 2–10 min: B: 12% 10–12 min B: 18% 12–20 min B: 30%.
4.3. Overexpression and purification of Slr1143 diguanylate
cyclase as an MBP (maltose binding protein) fusion
Acknowledgments
4.3.1. Large scale overexpression of protein
We thank Professor M. Gomelsky for supplying the plasmid
coding Slr1143. We also thank the National University of Singapore
for financial support of this work (ARF: R-143-000-308-112).
One liter rich LB broth with glucose (2 g) and ampicillin
(100
E. coli DH5
l
g/mL) was inoculated with 10 mL of overnight culture of
and incubated at 37 °C to an OD of 0.6–0.8 at
a

S. M. Ching et al. / Bioorg. Med. Chem. 18 (2010) 6657–6665
6665
14. (a) Ross, P.; Weinhouse, H.; Aloni, Y.; Michaeli, D.; Weinberger-Ohana, P.
Nature 1987, 325, 279; (b) Romling, U.; Gomelsky, M.; Galperin, M. Y. Mol.
Microbiol. 2005, 57, 629; (c) Ross, P.; Mayer, R.; Weinhouse, H.; Amikam, D.;
Huggirat, Y.; Benziman, M.; de Vroom, E.; Fidder, A.; de Paus, P.; Sliedregt, L. A.
J. M.; van der Marel, G. A.; van Broom, J. J. Biol. Chem. 1990, 265, 18933.
15. Ryjenkov, D. A.; Tarutina, M.; Moskvin, O. V.; Gomelsky, M. J. Bacteriol. 2005,
187, 1792.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.07.068.
References and notes
16. Jenal, U.; Malone, J. Annu. Rev. Genet. 2006, 40, 385.
17. (a) Karolis, D. K. R.; Newstead, M. W.; Zeng, X.; Hyodo, M.; Hayakawa, Y.; Bhan,
U.; Liang, H.; Standiford, T. J. Infect. Immun. 2007, 75, 4942; (b) Ogunniyi, A. D.;
Paton, J. C.; Kirby, A. C.; McCullers, J. A.; Cook, J.; Hyodo, M.; Hayakawa, Y.;
Karaolis, D. K. R. Vaccine 2008, 26, 4676.
1. del Pozo, J. L.; Patel, R. Clin. Pharmacol. Ther. 2007, 82, 204.
2. Costerton, J. W.; Stewart, P. S.; Greenberg, E. P. Science 1999, 284, 1318.
3. Rashid, M. H.; Rajanna, C.; Ali, A.; Karaolis, D. K. R. FEMS Microbiol. Lett. 2003,
227, 113.
18. Amiot, N.; Heintz, K.; Giese, B. Synthesis 2006, 24, 4230.
19. Hyodo, M.; Hayakawa, Y. Bull. Chem. Soc. Jpn. 2004, 77, 2089.
20. (a) Corey, E. J.; Hopkins, P. B. Tetrahedron Lett. 1982, 23, 4871; (b) Trost, B. M.;
Caldwell, C. G. Tetrahedron Lett. 1981, 22, 4999.
4. Tischler, A. D.; Camill, A. Mol. Microbiol. 2004, 53, 857.
5. Kulesekara, H.; Lee, V.; Brencic, A.; Liberati, N.; Urbach, J.; Miyata, S.; Lee, D. G.;
Neely, A. N.; Hyodo, M.; Hayakawa, Y.; Ausubel, F. M.; Lory, S. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 2839.
21. Nielsen, J.; Dahl, O. Nucleic Acids Res. 1987, 15, 3626.
22. Hayakawa, Y.; Kawai, R.; Hirata, A.; Sugimoto, J.-I.; Kataoka, M.; Sakakura, A.;
Hirose, M.; Noyori, R. J. Am. Chem. Soc. 2001, 123, 8165.
6. Karaolis, D. K. R.; Rashid, M. H.; Chythanya, R.; Luo, W.; Hyodo, M.; Hayakawa,
Y. Antimicrob. Agents Chemother. 2005, 49, 1029.
23. Christen, B.; Christen, M.; Paul, R.; Schmid, F.; Folcher, M.; Jenoe, P.; Meuwly,
M.; Jenal, U. J. Biol. Chem. 2006, 281, 32015.
7. National Nosocomial Infections Surveillance System, Am. J. Infect. Control 1998,
26, 522.
24. (a) Pietta, P. G.; Mauri, P. L.; Gardana, C.; Benazzi, L. J. Chromatogr., B 1997, 690,
343; (b) Rubio, I.; Pusch, R.; Wetzker, R. J. Biochem. Biophys. Methods 2004, 58,
111; (c) Arezzo, F. Anal. Biochem. 1987, 160, 57.
8. Walters, M. C.; Roe, F.; Bugnicourt, A.; Franklin, M. J.; Stewart, P. S. Antimicrob.
Agents Chemother. 2003, 47, 317.
9. Yarwood, J. M.; Bartels, D. J.; Volper, E. M.; Greenberg, E. P. J. Bacteriol. 2004,
186, 1838.
}
25. (a) Simm, R.; Morr, M.; Kader, A.; Nimtz, M.; Romling, U. Mol. Microbiol. 2004,
53, 1123; (b) Rao, F.; Yang, Y.; Qi, Y.; Liang, Z. J. Bacteriol. 2008, 190, 3622.
26. (a) De, N.; Navarro, M. V. A. S.; Raghavan, R. V.; Sondermann, H. J. Mol. Biol.
2009, 393, 619; (b) Stock, A. M. Structure 2007, 15, 887; (c) Paul, R.; Abel, S.;
Wassmann, P.; Beck, A.; Heerklotz, H.; Jenal, U. J. Biol. Chem. 2007, 282, 29170.
27. (a) Schirmer, T.; Jenal, U. Nat. Rev. Microbiol. 2009, 7, 724; (b) Chan, C.; Paul, R.;
Samoray, D.; Amiot, N. C.; Giese, B.; Jenal, U.; Schirmer, T. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 17084; (c) Hengge, R. Nat. Rev. Microbiol. 2009, 7, 263.
10. March, J. C.; Bentley, W. E. Curr. Opin. Biotechnol. 2004, 15, 495.
11. Lewis, K. Antimicrob. Agents Chemother. 2001, 45, 999.
12. Lynch, A. S.; Robertson, G. T. Annu. Rev. Med. 2008, 59, 415.
13. Balaban, N.; Cirioni, O.; Giacometti, A.; Ghiselli, R.; Braunstein, J. B.; Silvestri, C.;
Mocchegiani, F.; Saba, V.; Scalise, G. Antimicrob. Agents Chemother. 2007, 51,
2226.