Deoxyribosyl analogues of methionyl and isoleucyl sulfamate adenylates as inhibitors of methionyl-tRNA and isoleucyl-tRNA synthetases

Article abstract of DOI:10.1016/j.bmcl.2005.05.035

2′-Deoxy, 3′-deoxy, and 2′,3′-dideoxyribosyl surrogates of isoleucyl and methionyl sulfamate adenylates have been investigated to identify the pharmacophoric importance of the ribose group for the inhibition of Escherichia coli methionyl-tRNA (MRS) and isoleucyl-tRNA (IRS) synthetases. Molecular modeling of 2′,3′-dideoxyribosyl Met-NHSO₂-AMP (9) with the crystal structure of E. coli MRS revealed that the lack of the two hydroxyl groups on ribose was compensated by the formation of an extra hydrogen bond between the ring oxygen and His24, resulting in a small activity reduction.

Full text of DOI:10.1016/j.bmcl.2005.05.035

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3389–3393
Deoxyribosyl analogues of methionyl and
isoleucyl sulfamate adenylates as inhibitors of methionyl-tRNA
and isoleucyl-tRNA synthetases
Sung Eun Kim,a Su Yeon Kim,a Sunghoon Kim,b Taehee Kangc and Jeewoo Leea,*
aLaboratory of Medicinal Chemistry, Research Institute of Pharmaceutical Sciences, College of Pharmacy,
Seoul National University, Shinlim-Dong, Kwanak-Ku, Seoul 151-742, Republic of Korea
bCenter for ARS Network, Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University,
Shinlim-Dong, Kwanak-Ku, Seoul 151-742, Republic of Korea
cImagene Co. Ltd., #406 Biotechnology Incubating Center, Seoul National University, Seoul 151-742, Republic of Korea
Received 22 March 2005; revised 6 May 2005; accepted 9 May 2005
Abstract—20-Deoxy, 30-deoxy, and 20,30-dideoxyribosyl surrogates of isoleucyl and methionyl sulfamate adenylates have been inves-
tigated to identify the pharmacophoric importance of the ribose group for the inhibition of Escherichia coli methionyl-tRNA (MRS)
and isoleucyl-tRNA (IRS) synthetases. Molecular modeling of 20,30-dideoxyribosyl Met-NHSO2-AMP (9) with the crystal structure
of E. coli MRS revealed that the lack of the two hydroxyl groups on ribose was compensated by the formation of an extra hydrogen
bond between the ring oxygen and His24, resulting in a small activity reduction.
Ó 2005 Elsevier Ltd. All rights reserved.
The aminoacyl-tRNA synthetases (aaRSs) are a family
of enzymes that precisely transfer amino acids to their
corresponding tRNA molecules to form aminoacyl-
tRNAs, which are substrates for translation in protein
synthesis and are pivotal in determining how the genetic
code is interpreted as amino acids.1 Although the activ-
ities of aminoacyl-tRNA synthetases are essential in all
living organisms, the selective inhibition of pathogen
synthetases over their human cellular counterparts pro-
vides an attractive antibacterial mode of action for dis-
covering novel classes of antibiotics, particularly for
the treatment of antibiotic-resistant bacterial strains,
such as methicillin resistant Staphylococcus aureus
(MRSA) and vancomycin resistant enterococci
(VRE).2–4
acyl moiety to the 30-terminal adenosine of tRNA. The
aminoacyl adenylate, an active intermediate, has been
exploited as a starting prototype in the search for novel
aaRS inhibitors because of its tight binding affinity,
which is generally 2 or 3 orders of magnitude greater
than those of the substrates, amino acid and ATP.
Modifications of aminoacyl adenylates have been
extensively investigated for the purpose of improving
chemical stability, tight binding, and pathogen selectiv-
ity. Many of the modifications have focused on the
linker region, because the mixed anhydride acylphos-
phate bond of the intermediate is readily susceptible
to hydrolysis. Substitutions with non-hydrolyzable
isosteres have been tested to identify stable surrogates
of aminoacyl adenylates as potent aaRS inhibitors. To
date, alkylphosphate,5–7 ester (aa-CO2-AMP),8–10
amide (aa-CONH-AMP),8 hydroxamate (aa-CON
(OH)-AMP),8–10 sulfamate (aa-NHSO3-AMP),5,6,11–14
sulfamide (aa-NHSO2NH-AMP),6 N-alkoxysulfamide
(aa-NHSO2NHO-AMP)15, and N-hydroxysulfamide
(aa-NHSO2N(OH)-AMP)15 surrogates have been stud-
ied as candidates for the replacement of the labile
acylphosphate of the aminoacyl adenylate. Among
them, the sulfamate analogues consistently exhibited
The aaRS reaction comprises two steps including the
formation of an aminoacyl adenylate (aa-AMP, 1) from
an amino acid and ATP and the transfer of the amino-
Keywords: Methionyl-tRNA synthetase inhibitor; Isoleucyl-tRNA
synthetase inhibitor.
*
Corresponding author. Tel.: +82 2 880 7846; fax: +82 2 888 0649;
e-mail: jeewoo@snu.ac.kr
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.05.035

3390
S. E. Kim et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3389–3393
Figure 1.
the most potent inhibition of enzymes, including
AlaRS,11 ArgRS,5 HisRS,5 IleRS,12,15 MetRS,15
ProRS,13 SerRS,14 ThrRS,5 and TyrRS.6
sulfamate adenylates. Interestingly, we found that the
20,30-dideoxyribosyl analogues were much more potent
than the 20-deoxy or 30-deoxy surrogates for both aaRSs.
Over the past few years, for the purpose of discovering
aaRS inhibitors as novel antibiotic candidates, we have
systematically modified isoleucyl and methionyl aden-
ylates based on three pharmacophoric regions, including
the adenine, ribose, and acylphosphate moieties (desig-
nated as the A, B, and C regions in Fig. 1, respectively),
and have reported that their sulfamate surrogates were
potent IRS and MRS inhibitors.8–10,15 In particular, 2-
iodo and 2-acetylenic Ile-NHSO2-AMP were found to
be highly potent E. coli IRS inhibitors, with IC50 values
of 14 and 39 nM, which were 7- and 2.5-fold more po-
tent than the parent sulfamate analogue, respectively.15
The syntheses of the 20-deoxyribosyl analogues (4, 5) of
isoleucyl and methionyl sulfamate adenylates are
outlined in Scheme 1. Commercially available 20-deoxy-
adenosine (10) was converted to 30-protected 20-deoxy-
adenosine (12) by two steps. Treatment of 12 with
N-(carbobenzyloxy)sulfamoyl chloride, prepared from
chlorosulfonyl isocyanate and benzyl alcohol at
ꢀ78 °C,16 and subsequent catalytic hydrogenation pro-
vided the 50-sulfamoyl intermediate 14. The condensa-
tion of 14 with N-Boc-isoleucine and N-Boc-
methionine by 1,3-dicyclohexylcarbodiimide afforded
the penultimate intermediates (15, 16), whose two pro-
tecting groups were hydrolyzed under acidic conditions
to produce the final 20-deoxyribosyl sulfamate adenyl-
ates (4, 5), respectively. The syntheses of the 30-deoxyri-
bosyl analogues (6, 7) are represented in Scheme 2, and
were accomplished from 20-protected 30-deoxyadenosine
(21), which was prepared from adenosine in four con-
As a continuation of these studies, we have now exam-
ined the effect of the 20- and 30-hydroxyl groups of ribose
on the inhibition of IRS and MRS, through the syntheses
of the 20-deoxyribosyl, 30-deoxyribosyl, and 20,30-dide-
oxyribosyl analogues of isoleucyl and methionyl
Scheme 1. Reagents and conditions: (a) TBSCl, imidazole, DMF, rt, 95%; (b) AcOH/H2O/THF (3:1:1), 0 °C to rt, 80%; (c) (i) BnOH, CSI, CH2Cl2,
ꢀ78 °C; (ii) NEt3, rt, 75%; (d) H2, Pd/C, MeOH, rt, 99%; (e) BocNHCHRCO2H, DCC, DMAP, DMF, rt, 70% for Ile, 72% for Met; (f) 0.1 M HCl in
MeOH, 0 °C to rt, 39% for Ile, 45% for Met.

S. E. Kim et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3389–3393
3391
Scheme 2. Reagents and conditions: (a) TBSCl, pyridine, DMF, rt, 40%; (b) thiocarbonylimidazole, DMF, rt, 80%; (c) (n-Bu)3SnH, AIBN, toluene,
reflux, 87%; (d) AcOH/H2O/THF (3:1:1), 0 °C to rt, 80%; (e) (i) BnOH, CSI, CH2Cl2, ꢀ78 °C; (ii) NEt3, rt, 71%; (f) H2, Pd/C, MeOH, rt, 99%;
(g) BocNHCHRCO2H, DCC, DMAP, DMF, rt, 70% for Ile, 72% for Met; (h) 0.1 M HCl in MeOH, 0 °C to rt, 40% for Ile, 45% for Met.
ventional steps, by following the same route described in
Scheme 1. The syntheses of the 20,30-dideoxyribosyl ana-
logues (8, 9), shown in Scheme 3, were completed by
employing 20,30-dideoxyadenosine (25), which was pre-
pared from the 30-deoxyadenosine intermediate (20) in
five steps, using the above protocol.
the presence of different chemical concentrations, as de-
scribed.17 The previously reported sulfamate analogues
of isoleucyl adenylate (Ile-NHSO3-AMP) (2) and
methionyl adenylate (Met-NHSO3-AMP) (3) were also
evaluated as references in our assay protocol, and
showed IC50 values of 0.265 and 0.083 lM, respectively
(see Table 1).
The synthesized methionyl and isoleucyl adenylate ana-
logues (4–9) were evaluated as inhibitors of E. coli
MetRS and IleRS, respectively. The inhibitory activities
were determined by measuring the decrease in the gener-
ation of the aminoacylation product, [3H]isoleucyl
E. coli-tRNAIle or [35S]methionyl E. coli-tRNAMet, in
In the deoxy series of isoleucyl sulfamate adenylate, we
found that the 20-deoxyribosyl isoleucyl sulfamate aden-
ylate displayed a three order of magnitude reduction in
enzyme inhibition (4, IC50 = 280 lM) and the 30-deoxy-
ribose analogue showed an over 100-fold reduction in
Scheme 3. Reagents and conditions: (a) Bu4NF, THF, rt, 98%; (b) TBDPSCl, imidazole, DMF, rt, 90%; (c) thiocarbonylimidazole, DMF, rt, 92%;
(d) (n-Bu)3SnH, AIBN, toluene, reflux, 63%; (e) CF3CO2H, THF, rt, 94%; (f) (i) BnOH, CSI, CH2Cl2, ꢀ78 °C; (ii) NEt3, rt, 77%; (g) H2, Pd/C,
MeOH, rt, 97%; (h) BocNHCHRCO2H, DCC, DMAP, DMF, rt, 78% for Ile, 81% for Met; (i) CF3CO2H, CH2Cl2, 0 °C to rt, 71% for Ile, 69% for
Met.

3392
S. E. Kim et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3389–3393
Table 1. Inhibitory activities of the synthesized compounds toward Escherichia coli MetRS and IleRS
Compound
R1
R2
IC50 (lM)
Ile-AMP
2
4
6
8
OH
H
OH
OH
H
0.265
280
35
OH
H
H
6.4
Met-AMP
3
5
7
9
OH
H
OH
OH
H
0.083
110
220
OH
H
H
0.383
activity (6, IC50 = 35 lM). However, the 20,30-dideoxyri-
bosyl congener exhibited a moderate 24-fold decrease in
activity (8, IC50 = 6.4 lM), which is relatively less loss as
compared to the 20-deoxyribose and 30-deoxyribose
surrogates.
hydrogen bond acceptors for the d-nitrogen atom of
the imidazole ring in His24 and water molecules. The
sulfur atom of the methionine interacted with the phe-
nolic hydroxyl of Tyr260 and the protonated N-epsilon
atom of His301 by hydrogen bonding. The ammonium
group coordinated with both the carbonyl of Leu13
and the carboxylate of Asp52 by dipole–dipole ionic
interactions.
The deoxy series of methionyl sulfamate adenylate
showed a similar SAR activity pattern. Whereas both
the 20-deoxyribose (5, IC50 = 110 lM) and 30-deoxyri-
bose analogues (7, IC50 = 220 lM) displayed dramatic
decreases in enzyme inhibition, of 1300-fold and 2600-
fold, respectively, the 20,30-dideoxyribose surrogate
showed a small loss in activity, with ca. 4-fold reduction
(9, IC50 = 0.383 lM).
In order to investigate the binding mode of 20,30-dide-
oxyribosyl Met-NHSO2-AMP (9), it was docked in the
methionyl adenylate-binding site of E. coli MetRS.19
The docked model of compound 9 is shown in Figure
2. In the proposed model, the adenine moiety forms
hydrogen bonds with water molecules. The distance
and the angle between the adenine N1 atom and the
main-chain NH group of Val326, which was forming
a hydrogen bond in compound 3, were measured as
These results indicate that the deoxyribose series of sul-
famate analogues displayed reduced potencies as com-
pared to the parent compounds, and interestingly, the
20,30-dideoxyribosyl analogues were found to be much
more potent than the 20-deoxy or 30-deoxy analogues
in both IRS and MRS.
˚
2.47 A and 141.13°, respectively. This result indicates
that the side-chains of the residues within the binding
site undergo a large conformational change, as com-
pared with the conformation within the X-ray struc-
ture of compound 3 bound into MetRS (weighted
Recently, the 3D structure of methionyl-tRNA
synthetase from E. coli complexed with compound 3
(Met-NHSO2-AMP) was determined by X-ray crystal-
lography.18 The methionine-binding pocket was report-
edly composed of residues Ala12, Leu13, Tyr15, Trp253,
Ala256, Pro257, Tyr260, Ile297, His301, and Trp305.
The adenine base formed two hydrogen bonds with
the main-chain NH group of Val326 and the d-nitrogen
atom of the His323 imidazole ring, respectively. The 20-
OH group of the ribose was hydrogen bonded to the
side-chain of Asp296, and the 30-OH group made bifur-
cated hydrogen bonds with Glu27 and Gly294. Several
water molecules formed a hydrogen-bond network with
the adenosine moiety. The sulfamoyl group acted as
˚
rms value = 1.47 A). The sulfonyl oxygen atom acts
as a hydrogen bond acceptor for a water molecule,
and the sulfonamide NH group interacts with the car-
˚
bonyl group of Leu13 by hydrogen bonding (1.8 A).
The ammonium group makes bifurcated hydrogen
bonds with both carbonyl groups of Leu13 and
Pro14. We also found a particular hydrogen bond in
the ribose ring, whose oxygen atom interacts with
˚
the NH of His24 (1.8 A). The inhibitory activity of
=
20,30-dideoxyribosyl Met-NHSO2-AMP (9, IC50
0.383 lM) represents only a 5-fold decrease as com-
pared to Met-NHSO2-AMP (3, IC50 = 0.083 lM),
while it lacks two hydroxyl groups. The high potency

S. E. Kim et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3389–3393
3393
Figure 2. Proposed model of compound 9 bound to the Escherichia coli MetRS binding site.
5. Forrest, A. K.; Jarvest, R. L.; Mensah, L. M.; OÕHanlon,
P. J.; Pope, A. J.; Sheppard, R. J. Bioorg. Med. Chem.
Lett. 2000, 10, 1871.
6. Brown, P.; Richardson, C. M.; Mensah, L. M.; OÕHanlon,
P. J.; Osborne, N. F.; Pope, A. J.; Walker, G. Bioorg.
Med. Chem. 1999, 7, 2473.
7. Bernier, S.; Dubois, D. Y.; Therrien, M.; Lapointe, J.;
Chenevert, R. Bioorg. Med. Chem. Lett. 2000, 10, 2441.
8. Lee, J.; Kang, S. U.; Kang, M. K.; Chun, M. W.; Jo, Y. J.;
Kwak, J. H.; Kim, S. Bioorg. Med. Chem. Lett. 1999, 9,
1365.
9. Lee, J.; Kang, S. U.; Kim, S. Y.; Kim, S. E.; Kang, M. K.;
Jo, Y. J.; Kim, S. Bioorg. Med. Chem. Lett. 2001, 11, 961.
10. Lee, J.; Kang, S. U.; Kim, S. Y.; Kim, S. E.; Jo, Y. J.;
Kim, S. Bioorg. Med. Chem. Lett. 2001, 11, 965.
11. Ueda, H.; Shoku, Y.; Hayashi, N.; Mitsunaga, J.; In, Y.;
Doi, M.; Inoue, M.; Ishida, T. Biochim. Biophys. Acta
1990, 1080, 126.
12. Brown, M. J. B.; Mensah, L. M.; Doyle, M. L.; Broom, N.
J. P.; Osbourne, N.; Forrest, A. K.; Richardson, C. M.;
OÕHanlon, P. J.; Pope, A. J. Biochemistry 2000, 39, 6003.
13. Heacock, D.; Forsyth, C. J.; Kiyotaka, S.; Musier-
Forsyth, K. Bioorg. Chem. 1996, 24, 273.
14. Belrhali, H.; Yaremchuk, A.; Tukalo, M.; Larsen, K.;
Berthet-Colominas, C.; Leberman, R.; Beijer, B.; Sproat,
B.; Als-Nielsen, J.; Grubel, G.; Legrand, J.-F.; Lehmann,
¨
of compound 9 may be acquired by the formation of
an extra hydrogen bond between the oxygen atom of
the ribose ring and His24.
In summary, we have synthesized 20-deoxy, 30-deoxy,
and 20,30-dideoxyribosyl surrogates of isoleucyl and
methionyl sulfamate adenylates, which were previously
described as potent IRS and MRS inhibitors, from
adenosine and 20-deoxyadenosine to identify the phar-
macophoric importance of the two hydroxyl groups on
the ribose for the inhibition of the E. coli methionyl-
tRNA and isoleucyl-tRNA synthetases. All of the deoxy
analogues displayed reduced potencies as compared to
the parent compounds. Interestingly, the 20,30-dideoxyri-
bosyl analogues were found to be much more potent
than the 20-deoxy or 30-deoxy analogues in both IRS
and MRS. The molecular modeling study of 20,30-dide-
oxyribosyl Met-NHSO2-AMP (9) with the crystal struc-
ture of E. coli MRS revealed that an extra hydrogen
bond between the ring oxygen of 9 and His24 compen-
sated for the dramatic loss of activity, due to the lack
of the two hydroxyl groups on ribose, resulting in only
a small reduction in the activity.
M.; Cusack, S. Science 1994, 263, 1432.
15. Lee, J.; Kim, S. E.; Lee, J. Y.; Kim, S. Y.; Kang, S. U.;
Seo, S. H.; Chun, M. W.; Kang, T.; Choi, S. Y.; Kim, H.
O. Bioorg. Med. Chem. Lett. 2003, 13, 1087.
Acknowledgment
16. US patent 5,726,195.
This work was supported by a grant (03-PJ2-PG4-
BD02-0001) from the Ministry of Health & Welfare,
Republic of Korea.
17. Jo, Y. J.; Lee, S. W.; Jo, M. K.; Lee, J.; Kang, M.-K.;
Yoon, J. H.; Kim, S. J. Biochem. Mol. Biol. 1999, 32, 547.
18. Crepin, T.; Schmitt, E.; Mechulam, Y.; Sampson, P. B.;
Vaughan, M. D.; Honek, J. F.; Blanquet, S. J. Mol. Biol.
2003, 332, 59.
References and notes
19. Compound 9 was modeled in the binding pocket of E. coli
MetRS, using the X-ray structure of the MetRS–com-
pound 3 complex. All of the hydrogen atoms were added
to MetRS, and then the obtained complex was fully
optimized by energy minimization using the Tripos force
1. Ibba, M.; So¨ll, D. Annu. Rev. Biochem. 2000, 69, 617.
2. van der Haar, F.; Gabius, H. J.; Cramer, F. Angew.
Chem., Int. Ed. Engl. 1981, 20, 217.
3. Schimmel, P.; Tao, J.; Hill, J. FASEB J. 1998, 12, 1599.
4. De Pouplana, L. R.; Schimmel, P. Cell. Mol. Life Sci.
2000, 57, 865.
field and Gasteiger–Huckel partial atomic charges. All
¨
computational work was done on a Silicon Graphics O2
R10000 workstation.