Optimisation of aryl substitution leading to potent methionyl tRNA synthetase inhibitors with excellent Gram-positive antibacterial activity

Article abstract of DOI:10.1016/S0960-894X(02)01027-2

Optimisation of the left-hand-side aryl moiety of a file compound screening hit against Staphylococcus aureus methionyl tRNA synthetase led to the identification of a series of potent nanomolar inhibitors. The best compounds showed excellent antibacterial activity against staphylococcal and enterococcal pathogens, including strains resistant to clinical antibiotics.

Full text of DOI:10.1016/S0960-894X(02)01027-2

Bioorganic & Medicinal Chemistry Letters 13 (2003) 665–668
Optimisation of Aryl Substitution Leading to Potent Methionyl
tRNA Synthetase Inhibitors with Excellent Gram-Positive
Antibacterial Activity
Richard L. Jarvest,a,* John M. Berge,a Murray J. Brown,a Pamela Brown,a
John S. Elder,a Andrew K. Forrest,a C. S. V. Houge-Frydrych,a Peter J. O’Hanlon,a
David J. McNair,a Stephen Rittenhouseb and Robert J. Shepparda
aGlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
bGlaxoSmithKline, 1250 South Collegeville Road, Collegeville, PA 19426, USA
Received 16 September 2002; revised 19 November 2002; accepted 19 November 2002
Abstract—Optimisation of the left-hand-side aryl moiety of a file compound screening hit against Staphylococcus aureus methionyl
tRNA synthetase led to the identification of a series of potent nanomolar inhibitors. The best compounds showed excellent anti-
bacterial activity against staphylococcal and enterococcal pathogens, including strains resistant to clinical antibiotics.
# 2003 Elsevier Science Ltd. All rights reserved.
Bacterial resistance to established antibiotics continues
to pose an increasing problem in clinical practice. This
is particularly evident in the hospital context, with the
rise of methicillin resistant Staphylococcus aureus
(MRSA)1 and of vancomycin resistant enterococci
(VRE),2 many strains of which have become resistant to
all antibiotics. The resulting medical need has stimu-
lated the search for new classes of antibacterial agents
that act at novel molecular targets.
activity. Here we report the structure–activity relation-
ships of the left-hand side moiety that greatly enhanced
MRS inhibition and resulted in potent Gram-positive
antibacterial activity.
Alternative modes of attachment of the aryl moiety
were investigated in order to identify the preferred link-
age and chemistry for array amplification of the lead 2.
Direct acylation of the key amine intermediate 3 affor-
ded the amide 4 (Scheme 1).
We recently reported the discovery of a new class of
inhibitors of bacterial methionyl tRNA synthetase
(MRS) that had useful activity against staphylococci
and enterococci, including in vivo efficacy.3 Compound
1 was identified as a high throughput screening hit with
an IC50 value of 350 nM against S. aureus MRS but no
Gram-positive antibacterial activity.
The ether and thioether linked analogues 11 and 12
were synthesised by first preparing the appropriate aryl
linker units 7 and 8 (Scheme 2). These were then reacted
with the methoxybenzyloxy quinoline 10,4 utilised to
allow unmasking of the quinolone functionality under
mild conditions (TFA at room temperature).
The compounds 4, 11 and 12 were tested for inhibition
of MRS in an aminoacylation assay.3 They were also
tested for antibacterial activity against S. aureus and
Enterococcus faecalis in a standard MIC assay (Table
1). The amide and ether analogues were clearly less
potent than the amine and were without significant
antibacterial activity.
Compound 2 was derived as an early chemistry lead with
improved MRS inhibition and target-related antibacterial
Array amplification was thus carried out around the
secondary amine scaffold of 2. The reductive amination
*Corresponding author. Tel.: +44-1279-627629; fax: +44-1279-
622260; e-mail: richard_l_jarvest@gsk.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(02)01027-2

666
R. L. Jarvest et al. / Bioorg. Med. Chem. Lett. 13 (2003) 665–668
Scheme 1. Reagents: (i) 3,4-diClPhCO2H/DIC/N-HO-succinimide/
DMF.
Scheme 3. Reagents: (i) CuCN/DMF/80 ꢀC.
Scheme 4. Reagents: (i) NaSMe/Cu2O/DMF/80 ꢀC; (ii) pyridinium
tosylate/H2O/acetone.
with MCPBA followed by deprotection gave the 5-sulf-
oxide and 5-sulfone aldehydes.
2-Amino 3,5-disubstituted aldehyde intermediates were
prepared by the specific chemistries shown in Scheme 5,
due to the aniline nitrogen of the 3,5-dibromo anthra-
nilate 19 being resistant to standard alkylation methods.
Scheme 2. Reagents: (i) 3,4-diClbenzyl-Cl/NaH/THF/Á; (ii) 3,4-diCl-
benzyl-Cl/NaH/THFthen TFA; (iii) 4-methoxybenzyl-OH/NaH/15-
crown-5/THF; (iv) 7 or 8 toluene/Á; (v) TFA.
The substituted phenyl analogues were tested in the
standard way (Table 2). Many of the compounds were
very potent MRS inhibitors. However, due to the
enzyme concentration in the assay (3 nM), many of the
IC50 values are approaching the observable tight bind-
ing limit and there is thus little differentiation for the
most potent compounds.
Table 1. MRS IC50 values and in vitro minimum inhibitory concen-
tration (MIC) of compounds against strains of S. aureus and E. fae-
calis
Compd
IC50 (nM)
MIC (mg/mL)
The unsubstituted phenyl derivative 25 was a poor
MRS inhibitor and had no antibacterial activity (Table
2). Of a set of dichlorophenyl isomers tested 26–30, the
3,5-isomer gave the most potent MRS inhibition and
the best antibacterial activity. More lipophilic
3,5-dihalo analogues 31–33 gave potent inhibition and
increased antibacterial activity.
S. aureus
MRS
S. aureus
Oxford
E. faecalis
1
2
4
11
12
16
140
50
4
>64
>64
>64
2
>64
>64
>64
170
of 3 was optimised for array synthesis using resin-bound
cyanoborohydride as the reducing agent.5 Purification
was performed by parallel chromatography6 and the
products were converted to their hydrochloride salts by
addition of 1 M HCl in methanol to a methanolic solu-
tion of the amine.
By elaboration of the 3,5-dichloro and 3,5-dibromo
analogues it was found that 2,3,5-trisubstitution was
optimal, for example 34. At the 2-position, small alkoxy
groups such as ethoxy (35 and 38) were preferred over
branched alkoxy or amino substituents. Variation at the
A range of commercially available aldehydes was uti-
lised in the initial arrays. As the preferred substitution
pattern became evident, targetted 2,3,5-trisubstituted
aldehydes were synthesised. 2-Alkoxy-3,5-substituted
benzaldehydes were prepared by three routes: (i) 2-O-
alkylation of 3,5-substituted salicaldehydes (or salicyclic
esters followed by conversion of the ester to aldehyde);
(ii) electrophilic substitution of 3-substituted salicalde-
hydes; (iii) copper-catalysed aryl substitution reactions.
An example of the copper-catalysed chemistry is the
formation of the 3-cyano analogue 14 (Scheme 3). In
some cases protection of the aldehyde functionality was
necessary, as in the synthesis of the 5-methylthio analo-
gue 17 (Scheme 4). Oxidation of the intermediate 16
Scheme 5. Reagents: (i) (CO2Me)2/KOBut/DMA/reflux; (ii) NaAl
(NEt2)3H/THF/À78 ꢀC; (iii) MeCHO/NaBH(OAc)3/CH2Cl2; (iv)
Br2/MeOH.

R. L. Jarvest et al. / Bioorg. Med. Chem. Lett. 13 (2003) 665–668
667
Table 2. MRS IC50 values and in vitro minimum inhibitory concen-
tration (MIC) of compounds against strains of S. aureus and E. fae-
calis
Table 3. MRS IC50 values and in vitro minimum inhibitory concen-
tration (MIC) of compounds against strains of S. aureus and E. fae-
calis
R
IC50 (nM)
MIC (mg/mL)
R
X
n
IC50 (nM)
MIC (mg/mL)
S. aureus E. faecalis
S. aureus
MRs
S. aureus
Oxford
E. faecalis
1,7
S. aureus
MRS
Oxford
1
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
—
2,3-diCl
2,4-diCl
2,5-diCl
3,4-diCl
3,5-diCl
3,5-diBr
3-Br, 5-I
300
15
15
9
16
3
<3
3.2
6.4
<3
14
7.2
29
4.4
11
9.1
11
18
4.3
8.8
4.9
4.6
3.1
61
4.7
23
72
5.9
<3
7.2
14
>64
8
>64
2a
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diBr
3-Cl,5-I
CH2
O
CH2
O
NH
O
O
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
5.1
6.0
3.8
<3
23
0.25
2
ꢁ0.06
0.125
4
4
4
1
4a
0.25
0.13
0.13
ꢁ0.06
0.13
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
1
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
0.25
16
2
0.13
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06a
ꢁ0.06
ꢁ0.06
0.5a
ꢁ0.06
ꢁ0.06
ꢁ0.06
0.13
ꢁ0.06
ꢁ0.06
ꢁ0.06
0.25
1
7.8
12
0.25
<0.06
<0.06
0.25
0.13
0.5
2
ꢁ0.06
0.25
0.13
0.5
1
0.25
ꢁ0.06
2
4
0.13
8
3-I,5-Cl
O
9.4
8.0
<3
<3
8.0
12
3,5-diI
2,3,5-triCl
3,5-diBr
3-Cl,5-Br
3-Br,5-Cl
3-Br,5-Me
3-I,5-Et
NH
NH
NH
NH
NH
NH
NH
2-EtO,3,5-diCl
2-AllylO,3,5-diCl
2-iPrO,3,5-diCl
2-EtO,3,5-diBr
2-CF3CH2O,3,5-diBr
2-NH2,3,5-diBr
2-NHMe,3,5-diBr
2-NHEt,3,5-diBr
2-EtO,3-Cl,5-MeO
2-EtO,3-Br,5-I
2-EtO,3-Br,5-Me
2-EtO,3-Br,5-CN
2-EtO,3-Br,5-MeO
2-EtO,3-Br,5-EtO
2-EtO,3-Br,5-MeS
2-EtO,3-Br,5-MeSO
2-EtO,3-Br,5-MeSO2
2-EtO,3-I,5-Br
2-EtO,3,5-diI
2-EtO,3-I,5-Me
2-EtO,3-I,5-CH2OH
2-EtO,3-I,5-CH2OMe
2-EtO,3-I,5-CO2Et
2-EtO,3-I,5-MeO
2-EtO,3-SMe,5-Br
2-EtO,3-Me,5-Cl
2-EtO,3-Me,5-Br
2-EtO,3-Me,5-I
2-EtO,3-Me,5-MeS
2-EtO,3-Me,5-CN
2-EtO,3-Me,5-CF3
2-EtO,3-CH2OH,5-I
2-EtO,3-CH2OMe,5-I
2-EtO,3-allyl,5-I
2-EtO,3-tBu,5-I
2-EtO,3-CN,5-Br
2-EtO,3-CO2Et,5-I
2-EtO,3-MeO,5-Br
2-EtO,3-MeO,5-I
0.25
0.5
0.25
ꢁ0.06
ꢁ0.06
ꢁ0.06
3-Me,5-Br
3-Et,5-Br
9.4
4.3
cyano and hydroxymethyl were reasonably well toler-
ated by the enzyme but generally resulted in significant
reductions in antibacterial activity. Larger substituents
resulted in loss of MRS inhibition, especially large polar
groups such as ester, sulfoxide and sulfone. Overall it
seems that the binding pockets for the 3- and 5-posi-
tions are quite constrained, preferring relatively com-
pact, non-polar substituents.
ꢁ0.06
4
ꢁ0.06
ꢁ0.06
2
8
ꢁ0.06
ꢁ0.06
0.5
>64
2
8
0.13
1
ꢁ0.06
ꢁ0.06
ꢁ0.06
ꢁ0.06
4
0.25
32
8
1
8
0.25
8
2
ꢁ0.06
4
>64
ꢁ0.06
ꢁ0.06
2
>64
8
Conceptual cyclisation of the 2-substituent onto the
benzylic a-position results in compounds of the general
structure shown in Table 3. Analogues of this type were
synthesised by reductive amination from cyclic ketones
under more forcing conditions than used for the alde-
hydes, typically in refluxing methanol.3 The cyclic
ketones were generally known compounds or prepared
by the route described for substituted tetra-
hydroquinolines.3 The tetralone starting material for 76
was isolated in low yield from the reaction of butyro-
lactone and m-dichlorobenzene7 while the iodine in the
precursor to 80 was introduced into the known chloro-
chromanone using iodine tris-trifluoroacetate.8
17
170
3.3
21
32
0.5
0.5
0.25
0.13
ꢁ0.06
ꢁ0.06
8
0.5
32
16
0.13
32
6.8
5.7
3.2
5.1
3.9
6.8
4.1
9.4
<3
120
4
0.5
32
4
21
13
12
From the series of 3,5-dichloro analogues 74–78, it was
found that the six-membered ring compounds were
more active than the five-membered (Table 3). The tet-
rahydro-naphthalene, chroman and tetrahydroquino-
line scaffolds all afforded compounds exhibiting potent
MRS inhibition and good antibacterial activity.
2
aMIC against E. faecalis 1 not usable; the MIC given is for the strain
E. faecalis 7.
3- and 5-positions showed halogen and methylthio to
give the best antibacterial activity.
A smaller range of substituents were investigated at the 3-
and 5-positions on the chroman and tetrahydroquinoline
scaffolds, based on those that were preferred in the phenyl
series. The preference for halogen was more pronounced
than in the phenyl series, with only an ethyl substituent
approaching the level of antibacterial activity of the
More polar substituents were investigated to reduce
lipophilicity. A methoxy group was tolerated in the
5-position, giving good antibacterial activity, 47. Other
small polar groups of slightly higher polarity, such as

668
R. L. Jarvest et al. / Bioorg. Med. Chem. Lett. 13 (2003) 665–668
Table 4. Antibacterial activity of selected compounds against panels
of staphylococci and enterococci, where MIC90 is the concentration
required to inhibit 90% of the organisms9
preference for non-polar substituents. Analogues from
both the phenyl series and the cyclised chroman and
tetrahydroquinoline series afford very good anti-
bacterial activity against panels of staphylococcal and
enterocccal clinical isolates. These results provide
encouragement for the potential of MRS as a target for
a novel class of Gram-positive antibacterial agents.
MIC90 (mg/mL)
S. aureus
S. epidermidis
E. faecalis
E. faecium
Amoxicillin
>16
2
8–16
2
>16
0.06
0.06
0.5
>16
35
47
49
81
82
86
84
ꢁ0.016
2
2
0.03
0.5
0.25
0.25
0.5
1
0.25
0.25
0.5
1
Acknowledgements
nd
nd
ꢁ0.016
0.03
0.25
0.06
We thank Mr. Alan Dyke for technical assistance.
0.06
ꢁ0.016
1
1
nd, not done.
References and Notes
1. Solberg, C. O. Scand. J. Infect. Dis. 2000, 32, 587.
2. Low, D. E.; Keller, N.; Barth, A.; Jones, R. N. Clin. Infect.
Dis. 2001, 32 (Suppl. 2), S133.
dihalo analogues. Mono-chloro substituted analogues
appeared optimal for MRS IC50 and antibacterial activity.
3. Jarvest, R. L.; Berge, J. M.; Berry, V.; Boyd, H. F.; Brown,
M. J.; Elder, J. S.; Forrest, A. K.; Fosberry, A. P.; Gentry,
D. R.; Hibbs, M. J.; Jaworski, D. D.; O’Hanlon, P. J.; Pope,
A. J.; Rittenhouse, S.; Sheppard, R. J.; Slater-Radosti, C.;
Worby, A. J. Med. Chem. 2002, 45, 1959.
4. In the preparation of 10, addition of the crown ether was
designed to increase the amount of 4-substitution by reducing
the postulated complexation of sodium to the quinoline nitro-
gen10 which favours 2-substitution. Under these conditions,
the crown ether increased the ratio of 4-substitution to 2-sub-
stitution from 1:7 to 1:1.
5. The preferred array procedure utilised the aldehyde (0.1
mmol), amine 3 (0.12–0.15 mmol), and resin-bound cyano-
borohydride in methanol containing 1% acetic acid.
6. The crude reaction mixture was adsorbed onto Biotage
Quad 3 samplets followed by drying in a dessicator overnight.
Subsequent parallel chromatography was carried out on silica
gel cartridges eluting with increasing concentrations of
methanolic ammonia in dichloromethane (0–10%).
7. Kerr, C. A.; Rae, I. D. Aust. J. Chem. 1978, 31, 341.
8. Fukuyama, N.; Nishino, H.; Kurosawa, K. Bull. Chem.
Soc. Jpn. 1987, 60, 4363.
The excellent Gram-positive antibacterial activity of
some of these compounds was confirmed by testing
against panels of clinical isolates to determine their
MIC90 values (the concentration required to inhibit
90% of the organisms; Table 4). These panels of
S. aureus, Staphylococcus epidermidis, E. faecalis and
Enterococcus faecium include a range of resistant organ-
isms.9 In all cases, the MRS inhibitors maintain good
activity against these pathogens. Excellent activity was
seen against the enterococci. Particularly good anti-
bacterial activity against staphylococci was seen with the
5-methylthiophenyl analogue 49 and the chroman 81.
All the MRS inhibitors described were highly selective
for the bacterial enzyme, with no significant inhibition
of mammalian (rat liver) MRS up to the highest con-
centration tested (either 1 or 10 mM).
In summary, the topology of a 2,3,5-trisubstituted benzyl-
amino group has been identified as an optimal left-hand
side moiety, affording potent nanomolar MRS inhibition
and Gram-positive antibacterial activity. Cyclisation to the
benzylic a-position is tolerated, affording very active
chroman and tetrahydroquinoline analogues. The 3- and
5-positions are relatively constrained, and display a
9. The number of isolates in each profile were S. aureus,
n=31, S. epidermidis, n=10, E. faecalis, n=10, E. faecium,
n=10. The resistance profile has been reported.3 Testing was
carried out using the NCCLS protocol.
10. Osborne, A. G.; Miller, L. A. D. J. Chem. Soc., Perkin
Trans. 1 1993, 181.