Nocathiacin analogs: Synthesis and antibacterial activity of novel water-soluble amides

Article abstract of DOI:10.1016/j.bmcl.2009.04.144

Novel water-soluble amide analogs were synthesized from nocathiacin I (1) through the formation of the carboxylic acid intermediate followed by coupling to primary or secondary amines. Several compounds with potent antibacterial activity and adequate wate

Full text of DOI:10.1016/j.bmcl.2009.04.144

Bioorganic & Medicinal Chemistry Letters 19 (2009) 3531–3535
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Nocathiacin analogs: Synthesis and antibacterial activity of novel
water-soluble amides
*
Libo Xu , Amy K. Farthing , James F. Dropinski, Peter T. Meinke, Christine McCallum, Penny S. Leavitt,
Emily J. Hickey, Lawrence Colwell, John Barrett, Kun Liu
Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel water-soluble amide analogs were synthesized from nocathiacin I (1) through the formation of the
carboxylic acid intermediate followed by coupling to primary or secondary amines. Several compounds
with potent antibacterial activity and adequate water solubility were identified. Of these, compound
19 was selected for more extensive evaluation because of its excellent in vitro antibacterial activity
and in vivo efficacy, as well as clean off-target screening.
Received 2 April 2009
Revised 27 April 2009
Accepted 30 April 2009
Available online 5 May 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Nocathiacin
Water-soluble
Thiazolyl peptide
Antibacterial
Nocathiacins are a novel class of antibiotics that belong to the
thiazolyl peptide family.1,2 Nocathiacin I (1) is structurally identi-
cal to MJ347-81F4 component A,3 and displays potent in vitro anti-
bacterial activity against a variety of Gram-positive bacteria,
including many drug-resistant strains, and exhibits in vivo efficacy
in a systemicStaphylococcus aureus infection mouse model.4,5 How-
ever, the poor aqueous solubility of nocathiacin I limits its poten-
tial for further development as a hospital i.v. drug for serious
infections. Nonetheless, with its novel structure and high intrinsic
potencies, nocathiacin I provides a promising lead for the develop-
ment of new antibacterial agents with broad spectrum efficacy
against resistant Gram-positive pathogens with minimal risk of
cross-resistance to currently marketed agents. We have under-
taken an investigation to chemically modify nocathiacin I (1)
through mild and selective chemistry to improve its aqueous solu-
bility while maintaining its biological activity.
Several approaches have been applied to nocathiacin I and ana-
logs to obtain compounds with increased water solubility.6–10 As
part of our own efforts in this area, we recently discovered11 a fac-
ile conversion of nocathiacin I to an important and versatile inter-
mediate nocathiacin acid (2) via selective degradation of the
dehydroalanine side chain. In this Letter, we describe the synthesis
and SAR of a variety of novel water-soluble amide analogs utilizing
the pivotal carboxylic acid intermediate (Fig. 1).
Amides with a diverse water-solubilizing functionality attached
were conveniently prepared in high yields by coupling nocathiacin
acid (2) with primary or secondary amines via standard peptide
chemistry (PyBop or EDC/HOBt) (Scheme 1). Alternatively, cou-
pling of 2 with N-Boc-ethylenediamine followed by deprotection
with TFA gave compound 12, which was used as an intermediate
to generate structurally diverse compounds through acylation,
S
OH
R
N
O
N
S
N
N
O
S
NH
NH
O
O
NH
OH
O
O
HO
N
N
S
O
N
O
O
S
O
NH
O
O
H
N
O
Nocathiacin I (1)
R =
NH2
O
Nocathiacin acid (2) R = OH
HO
N
* Corresponding author. Tel.: +1 732 5944963; fax: +1 732 5949556.
E-mail address: libo_xu@merck.com (L. Xu).
Present address: CryoLife, Kennesaw, GA 30144, USA.
Figure 1. Structures of nocathiacins I and nocathiacin acid 2.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.04.144

3532
L. Xu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3531–3535
S
OH
S
OH
NR1R2
OH
N
b
N
a
O
O
1
N
N
2
3 -11
c, d
S
S
OH
OH
H
H
N
e or f or g
N
R
NH2
N
N
N
H
O
O
N
N
12
13 -17
Scheme 1. Reagents: (a) TFAA, Py, THF, then aq NaHCO3; (b) R1R2NH, EDC, HOBt, DMF; (c) NH2(CH2)2NHBoc, EDC, HOBt, DMF; (d) TFA/DCM; (e) RCOOH, EDC, HOBt, DMF; (f)
RCHO, NaCNBH3, MeOH, AcOH; (g) (CH3)3SiNCO, TEA, DMF.
reductive amination or urea formation under mild and selective
conditions.
The acid (2) itself exhibited little antibacterial activity, but
using it as an intermediate, we have generated novel amides bear-
ing different water-solubilizing groups. Previous SAR study in the
dehydroalanine region9 suggests acidic functional groups or poly-
hydroxyls are poorly tolerated, therefore we primarily focused on
substituents with various basic groups. Table 1 lists some repre-
Table 1
Antibacterial activity of nocathiacin analogs 3–17
S
OH
N
NR1R2
O
N
Compound
NR1R2
MICa
(
l
g/mL)
ED99b (mg/kg)
Solubilityc (mg/mL)
0.34
AT2 IC50d
NDe
(lM)
S. aureus MB2865
S. pneumo CL2883
E. faecalis C21560
1
Nocathiacin I
0.007
0.002
0.03
2
3
4
5
6
7
8
9
Nocathiacin acid
8
2
>8
ND
ND
1.12
0.190
ND
N(CH3)2
0.015
0.06
0.015
0.011
0.0038
0.06
0.06
0.00046
0.00095
0.00095
0.007
0.06
0.12
0.12
0.016
0.03
0.12
0.06
0.17
> 0.5
0.13
0.24
0.13
>0.5
0.24
>10
>10
>10
>10
>10
>10
>10
N
H
N
H
N(CH3)2
N
H
N
N
<0.010
0.260
ND
N
H
N
O
N
N
0.00048
0.0038
0.015
N
H
O
N
N(CH3)2
ND
N
N
0.102

L. Xu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3531–3535
3533
Table 1 (continued)
Compound
NR1R2
MICa
(l
g/mL)
ED99b (mg/kg)
Solubilityc (mg/mL)
AT2 IC50d
(lM)
S. aureus MB2865
S. pneumo CL2883
E. faecalis C21560
N
O
11
12
13
14
0.12
0.0038
0.25
ND
>10
>10
>10
>10
ND
N
N
NH2
0.5
0.06
2
ND
ND
N
H
H
N
0.019
0.0038
0.00059
0.00048
0.09
0.06
0.20
0.05
0.078
0.250
N
H
N
O
H
N
N
H
N
O
O
OH
15
16
H
N
>1
ND
>1
2
ND
ND
ND
10
ND
ND
N
H
OH
O
0.5
0.125
OH
H
N
OH
N
H
H
17
0.06
0.0075
0.5
ND
<10
ND
N
NH2
N
H
O
a
MIC: minimum inhibitory concentration; the minimum concentration at which 100% inhibition of bacterial growth was observed in liquid culture by turbidity method.
ED99: Dosage that induces P 2 log10 cfu kidney burden reduction in a systemic S. aureus infection mouse model (SATOA). Vancomycin is included for comparison and its
b
ED99 is normally 0.17–0.27 mg/kg.
c
Aqueous solubility data were derived from amorphous powder of lyophilizing materials as their TFA salts and measured by nephelometry.
d
AT2 IC50 values determined with four dose levels of each compound according to the literature.18
e
ND: not determined.
sentative compounds and their antibacterial activities against
three selected bacterial strains.
The potent compounds in Table 1, especially those which dem-
onstrated good in vivo efficacy, were screened for off-target liabil-
ities. These efforts revealed that a majority of the compounds
tested displayed potent binding to the angiotension II receptor
type 2 (AT2) with IC50 values ranging from below 10 nM (5) to
459 nM (10). Although AT2 receptor is expressed at low levels in
the cardiovascular system in adults, it is up-regulated during cer-
tain pathological conditions, and recent investigations have impli-
cated a role for the AT2 receptor in cardiovascular, brain and renal
function as well as in the modulation of various biological pro-
cesses involved in development, cell differentiation, tissue repair
and apoptosis.13–15 Significant efforts were directed at addressing
this off-target activity, which led to the discovery that the observed
AT2 binding can be significantly attenuated by introducing substi-
tutions at the pyridyl hydroxyl and/or indole hydroxyl group. More
specifically, methylation of pyridyl hydroxyl in compound 6 affor-
ded 19, which exhibited a 20-fold reduction in AT2 activity (IC50
A number of secondary amides bearing tertiary nitrogen as the
solubility-enhancing element, as exemplified by 3-7, exhibited po-
tent in vitro antibacterial activity with MIC ranges comparable to
nocathiacin I. Compounds 3, 5, 6 and 7 also demonstrated excellent
in vivo efficacy. In the systemic S. aureus infection mouse model
(SATOA),12 5 at 0.5 mg/kg iv produced 5.8 log reduction in bacterial
load relative to control, and the calculated ED99 is 0.13 mg/kg. Sim-
ilarly, compound 6 reduced the bacterial load by 4.68 log units
with an ED99 of 0.24 mg/kg.
Open-chain tertiary amides, exemplified by 8, demonstrated
similar in vitro antibacterial activity as compounds in the primary
amide series, but they were less efficacious when tested in the SA-
TOA model. However, some cyclic secondary amides derived from
piperazine (9–10) displayed potent antibacterial activity, both
in vitro and in vivo. Compound 11 was less potent than 9 and 10,
especially against S. aureaus and Enterococcus faecalis.
5.07 lM, Table 2), while maintaining potent antibacterial activity.
It is interesting to note that compound 12, with a terminal pri-
mary amino group, showed significantly reduced antibacterial
activity relative to compound 3. Two amides prepared from 12,
namely 13 and 14, displayed excellent activity both in vitro and
in vivo. Incorporation of non-basic polar moieties, such as bis-phe-
nol in 15, attenuated activity. Compounds obtained from reductive
amination, such as 16, showed weak antibacterial activity.
Although compound 17 displayed modest in vitro antibacterial
activity, it has limited aqueous solubility.
Table 2
Effect of substitution on AT2 binding and antibacterial activity
Compounds
MIC (
lg/mL) (S. aureus 2865)
AT2 IC50 (lM)
5
6
18
19
0.015
0.011
0.09
<0.01
0.26
0.85
5.07
0.010

3534
L. Xu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3531–3535
O
S
O
O
S
H
OH
N
NH2
N
N
b
O
a
O
N
N
1
21
20
c
X
S
O
H
N
N
N
O
N
18 X = NMe
19 X = O
Scheme 2. Reagents: (a) TMSCH2N2, THF/MeOH; (b) TFAA, Py, THF, then aq NaHCO3; (c) RNH2, EDC, HOBt, DMF.
Although modification of indole hydroxyl also mitigated AT2 bind-
ing, it often resulted in deleterious effects to antibacterial activity
(data not shown).
IC50 value of 71 lM for CYP2C8. It was neither a time-dependent
inhibitor of CYP3A4, nor an activator of human PXR.
In summary, we have synthesized a variety of amide analogs of
nocathiacin I through a versatile carboxylic acid intermediate.
Many of the newly synthesized analogs display comparable po-
tency to nocathiacin I against a wide spectrum of bacteria, but with
dramatically improved aqueous solubility at acidic pH, and are not
accessible through previously established chemistry. Compound
19 was chosen to be further evaluated because of its excellent
in vitro antibacterial activity and in vivo efficacy.
The synthesis of compounds 18 and 19 is summarized in
Scheme 2. Methylation with TMS diazomethane in THF and meth-
anol gave almost exclusively pyridyl OMe nocathiacin I (20).16
Treatment with TFAA/pyridine in THF converted 20 to the corre-
sponding acid 21, which was coupled to the requisite amine com-
ponents to form 18 and 19, respectively. Direct methylation of 5
and 6 with TMS diazomethane can also yield the desired products,
but it was difficult to separate them from bismethylated products,
especially on a large scale.
The HCl salt of compound 19 has a solubility of 68 mg/mL in
water at a native pH of 3.4. It is among the most potent nocathiacin
analogs synthesized so far. It exhibits excellent in vitro antibacte-
rial activity against a variety of Gram-positive pathogens, including
methicillin-resistant S. aureus (MRSA) and Vancomycin-resistant
Acknowledgements
We wish to acknowledge the skills and hard work of the many
scientists who provided biological support including Gowri Bhat,
Nichelle Newsome, Rosalind Jenkins, Suzy Kwon, Andrew S. Misura
and Charles J. Gill. The authours would also like to thank members
of Fermentation Development & Operations and Biopurification
Development for the supply of nocathiacin I.
enterococci (VRE) (MICs 0.00095 and 0.06
is comparable to nocathiacin I in vitro against a panel of E. faecalis
(MICs 0.015–0.06 g/mL), which is the limiting strain for this class
lg/mL, respectively). It
l
of compounds and becoming more prevalent in the clinic. In a sys-
temic S. aureus mouse infection model (SATOA, i.v., bid), the dose
References and notes
required to reduce kidney colony forming units by 99% (ED99
)
was 0.27 mg/kg for 19, versus 12 mg/kg for linezolid. In a more
stringent mouse MRSA COL thigh model,19 19 (i.v., bid) @ 1 mg/
kg dosage produced 1.22 log reduction in bacterial load relative
to control over a 24 h period, comparable to the efficacy of vanco-
mycin at 100 mg/kg.
The pharmacokinetics of 19 in preclinical species is summa-
rized in Table 3. It has a terminal half life of ꢀ5–7 h in mice, rats,
dogs and monkeys. The clearance ranges from low in dogs and
monkeys to high in rats. In bile duct-cannulated male rats dosed
i.v. with 19 at 2 mg/kg, biliary and urinary excretion of intact
parent constituted one of the elimination routes, while
metabolism accounted for another clearance pathway (data not
shown).
1. Li, W.; Leet, J. E.; Ax, H. A.; Gustavson, D. R.; Brown, D. M.; Turner, L.; Brown, K.;
Clark, J.; Yang, H.; Fung-Tomc, J.; Lam, K. S. J. Antibiot. 2003, 56, 226.
2. Leet, J. E.; Li, W.; Ax, H. A.; Matson, J. A.; Huang, S.; Huang, R.; Cantone, J. L.;
Drexler, D.; Dalterio, R.; Lam, K. S. J. Antibiot. 2003, 56, 232.
3. Sasaki, T.; Otani, T.; Matsumoto, H.; Hamada, M.; Takeuchi, T.; Hori, M. J.
Antibiot. 1998, 51, 715.
4. Pucci, M. J.; Bronson, J. J.; Barrett, J. F.; DenBleyker, K. L.; Discotto, L. F.; Fung-
Tomc, J.; Ueda, Y. Antimicrob. Agents Chemother. 2004, 48, 3697.
5. Ueda, Y.; Barrett, J.; DenBleyker, K.; Fung-Tomc, J.; Discotto, L.; Stickle, T.;
Kolek, B.; Pucci, M. Abstracts of Papers, 43rd Annual ICAAC Meeting, Chicago, IL,
Sept 14-17; American Society for Microbiology: Washington, DC, 2003;
F-1842.
6. Regueiro-Ren, A.; Naidu, B. N.; Zheng, X.; Hudyma, T. W.; Connolly, T. P.;
Matiskella, J. D.; Zhang, Y.; Kim, O. K.; Sorenson, M. E.; Pucci, M.; Clark, J.;
Bronson, J. J.; Ueda, Y. Bioorg. Med. Chem. Lett. 2004, 14, 171.
7. Naidu, B. N.; Sorenson, M. E.; Zhang, Y.; Kim, O. K.; Matiskella, J. D.;
Wichtowski, J. A.; Connolly, T. P.; Li, W.; Lam, K. S.; Bronson, J. J.; Pucci, M. J.;
Clark, J. M.; Ueda, Y. Bioorg. Med. Chem. Lett. 2004, 14, 5573.
8. Naidu, B. N.; Li, W.; Sorenson, M. E.; Connolly, T. P.; Wichtowski, J. A.; Zhang, Y.;
Kim, O. K.; Matiskella, J. D.; Lam, K. S.; Bronson, J. J.; Ueda, Y. Tetrahedron Lett.
2004, 45, 1059.
Compound 19 in vitro exhibited IC50 values P100 lM for
reversible inhibition of CYP1A2, 2C9, 2C19, 2D6 and 3A4, and an
9. Naidu, B. N.; Sorenson, M. E.; Matiskella, J. D.; Wichtowski, J. A.; Li, W.; Sauaker,
J. B.; Zhang, Y.; Connolly, T. P.; Lam, K. S.; Bronson, J. J.; Pucci, M. J.; Yang, H.;
Ueda, Y. Bioorg. Med. Chem. Lett. 2006, 16, 3545.
10. Hrnciar, P.; Ueda, Y.; Huang, S.; Leet, J. E.; Bronson, J. J. J. Org. Chem. 2002, 67,
8789.
11. Xu, L.; Farthing, A. K.; Shi, Y. J.; Meinke, P. T.; Liu, K. J. Org. Chem. 2007, 72,
7447.
12. Mice were challenged IP with S. aureus MB 2865 at 1.8 Â 103 cfu/mouse. Mice
(3/group) were treated with test compounds (i.v., b.i.d.). Kidneys were
aseptically collected 24 h after challenge.
Table 3
Mean pharmacokinetic parameters of 19 in preclinical species
Parametersa
Mouseb
Ratb
Dogb
Monkeyb
Clp (mL/min/kg)
Vdss (L/kg)
T1/2 (h)
32.1
18.7
5.0
51.3
15.5
6.4
15.6
3.3
5.4
15.0
1.2
6.5
a
Clp, plasma clearance (blood clearance for mice); Vdss, steady-state volume of
distribution; T1/2, terminal half-life.
13. Matsubara, H. Circ. Res. 1998, 83, 1182–1191.
14. Wolf, G. Nephrol. Dial. Transplant. 2002, 17, 195.
15. Kaschina, E.; Unger, T. Blood Press 2003, 12, 70.
b
The iv dose was 1 mg/kg in EtOH/PEG400/H2O (10:15:75) for all species.

L. Xu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3531–3535
3535
16. The reaction product of 1 with TMS diazomethane was previously assigned to
be indole-N-OMe nocathiacin I.6 However, several lines of evidence from our
NMR and chemical derivatization studies support the assignment of pyridyl-
OH methylated 20 as the reaction product. The NMR peak of the methyl
protons appears as a singlet at 4.3 ppm in CD3OD. In 2D NMR, NOE between
methyl protons and H-4 of pyridine moiety was observed. We also observed
coupling of methyl protons to C-3 of pyridine in HMBC. Nocathiacin I can be
converted to its deoxy indole analog nocathiacin II17 when treated with ethyl
bromopyruvate and BTPP in DMF. 20 should not undergo similar indole-OH
cleavage if the indole-OH was methylated. However, we found that 20 was
converted cleanly to the deoxy indole product, identical in all respects to that
of methylation product of nocathiacin II by treatment with TMSCH2N2.
17. Connolly, T. P.; Regueiro-Ren, A.; Leet, J. E.; Springer, D. M.; Goodrich, J.;
Huang, X.; Pucci, M. J.; Clark, J. M.; Bronson, J. J.; Ueda, Y. J. Nat. Prod. 2005,
68, 550.
18. Whitebread, S. E.; Taylor, V.; Bottari, S. P.; Kamber, B.; de Gasparo, M. Biochem.
Biophys. Res. Commun. 1991, 181, 1365.
19. Mice were challenged IM with Staphylococcus aureus ‘COL’ at 2.3 Â 107 cfu/
mouse. Mice (4/group) were treated with test compounds intravenously
(0.2 ml, iv, bid.). Thighs were collected aseptically 24 h after challenge.