Design and synthesis of 2-aminothiazole based antimicrobials targeting MRSA

Article abstract of DOI:10.1016/j.bmcl.2012.09.095

Privileged structure-based libraries have been shown to provide high affinity lead compounds for a variety of important biological targets. The present study describes the synthesis and screening of a 2-aminothiazole based compound library to determine th

Full text of DOI:10.1016/j.bmcl.2012.09.095

Bioorganic & Medicinal Chemistry Letters 22 (2012) 7719–7725
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of 2-aminothiazole based antimicrobials targeting MRSA
Sivakumar Annadurai a, Rogelio Martinez a, Daniel J. Canney a, Tess Eidem b, Paul M. Dunman b,
Magid Abou-Gharbia a,
⇑
a Temple University School of Pharmacy, Moulder Center for Drug Discovery Research, Philadelphia, PA, USA
b University of Rochester Medical Center, School of Medicine and Dentistry, Rochester, NY, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Privileged structure-based libraries have been shown to provide high affinity lead compounds for a vari-
ety of important biological targets. The present study describes the synthesis and screening of a 2-ami-
nothiazole based compound library to determine their utility as antimicrobials, focusing on MRSA.
Several of the compounds in this series demonstrated improved antimicrobial activity as compared to
ceftriaxone (CTX), a b-lactam antibiotic. The most potent compound (21) had MICs in the range of 2–
Received 9 August 2012
Revised 20 September 2012
Accepted 25 September 2012
Available online 3 October 2012
4
l
g/ml across a panel of Staphylococcus aureus strains. In addition, trifluoromethoxy substituted amino-
Keywords:
thiazoles and aminobenzothiazoles were found to be potent antimicrobials with MICs of 2–16 g/ml.
l
Privileged structure
2-Aminothiazole
Antimicrobial
MIC
Ó 2012 Elsevier Ltd. All rights reserved.
MRSA
More deaths are caused by methicillin resistant Staphylococcus
aureus (MRSA) strains in the US each year than by HIV/AIDS.1
The emergence of antibacterial resistance due to metallo-b-
lactamases neutralizing carbapenems is an example of the serious
challenges that exist for antibacterial drug discovery. This chal-
lenge is further complicated by the exit of major pharmaceutical
companies from the antibacterial arena leading to a paucity of
new antibiotics to combat resistant bacteria.2 Hence, there is a
need for increased drug discovery efforts in this important area.
One commonly used approach for lead generation is the screening
of commercial libraries and natural product-based libraries. Com-
mercial libraries commonly result in low hit rates due to false pos-
itives, lack of structural diversity and poor physicochemical
properties.3 Natural product-based libraries are often comprised
of derivatives of parent compounds and seldom lead to activity
that is truly distinct from the parent.3 Privileged structure-based li-
braries however, have been shown to be capable of providing high
affinity ligands for diverse biological targets. This approach has
been used successfully for a variety of biological targets, especially
G-protein coupled receptors (GPCRs), which has seen a number of
successful developments over the last decade.4,5
molecules.6–9 The Kyoto Encyclopedia of Genes and Genomes
(KEGG) database reveals the extensive use of the 2-aminothiazole
structural motif in numerous drugs, as well as, preclinical and clin-
ical candidates (Fig. 1). In addition there is literature evidence
showing that 2-aminothiazole-based compounds serve as musca-
rinic and serotonergic ligands.10–12 Herein we report the synthesis
and antimicrobial screening of a 2-aminothiazole-based library of
35 compounds.
Our structurally diverse privileged structure-based library was
synthesized using both commercially available bromo-2-
aminobenzothiazoles (5-bromo and 6-bromosubstituted) and in-
house synthesized bromo substituted 2-aminothiazole scaffolds.
Both microwave13 and thermal14 synthetic approaches were
employed to generate the 2-aminothiazole scaffolds (Scheme 1).
The microwave assisted synthesis of the scaffold involving
a-bromoketones and thiourea or substituted thiourea proceeded
in moderate to high yields. Changing the solvent from ethanol to
methanol for piperidine, morpholine and pyrrolidine containing
a-bromoketones led to a higher conversion to the target amino-
thiazoles. A fused aminothiazole scaffold (26a) was generated from
2-bromocyclohexanone under microwave conditions (refer
Supplementary data). In the absence of a commercially available
Our efforts to develop a novel series of antibacterial agents led
to the preparation and study of privileged structure-based li-
braries. We hypothesized that the 2-aminothiazole scaffold (fused
and non-fused) would serve as a privileged structure due to their
prevalence in antibacterial agents and other biologically active
a-bromoketone, the corresponding substituted acetophenone
was subjected to thermal conditions with iodine and thiourea to
generate the target 2-aminothiazole scaffold. Aminothiazoles
containing methyl ester groups (30) and (33) were synthesized
by methylating the corresponding acetyl benzoic acid starting
materials with trimethylsilyl diazomethane (refer Supplementary
⇑
Corresponding author.
E-mail address: aboumag@temple.edu (M. Abou-Gharbia).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.09.095

7720
S. Annadurai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7719–7725
H
N
N
S
O
S
NH
O
S
N
NH2
N
N
H
Sulfathiazole
antibacterial
CP-146662 (discovery)
5-HT1A agonist, dopamine uptake
O
O
N
H
N
S
S
F3CO
S
N
S
N
N
N
O
S
N
O
H2N
NH2
NH2
NH
O
Riluzole
Treat Amyotropic lateral
sclerosis (ALS)
N
OH
O
Ceftriaxone
antibacterial
(GLT-1 modulator)
O
(GLT-1 modulator)
OH
O
O
HO
H3N+
S
N
N
H
N
N
N
H
H
N
S
N
O
O
O
O
O
Aztreonam
antibacterial
S
O
O
O-
MZ-338 (phase II)
Adenosine A1, Muscarinic M1
and M2 antagonist, AChE inhibitor
Figure 1. Use of 2-aminothiazole (non-fused and fused) scaffold in ligands/drugs pertaining to the biological targets of interest (adapted and modified from KEGG database).
O
X
a
S
X
= 2-bromo (5), 2-fluoro (27),
3-methylester(29),
4-methylester(32)
NHR2
O
O
S
c
R1
N
S
R3
N
H
R1
N
b
(R3 = H or CF3)
(2), (4), (6), (8), (10), (12), (14),
(16), (17), (19), (21), (23). (25), (26a), (27a), (30)
(33)
O
Br
(2a, 4a, 6a, 25a, 26b, 27b)
R1
R1 = 3-bromophenyl (1), 4-bromophenyl (3), 4-trifluoromethoxyphenyl (7), 4-methoxyphenyl (9),
2-benzofuranyl (11), 2-naphthyl (13), 4-morpholinophenyl (15), 4-pyrrolidinophenyl (18),
4-piperidinophenyl (22), adamantan-1yl (24)
Scheme 1. Synthesis of 2-aminothiazole scaffolds and functionalization. Reagents and conditions: (a) iodine (1 equiv) thiourea (2 equiv) heat at 100 °C or reflux in THF (12–
52%); (b) thiourea or 3-fluorophenyl thiourea, ethanol or methanol, MW 50 °C, 10 min (closed) or 30 min (open vessel) (20–96%); (c) benzene sulfonyl chloride or substituted
benzene sulfonyl chloride, pyridine, 0 °C to rt (12–64%).
data) and then heated with iodine and thiourea to provide the
desired aminothiazole scaffold.
oles and bromoamino benzothiazoles were coupled with the corre-
sponding substituted boronic acids in the presence of palladium
catalysts (Schemes 2 and 3). trans-Dichlorobis(triphenyl phos-
phine) palladium(II) provided better reaction yields when com-
pared to Tetrakis (triphenylphosphine)palladium(0) in the
present series.
There are several methods reported in the literature for cou-
pling the amino group of the b-lactams with carboxyl derivatives.
A commonly used method is the activation of the acid by formation
of the Vilsmeier reagent using phosphorus oxychloride (POCl3),
N,N-dimethylformamide and tetrahydrofuran.18 Other methods
include the use of coupling reagents like diethyl chlorophosphate
and 2-chloro-4,6-dimethoxy-1,3,5-triazine.19,20 After unsuccessful
attempts with the Vilsmeier activation, 2-chloro-4,6-dimethoxy-
1,3,5-triazine was used for activation of the carboxyl containing
aminothiazoles. An advantage of using 2-chloro-4,6-dimethoxy-
1,3,5-triazine is the ability to monitor the formation of the
activated ester using LC-MS. Using this approach, b-lactam-based
aminothiazoles were synthesized in four steps (Schemes 4 and
5). In the first step the amine functionality was protected with a
The aminothiazole scaffolds (2), (4), (6), (25), (26), (27), (30) and
(33) were chosen for further functionalization. An aryl sulfonamide
fragment was incorporated at the amino end of the scaffolds (2),
(4), (6), (25), (26a) and (27a) by condensation with either benzene
sulfonyl chloride or substituted benzene sulfonyl chloride in the
presence of base (Scheme 1). The yields of the sulfonamidation
reactions were found to be low to moderate. LC–MS analysis of
the reaction mixture revealed the simultaneous formation of the
bis-sulfonamide product in some cases. Literature reports suggest
the simultaneous formation of bis-sulfonamide during sulfonami-
dation reactions may be responsible for the low yields of monosulf-
onamide product recovered in our hands.15–17 Alkali mediated
recovery of the monosulfonamides has been reported to improve
the final yield of the monosulfonamide product but was not at-
tempted here.
Microwave promoted Suzuki coupling reactions were employed
in order to incorporate carboxyphenyl and trifluoromethoxyphenyl
fragments on the aminothiazole scaffold. The bromoaminothiaz-

S. Annadurai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7719–7725
7721
HO
OH
S
S
B
NH2
NH2
R4
N
N
Br
R4
b
2f, 2g, 2h, 2i, 2j, 4d, 4e, 6b, 6c
a
2b, 2d, 4b
2c, 2e, 4c
(2) - 3-bromo
(4) - 4-bromo
(6) - 2-bromo
Scheme 2. Synthesis of 2-aminothiazole derivatives. Reagents and conditions: (a) substituted boronic acids or boronic pinacol ester, (Ph3P)4Pd or PdCl2(PPh3)2, anhydrous
potassium carbonate or sodium carbonate (2 M), THF:water (1:1), MW 140 °C, 30 min (10–49%); (b) 2 M NaOH (40–76%).
HO
OH
determined to be the lowest compound concentration that inhibits
bacterial growth (turbidity). MIC data for the test compounds is
shown in Tables 1–3. Aminothiazole derivatives possess compara-
ble or superior antimicrobial activity against S. aureus strains when
B
S
N
R4
S
N
R4
Br
NH2
NH2
a
(1a) - 5-bromo
(1d) - 6-bromo
(1b), (1c), (1e)
compared to the standard CTX [8
SA300);16 g/ml (RN8175)].
Compound (21) was found to be the most potent in the series
with MIC values in the range of 2–4 g/ml across all S.aureus
lg/ml(UAMS-1);128 lg/ml(U-
l
Scheme 3. Synthesis of 2-aminobenzothiazole derivatives. Reagents and condi-
tions: (a) substituted boronic acids or boronic pinacol esters, (Ph3P)4Pd or
PdCl2(PPh3)2, sodium carbonate (2 M), THF:water (1:1), MW 140 °C,30 min (32–
49%)
l
strains. Trifluoromethoxyphenyl substituted 2-aminothiazoles
(2j), (4d), (6c) and trifluoromethoxy substituted aminobenzothiaz-
oles (1c) and (1e) showed MIC values in the range of 2–16
against S. aureus strains. Compound (8) containing a 4-trifluoro-
methoxy phenyl group showed moderate MIC value of
64 g/ml against S. aureus (USA 300 and RN8175) compared to
CTX. Replacement of the 4-trifluoromethoxy phenyl group with a
4-methoxy phenyl group resulted in loss of antimicrobial activity
lg/ml
trityl group followed by alkali hydrolysis of the ester to the corre-
sponding acid. The acid was then activated using 2-chloro-4,6-
dimethoxy-1,3,5-triazine and the activated ester was coupled to
the silylated b-lactam fragment (7-aminocephalosporanic acid)
followed by deprotection of the trityl group. The compounds syn-
thesized as described above were evaluated in antimicrobial assays
and MIC values determined.
a
l
(>256 lg/ml). Additional studies will be required to determine
the role of the trifluoromethoxy phenyl fragment in improving
antimicrobial potency of 2-aminothiazoles. In addition to the
promising effects observed with the trifluoromethoxy phenyl frag-
ment, there were other fragments which resulted in MIC values
comparable to CTX. Compound (20), a pyrrolidine based com-
Thirty five compounds were screened for antimicrobial activity
against a panel of methicillin resistant or susceptible microorgan-
isms including S. aureus (USA300, UAMS-1 & RN8175), Escherichia
coli C600 N/puc19 and Acinetobacter baumanii UNMC 8872. MIC
values were determined using the broth microdilution method
according to the Clinical and Laboratory Standards Institute (CLSI)
guidelines using cation adjusted Mueller–Hinton broth. Microtiter
plates containing serial dilutions of compounds (0, 2, 4, 8, 16, 32,
pound, demonstrated an MIC value of 2
(UAMS-1). The MIC for the b-lactam based compound (37) was
determined to be 16 g/ml against S. aureus (RN8175). Sulfona-
lg/ml against S. aureus
l
mides (4a) and (6a) exhibited MIC values comparable to CTX
against S. aureus strains. The MIC value for ethylbenzoate (6b)
was two fold lower than CTX against S. aureus (RN8175). The
64, 128, or 256
units (CFU/ml) and incubated for 24 h at 37 °C. MIC values were
l
g/ml) were inoculated with 105 colony forming
NH2
O
S
O
N
H
N
O
OCH3
S
O
S
O
OH
S
S
O
N
N
O
O
S
O
N
NH2
NHC(C6H5)3
NHC(C6H5)3
O
N
N
N
O
N
a
b
c
H3CO
N
O
O
NH2
O
O
OH
d,
e
(37)
(30)
(34)
(35)
Scheme 4. Synthesis of compound (37). Reagents and conditions: (a) Trityl chloride, pyridine, 40 °C; (b) 2 M NaOH, 50 °C; (c) 2-chloro-4, 6-dimethoxy,-1,3,5-triazine, N, N-
dimethylformamide, 4-methylmorpholine, rt 30 min; (d) N,O-bis(trimethyl silyl) acetamide (2 equiv), tetrahydrofuran, 0 °C, 30 min (silylation of 7-ACA) followed by coupling
at 0– À10 °C overnight; (e) Trifluoroacetic acid, anisole (21%).
NH2
O
S
H
N
O
O
N
S
S
S
O
S
O
N
O
OH
NHC(C6H5)3
NH2
a
O
NHC(C6H5)3
N
N
O
N
O
OH
b
O
N
O
O
d,
e
c
N
H3CO
N
O
S
O
O
(39)
(33)
(38)
(40)
H2N
N
OCH3
Scheme 5. Synthesis of compound (40). Reagents and conditions: (a) Trityl chloride, pyridine, 40 °C; (b) 2 M NaOH, 50 °C; (c) 2-chloro-4, 6-dimethoxy,-1,3,5-triazine, N,N-
dimethylformamide, 4-methylmorpholine, rt 30 min; (d) N, O-bis(trimethyl silyl) acetamide (2 equiv), tetrahydrofuran, 0 °C, 30 min (silylation of 7-ACA) followed by
coupling at 0–À10 °C overnight; (e) trifluoroacetic acid, anisole (48%).

7722
S. Annadurai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7719–7725
Table 1
Minimum inhibitory concentration (MIC) data of 2-aminothiazole derivatives
S
N
H
R2
R1
N
Compd. No.
R1
R2
S.aureus
E.Coli
A.baumanii
UAMS-1
g/ml
USA-300
RN8175
g/ml
C600 N/puc19
UNMC 8872
lg/ml
l
lg/ml
l
l
g/ml
8
H
H
N.T.
N.T.
64
64
>256
N.T.
N.T.
F3CO
H3CO
10
>256
>256
> 256
12
14
H
H
256
256
>256
>256
>256
>256
>256
>256
N.T.
N.T.
O
16
17
H
128
256
>256
>256
256
>256
>256
N.T.
N.T.
N
O
O
F
>256
N
19
20
21
H
256
2
>256
>256
4
>256
>256
4
>256
>256
>256
N.T
N
F
N.T.
N.T.
N
F
Br
2
O
S
S
O
O
25a
27b
128
N.T.
>256
128
>256
32
>256
N.T.
N.T.
CF3
F
O
> 256
O
30
33
H
H
N.T.
N.T.
>256
>256
>256
128
>256
>256
N.T.
N.T.
O
O
O
O
OH
2c
H
N.T.
>256
>256
>256
>256
O
HO
2e
2f
H
H
N.T.
>256
>256
>256
>256
>256
>256
>256
N.T.
O
O
>256

S. Annadurai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7719–7725
7723
Table 1 (continued)
Compd. No.
R1
R2
S.aureus
E.Coli
A.baumanii
UNMC 8872
g/ml
UAMS-1
g/ml
USA-300
RN8175
g/ml
C600 N/puc19
l
lg/ml
l
l
g/ml
l
O
O
O
2g
2h
H
H
N.T.
N.T.
>256
>256
>256
256
>256
>256
N.T.
N.T.
O
O
O
O
O
2i
H
N.T.
>256
256
>256
N.T.
OCF3
2j
H
N.T.
N.T.
8
4
>256
>256
N.T.
O
S
O
4a
64
16
>256
Br
O
4b
4c
H
H
N.T.
N.T.
>256
N.T.
>256
N.T.
>256
N.T.
>256
N.T.
O
O
OH
4d
4f
H
N.T
>256
N.T.
128
2
>256
N.T.
N.T.
N.T.
N.T.
F3CO
O
S
S
O
O
N.T.
N.T.
N.T.
32
Br
O
6a
>256
O
O
6b
6c
H
H
N.T.
N.T.
256
16
32
16
>256
>256
>256
N.T.
OCF3
O
37
H
256
256
16
>256
N.T.
HN
S
N
O
O
O
O
OH
(continued on next page)

7724
S. Annadurai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7719–7725
Table 1 (continued)
Compd. No.
R1
R2
S.aureus
E.Coli
A.baumanii
UAMS-1
g/ml
USA-300
RN8175
g/ml
C600 N/puc19
UNMC 8872
lg/ml
l
lg/ml
l
l
g/ml
O
HN
O
40
H
N.T.
N.T.
N.T.
N.T.
N.T.
S
N
O
O
O
OH
Table 2
Minimum inhibitory concentration (MIC) data of fused 2-aminothiazole
S
N
H
N
R2
Compd. No.
R2
S. aureus
E. coli
C600 N/puc19
g/ml
A. baumanii
UAMS-1
g/ml
USA-300
RN8175
g/ml
UNMC 8872
l
l
g/ml
l
l
lg/ml
CF3
O
S
26b
>256
>256
>256
>256
N.T
O
Table 3
Minimum inhibitory concentration of 2-aminobenzothiaozles
R1
R2
N
NHR3
S
Compd. No.
R1
R2
R3
S. aureus
E. coli
C600 N/puc19
g/ml
A. baumanii
UAMS-1
USA-300
RN8175
UNMC 8872
l
g/ml
lg/ml
lg/ml
l
lg/ml
O
O
1b
1c
H
H
H
>256
N.T.
>256
>256
>256
>256
N.T.
N.T.
OCF3
H
H
8
8
OCF3
1e
H
N.T.
8
16
8
>256
2
N.T.
8
CTX
128
16
CTX-ceftriaxone.
N.T.—not tested.
preliminary antimicrobial data presented herein supports the priv-
ileged nature of 2-aminothiazole and 2-aminobenzothiazole scaf-
folds. Incorporation of suitable peripheral fragments could be a
useful approach to generate promising antimicrobial leads espe-
cially against methicillin resistant S. aureus strains.
focused on the aminothiazole (fused and non-fused) ring system.
The work provides useful synthetic methods for the preparation
of aminothiazole-based ligands that can be used to prepare novel
compounds for screening in antibacterial assays as well as for other
important biological targets. Biological evaluation of the present li-
brary has identified promising antimicrobial leads against methi-
cillin resistant S. aureus strains. Efforts are currently underway to
In conclusion, the present work describes the synthesis and
antimicrobial screening of a privileged structure-based library

S. Annadurai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7719–7725
7725
4. Guo, T.; Hobbs, D. W. Assay Drug Dev. Technol. 2003, 1, 579.
5. Meng, T.; Wang, J.; Peng, H.; Fang, G.; Li, M.; Xiong, B.; Xie, X.; Zhang, Y.; Wang,
X.; Shen, J. Eur. J. Med. Chem. 2010, 45, 1133.
6. Vukovic, N.; Sukdolak, S.; Solujic, S.; Milosevic, T. Arch. Pharm. (Weinheim, Ger.)
2008, 341, 491.
expand the library and screen the compounds for antibacterial
activity and to assess their utility in additional biologically relevant
targets.
7. Palkar, M.; Noolvi, M.; Sankangoud, R.; Maddi, V.; Gadad, A.; Nargund, L. V.
Arch. Pharm. (Weinheim, Ger.) 2010, 343, 353.
Acknowledgement
8. Saeed, A.; Rafique, H.; Hameed, A.; Rasheed, S. Pharm. Chem. J. 2008, 42, 191.
9. Hashiguchi, T.; Yoshida, T.; Itoyama, T.; Taniguchi, Y. U.S. Patent 5,856,347,
1999.
10. Lebel, L. A.; Nowakowski, J. T.; Macor, J. E.; Fox, C. B.; Kenneth Koe, B. Drug Dev.
Res. 1994, 33, 413.
The author S.A. would like to acknowledge Office of the Dean,
School of Pharmacy for continued financial support.
11. Sabb, L. A. U.S. Patent 5,712,270, 1996.
12. Sagara, Y.; Kimura, T.; Fujikawa, T.; Noguchi, K.; Ohtake, N. Bioorg. Med. Chem.
Lett. 2003, 13, 57.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.
09.095.
13. Kabalka, G. W.; Mereddy, A. R. Tetrahedron Lett. 2006, 47, 5171.
14. Siddiqui, H. L.; Iqbal, A.; Ahmad, S.; Weaver, W. Molecules (Basel, Switzerland)
2006, 11, 206.
15. Greenfield, A.; Grosanu, C. Tetrahedron Lett. 2008, 49, 6300.
16. Parlow, J. J.; Stevens, A. M.; Stegeman, R. A.; Stallings, W. C.; Kurumbail, R. G.;
South, M. S. J. Med. Chem. 2003, 46, 4297.
17. Su, B.; Landini, S.; Davis, D. D.; Brueggemeier, R. W. J. Med. Chem. 2007, 50,
1635.
18. Jarrahpour, A.; Zarei, M. Tetrahedron Lett. 2007, 48, 8712.
19. Lee, H.-W.; Kang, T. W.; Cha, K. H.; Kim, E.-N.; Choi, N.-H.; Kim, J.-W.; Hong, C.,
II. Synth. Commun. 1998, 28, 1339.
References and notes
1. Olson, P. D.; Kuechenmeister, L. J.; Anderson, K. L.;Daily, S.; Beenken, K. E.;
Roux, C. M.; Reniere, M. L.; Lewis, T. L.; Weiss, W. J.; Pulse, M.; Nguyen, P.;
Simecka, J. W.; Morrison, J. M.; Sayood, K.; Asojo, O. A.; Smeltzer, M. S.;
Skaar, E. P.; Dunman, P. M. PLoS Pathog, 2011, 7, e1001287.
2. Jones, D. Nat. Rev. Drug Disc. 2010, 9, 751–752.
3. Welsch, M. E.; Snyder, S. A.; Stockwell, B. R. Curr. Opin. Chem. Biol. 2010, 14, 347.
20. Lee, H.-W.; Kang, T. W.; Cha, K. H.; Kim, E.-N.; Choi, N.-H.; Kim, J.-W.; Ii, H. C.
Synth. Commun. 1998, 28, 35.