Novel pyridopyrimidine derivatives as inhibitors of stable toxin a (STa) induced cGMP synthesis

Article abstract of DOI:10.1016/j.bmcl.2009.04.024

A series of pyridopyrimidine derivatives were synthesized and evaluated for their ability to inhibit cyclic nucleotide synthesis in the presence of stable toxin a of Escherichia coli. The structure activity relationships around the basic core structure we

Full text of DOI:10.1016/j.bmcl.2009.04.024

Bioorganic & Medicinal Chemistry Letters 19 (2009) 3067–3071
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel pyridopyrimidine derivatives as inhibitors of stable
toxin a (STa) induced cGMP synthesis
Eric A. Tanifum a, Alexander Y. Kots b, Byung-Kwon Choi b, Ferid Murad b, Scott R. Gilbertson a,
*
a Chemical Biology Program, Department of Pharmacology and Toxicology, The University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0650, USA
b Institute of Molecular Medicine, University of Texas Health Science Center, 1825 Pressler Street, Houston, TX 77030, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of pyridopyrimidine derivatives were synthesized and evaluated for their ability to inhibit cyclic
nucleotide synthesis in the presence of stable toxin a of Escherichia coli. The structure activity relation-
ships around the basic core structure were examined and examples with better activity and potentially
better pharmacological properties are presented.
Received 29 January 2009
Revised 31 March 2009
Accepted 3 April 2009
Available online 12 April 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
STa
Pyridopyrimidine
Diarrhea
cGMP
Diarrhea is a major health problem throughout the world, with
22% of all deaths of children in sub-Saharan Africa, and 23% in
south Asia, being attributed to diarrheal diseases in 2000.1 It is
generally caused by gastrointestinal infections ranging from bacte-
rial and viral to parasitic. Typical treatment is to give fluids to pre-
vent dehydration and continued feeding while administering drugs
for the underlying cause. The development of drugs that are effec-
tive against the physiological mechanisms that cause the imbal-
ance of fluids in the intestine would be a significant addition in
our therapeutic arsenal. We report our efforts to develop inhibitors
of cyclic nucleotide synthesis caused by stable toxin a of Escherichia
coli (STa).
STa induces diarrhea when it binds to intestinal epithelial cell
membrane receptor, guanylyl cyclase type C (GC-C).2 This activates
the enzyme to convert guanosine triphosphate (GTP) to cyclic gua-
nosine 30,50-monophosphate (cGMP), causing intracellular levels of
cGMP to spike.3–5 This in turn induces activation of a cGMP-depen-
dent protein kinase and chloride-ion channel, cystic fibrosis trans-
membrane conductance regulator (CFTR). Activation of CFTR
triggers the flux of chloride ions into the intestinal lumen and the
accumulation of water and sodium ions, thus causing diarrhea.5
In an effort to develop a novel approach to the treatment of acute
diarrhea based on inhibition of stimulated cyclic nucleotide synthe-
sis, a small library of compounds was screened, from which com-
identified, as a promising lead.6 It was determined that BPIPP (1)
suppresses cyclic nucleotide synthesis in the presence of STa (with
IC50 3.4 1.6 lM at 100 nM STa), and is active in vivo in an intestinal
loop animal model of acute diarrhea.4 However, even though BPIPP
(1) has a CLog P of 4.02,7 within the generally acceptable range for
pharmaceuticals, it was found to have limited solubility. Addition-
ally at this time, its mechanism of inhibition of STa-stimulated cyclic
nucleotide synthesis is unclear. In an effort to address these issues,
we designed, synthesized and tested a series of new analogs of BPIPP
aimed at, improving activity, finding molecules with improved solu-
bility and identifying a location where a fluorescent tag can be at-
tached while maintaining activity. The latter will enable initial
tracking of the binding site of BPIPP in intact cells.
In the initial screen (Table 1),6 it was clear that certain portions
on the molecule are critical for activity, consequently those sites
were not significantly altered in the new derivatives. For instance,
an electronegative atom at the meta position of the phenyl ring at
C-5 appears to be necessary. Absence or replacement of this group
on this phenyl ring with another functional group led to significant
or complete loss of activity (compounds 2 or 5). The indene moiety
was tolerant to minor changes while capping of N-11 (compound
8) resulted in considerable loss of activity. At the time of the initial
study, changes to the pyrimidine moiety were not explored. With
this data, and the above stated goals, a more extensive SAR study
of the scaffold was undertaken.
pound
1,
5-(3-bromophenyl)-1,3-dimethyl-5,11-dihydro-1H-
indeno-[20,10:5,6]pyrido[2,3-d]pyrimidine-2,3,6-trione (BPIPP) was
All compounds in Table 2 were synthesized from commercially
available starting materials using the Hantzsch dihydropyridine
three component cyclization.8–11 Reaction of 6-amino-uracil (11)
derivative with an aldehyde (10) and a 1,3-dicarbonyl compound
* Corresponding author.
E-mail address: srgilber@utmb.edu (S.R. Gilbertson).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.04.024

3068
E. A. Tanifum et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3067–3071
Table 1
Table 1 (continued)
Summary of initial SAR studies on BPIPP5
Compd #
Compound
Inhibition (%)
Me
O
Me
N
Me
O
Me
N
H
N
N
O
O
N
8
22
4
N
Me
1
86
4
Me
O
O
O
Br
Cl
Me
N
H
N
O
(9, 13, 15) provided the desired products in generally good yields. A
frequent byproduct of this reaction is from air oxidation of the de-
sired product. This impurity was greatly reduced by prior degas-
sing of the reaction solution (Scheme 1).
N
2
3
4
7
1
Me
O
O
Compounds were assayed for activity in T84 cells stimulated
with STa as described.6 With the goal of maintaining the electron
deficient nature of the aromatic ring using a group that is smaller
and less hydrophobic than bromide, a series of compounds with
Me
N
H
N
O
N
Table 2
Me
63 16
Inhibition of STa-stimulated cGMP accumulation in T84 cells
O
O
F
Compd #
Yield CLog P
% Inhibition
Me
H
N
11
10
7
N
1
O
2
Me
N
9
N
O
N
4
6
5
8
1
70% 4.0
85
9
Me
3
N
O
O
2
1
Me
O
O
Br
Br
Me
H
N
N
O
Me
N
H
N
O
N
17
85% 3.4
63
6
Me
N
O
F
O
F
Me
5
6
7
26
7
O
O
OH
Me
N
OH
H
N
O
Me
N
N
H
N
Me
O
18
78% 3.4
42 10
O
F
O
F
N
56 11
Me
O
O
F
Br
Me
N
H
N
O
Me
N
H
N
O
N
Me
19
83% 3.3
62
4
O
O
F
N
Me
29
6
O
O
F
F
OMe
OH

E. A. Tanifum et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3067–3071
3069
Table 2 (continued)
Table 2 (continued)
Compd #
Yield CLog P
% Inhibition
Me
N
H
N
O
NH
Me
N
H
N
O
27
28
29
30
31
68% 3.0
86% 2.3
74% 6.6
43% 2.4
56% 3.3
19
8
N
O
O
Me
20
74% 2.7
23
8
O
O
F3C
CF3
CN
Me
N
H
N
O
NH
Me
H
N
N
O
ꢀ6 13
O
O
F
N
21
50% 3.8
73
8
Me
O
O
F3C
CF3
H
N
H
N
Me
N
N
H
N
O
ꢀ2
7
O
S
N
Me
22
23
24
25
26
73% 3.9
83% 3.7
65% 4.7
78% 3.6
82% 2.4
80
93
93
9
6
9
O
O
F
F3C
CF3
Me
CF3
H
N
Me
H2N
N
O
Me
N
N
H
N
8
8
Me
O
O
O
N
O
Me
F3C
CF3
O
F3C
F
Me
H
Me
N
N
O
H
Me
N
N
N
H
N
S
11
7
Me
O
N
O
O
N
O
Me
Br
O
F3C
CF3
Me
N
H
Me
N
O
H
H
N
H
N
N
N
32
54% 4.7
22 14
Me
O
NH
O
O
56 14
O
O
F3C
CF3
F3C
CF3
Me
N
H
N
O
O
N
O
Me
N
33
68% 3.1
2
8
Me
H
N
O
NH
O
1
4
Br
O
O
Br
(continued on next page)

3070
E. A. Tanifum et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3067–3071
Table 2 (continued)
comp 1; IC50=9.6 2.3 µM
comp 23; IC50=3.4 0.5 µM
comp 24; IC50=2.5 0.2 µM
Compd #
Yield CLog P
% Inhibition
60
40
20
0
Me
N
H
N
O
O
N
34
81% 2.5
ꢀ16
8
Me
O
F
O
F
Me
N
H
N
O
O
-1.0
-0.5
0.0
0.5
1.0
1.5
N
Me
35
65% 2.4
ꢀ3 13
Log concentration, µM
O
F
O
Figure 1. Inhibition of cGMP accumulation in T84 cells treated with 1
compounds 1, 23, and 24.
lM STa by
F
F
24 inhibit cGMP accumulation with IC50 values of 3.4 0.5
and 2.5 0.2 M, respectively, compared to 9.6 2.3 M for BPIPD
at the above mentioned STa concentration.
lM
T84 cells were treated for 10 min with 50 lM derivatives and then were either left
l
l
unstimulated (basal cGMP level), or were stimulated with 250 nM STa for 10 min.
cGMP was extracted and assayed by ELISA. Each plate had a vehicle control and the
percentage of inhibition was calculated for each plate. None of the compounds
significantly influenced basal cGMP levels.
As mentioned earlier, changes in the pyrimidine portion of the
lead had not been explored prior to this work. Removal of the
methyl groups at positions 1 and 3 add two hydrogen-bond donor
sites to the molecule and should increase aqueous solubility.
Unfortunately, demethylation of both positions (25), was accompa-
nied by considerable loss in activity. Demethylation at only posi-
tion 3 (26–28), led to derivatives that had diminished stability in
aqueous medium and showed negligible activity. Replacement of
the six-membered pyrimidine with a thiophine containing pyra-
zole ring, (29), led to complete loss of activity. The results from
these changes indicate that methyl substituted pyrimidine moiety
with its substituents, appear to be necessary for the inhibitory
activity of BPIPP.
The extended conjugation of the indene moiety may render the
scaffold a participant in p-stacking or simple hydrophobic interac-
tions. To determine the structure activity relationship for this sec-
tion of the molecule, derivatives where synthesized that replace
the indene section with a cyclohexanone moiety. It was hypothe-
sized that this change with a more flexible construction or remov-
different fluoride substituted benzene rings was synthesized and
examined (17–19). This modification resulted in a moderate
improvement in solubility, but with a decrease in activity. Substi-
tution of the bromide at the meta position with a cyano group (20)
resulted in an almost inactive compound. However, substitution
with a trifluormethyl group (21), provided significant activity, con-
firming our initial observation that substitution of an electron
withdrawing group at this site appears to be important for activity.
Incorporation of fluorine in the ortho position next to the meta tri-
fluoromethyl (22), results in a derivative with activity that is com-
parable to that of BPIPP. When this fluorine or another
trifluoromethyl group is added to the other meta position (23
and 24, respectively), derivatives with activity superior to BPIPP
are generated. The improved activity of these two compounds
was further corroborated by their IC50 values relative to BPIPP,
on T84 cells treated with 1
lM STa. As shown in Figure 1, 23 and
R1
H
O
R1
N
O
N
N
O
H2N
O
AcOH, 110 °C, 8 h
+
Ar
+
H
N
R2
10
N
9
R2
11
O
12
Ar
O
O
O
OR
O
R1
N
R1
N
H
H
O
N
O
H2N
O
O
AcOH, 110 °C, 8 h
Ar
H
+
+
N
O
R2
N
10
R2
14
13
11
O
Ar
O
O
O
OR
R1
N
O
R1
N
O
H
H3C
R3HN
16
N
O
H3C
H2N
O
AcOH, 110 °C, 8 h
+
Ar
H
+
R3HN
O
N
10
R2
N
11
R2
15
O
Ar
O
O
Scheme 1.

E. A. Tanifum et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3067–3071
3071
ing it completely would yield derivatives with better solubility.
Supplementary data
While compounds that have that ring removed (30–32) had better
solubility, they were found to be inactive. Replacement of the in-
dene portion of the molecule with a cyclohexanone ring and a fur-
an ring, (33–35), led to derivatives that appear to insignificantly
stimulate cGMP production in our assay.
To conclude, through SAR studies on BPIPP, we have identified
three new potent inhibitors (22–24) of STa induced cyclic nucleo-
tide synthesis. These lead compounds will be tested to determine
their ability to suppress STa-induced fluid accumulation in an
in vivo rabbit intestinal loop model. This is the same model used
to verify that the original hit (cpd 1) was active in an animal mod-
el.6 Physiology studies on these compounds as well as the possible
mechanism of inhibition are ongoing and will be reported in due
course.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.04.024.
References and notes
1. Morris, S. S.; Black, R. E.; Tomaskovic, L. Int. J. Epidemiol. 2003, 32, 1041.
2. Qadri, F.; Svennerholm, A. M.; Faruque, A. S.; Sack, R. B. Clin. Microbiol. Rev.
2005, 12, 465.
3. Field, M.; Graf, L. H.; Laird, W. J.; Smith, P. L. Proc. Natl. Acad. Sci. U.S.A. 1978, 75,
2800.
4. Hughes, J. M.; Murad, F.; Chang, B.; Guerrant, R. L. Nature 1978, 271, 755.
5. Vaandrager, A. B. Mol. Cell Biochem. 2002, 230, 73.
6. Kots, A. Y.; Choi, B.-K.; Jimenez, M. E.; Warren, C. A.; Gilbertson, S. R.; Guerrant,
R. L.; Murad, F. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 8440.
7. CLog P values were calculated using ACD/ChemSketch software.
8. Hantzsch, A. Chem. Ber. 1881, 14, 1637.
9. Agarwal, A.; Chauhan, P. M. S. Synth. Commun. 2004, 34, 8440.
10. Agarwal, A.; Ashutosh, R.; Goyal, N.; Chauhan, P. M.; Gupta, S. Bioorg. Med.
Chem. 2005, 13, 6678.
11. Manpadi, M.; Uglinskii, P. Y.; Rastogi, S. K.; Cotter, K. M.; Wong, Y.-S. C.;
Anderson, L. A.; Ortega, A. J.; Van Slambrouck, S.; Steelant, W. F. A.; Rogelj, S.;
Tongwa, P.; Antipin, M. Y.; Magedov, I. V.; Kornienko, A. Org. Biomol. Chem.
2007, 5, 3865.
Acknowledgments
This work was supported in part by the Robert A. Welch Foun-
dation, University of Texas, and UT-Houston Health Science Center
Office of Biotechnology.