IKCa-channel blockers. Part 2: Discovery of cyclohexadienes

Article abstract of DOI:10.1016/j.bmcl.2004.10.063

Novel cyclohexadienes have been identified as potent and specific IK _Ca-channel blockers. In this communication we describe their synthesis as well as their chemical and biological properties. A selected derivative distributes well into brain and reduces the infarct volume, intracranial pressure as well as the water content in a rat subdural hematoma model of traumatic brain injury after iv administration. Novel cyclohexadienes have been identified as potent and specific IK_Ca-channel blockers. In this communication we describe their synthesis as well as their chemical and biological properties. A selected derivative is being enriched in rat brain and reduces the infarct volume, intracranial pressure as well as the water content in a rat subdural hematoma model of traumatic brain injury after iv administration.

Full text of DOI:10.1016/j.bmcl.2004.10.063

Bioorganic & Medicinal Chemistry Letters 15 (2005) 401–404
IKCa-channel blockers. Part 2: Discovery of cyclohexadienes
a
a
b
Klaus Urbahns,a, Siegfried Goldmann, Jochen Kruger, Ervin Horvath,
*
´
¨
Joachim Schuhmacher,c Rolf Grosser,a Volker Hinzb and FrankMauler
b
aInstitute of Medicinal Chemistry, Pharma Research Center, Bayer Health Care, D-42096 Wuppertal, FRG
bInstitute of CNS Research, Pharma Research Center, Bayer Health Care, D-42096 Wuppertal, FRG
cInstitute of Pharmakokinetics, Pharma Research Center, Bayer Health Care, D-42096 Wuppertal, FRG
Received 23 July 2004; revised 15 October 2004; accepted 21 October 2004
Available online 18 November 2004
Abstract—Novel cyclohexadienes have been identified as potent and specific IKCa-channel blockers. In this communication we
describe their synthesis as well as their chemical and biological properties. A selected derivative is being enriched in rat brain
and reduces the infarct volume, intracranial pressure as well as the water content in a rat subdural hematoma model of traumatic
brain injury after iv administration.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The intermediate conductance Ca2+-activated potassium
ion channel (IKCa) is involved in inflammatory cell acti-
vation and abundantly expressed in human brain.1 In an
earlier communication we described a novel 4-phenyl
pyran dicarboxylic ester lead structure (A) derived from
the dihydropyridine Nifedipine as nanomolar IKCa
inhibitors. These compounds resulted from exchanging
the dihyropyridine NH group with an isoelectronic oxy-
gen atom and reduced the infarct volume in a rat sub-
dural hematoma model of traumatic brain injury
without cardiovascular side effects (Scheme 1).2
Whereas the NH/O exchange furnished potent and spe-
cific IKCa blockers, we were wondering whether a corre-
sponding NH/CH2 exchange would give similar results.
In this report we describe the synthesis and pharmaco-
logical properties of a novel class of cyclohexadienes (B).
In order to get synthetic access to cyclohexadienes 1–4
and rac-5–7, we used a Carba-Hantzsch reaction re-
cently developed in our laboratories.3 This reaction uses
phosphora I and provides cyclohexadienes as mixtures
of two double bond isomers each of which can be
isolated and/or interconverted (cf. Scheme 2).
The synthesis of N-methyl amino-substituted cyclohexa-
dienes rac-8–11 was accomplished starting from the
known carbocycles III as a mixture of diastereomers
and subsequent functional group interconversion. The
final dehydration step yields the corresponding cyclo-
hexadienes as mixtures of 3,6- (i.e., rac-8–11) and 3,5-
double bond isomers, which were separated and
isolated.4
O2N
MeOOC
COOMe
R4
R2
R1
A: X=O
B: X=CH2
N
H
X
Scheme 1. Design of cyclohexadienes as IKCa-channel blockers
starting from the lead structure Nifedipine. Isoelectronic replacement
of NifedipineÕs NH fragment leads to pyrans (A) and cyclohexadienes
(B).
The bicyclic derivative rac-16 was obtained in a two step
sequence in close analogy to the Carba-Hantzsch proto-
col. First I was reacted with VII in a Michael type reac-
tion using NaHMDS in toluene at À78ꢁC. For this
transformation toluene was the solvent of choice to sup-
press side reactions such as Diels–Alder dimerization of
VII. Subsequently, NaOMe/MeOH was added to affect
formation of the cyclohexadiene core and lactone ring
closure. Using these optimized conditions, an alternative
Keywords: IKCa channel; Potassium channel; Cyclohexadiene; Brain
edema; Water content; Cerebral infarction; Subdural hematoma; Rat.
*
Corresponding author at present address: AstraZeneca, R&D, Lund,
Sweden. Tel.: +46 46 33 78 25; fax: +46 46 33 71 19; e-mail:
klaus.urbahns@astrazeneca.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.063

402
K. Urbahns et al. / Bioorg. Med. Chem. Lett. 15 (2005) 401–404
MeOOC
fects the reactivity of the aminocrotonic ester such that
R3
R3
O
a)
it turns the adjacent methyl group into a better nucleo-
phile than nitrogen, leading to carbocycles rather than
dihydropyridines. The resulting carbocyclic enamines
were hydrolyzed under concomitant cyclization of the
lactone ring to yield rac-12–13. Subsequent treatment
with N-methyl ammonium formate furnished rac-14–
15.5
R4
MeOOC
R4
+
PPh3
1-4: R4=COOMe
rac-5: R4=COOtBu
rac-6: R4=COOH
rac-7: R4=CONH2
I
II
b)
c)
R3
R3
d,e)
R4
R2
R4
By and large, the SAR of cyclohexadienes follows trends
similar to those observed in the 4-phenyl pyran series.2
Direct comparison of 4 (1.5nM) and the corresponding
isolectronic pyran (24nM) indicates however the superi-
ority of cyclohexadienes in terms of potency. Whereas
the cyclohexyl and pyridine derivatives were inactive
(1, 2), electron withdrawing substituents in meta posi-
tion clearly enhanced potency (3).
R2
HO
N
H
OH
rac-8-11
III
COOEt
OAc
R3
O
f)
R2
R2
R1
R3
O
O
NHnBu
O
rac-12-13: R1=OH
g)
IV
V
VI
rac-14-15: R1=NHMe
The most potent aromatic substitution pattern was
the 4-Cl, 3-CF3 motif, already identified in the pyran
series. The presence and nature of the ester side chain
was also important for biological activity. In contrast
to its methyl counterpart, the bulky ester rac-5 was
inactive as well as the carboxylic acid rac-6. Interest-
ingly, the corresponding amide rac-7 showed retained
activity. The corresponding N-methyl amino-substi-
tuted cyclohexadienes rac-8–11 have also been investi-
gated and showed potencies comparable to their
symmetric counterparts. The methyl ester rac-11 was
more potent than its ethyl counterpart rac-10, again
confirming the importance of a small ester side chain
(Table 1).
R3
COOEt
OAc
MeOOC
R3
h)
O
MeOOC
O
O
PPh3
I
VII
rac-16
Scheme 2. Synthesis of cyclohexadienes. (a) 3equiv NaOMe, MeOH,
12h, rt, 49–98% (mixture of double bond isomers, ratios: see3); (b) 5%
CF3COOH, CH2Cl2, 2h, rt, 67%; (c) SOCl2, 2h, reflux, evaporate, then
THF/aq NH3 (1+2), 0ꢁC, 40% (+19% biphenyl amide); (d) MeOH/
MeNH2 (11N), cat. p-Tos-OH, 2h, reflux, 70–80%; (e) cat. p-Tos-OH,
toluene, reflux, 2h, 7–33%; (f) EtOH, reflux, 6h, then: 10% aq HCl
(33%), reflux, 2h, 48–66%; (g) 10equiv N-methyl ammonium formate,
EtOH, reflux, 85–90%; (h) 1equiv NaHMDS, toluene, À78ꢁC to rt,
3days, then: 3equiv NaOMe, MeOH, 0ꢁC to rt, 1h, 21%.
The corresponding 3,5-cyclohexadienes, which were iso-
lated as side products, have also been investigated,
but did not show activity below 1000nM (data not
shown).
3,5-double bond isomer could not be detected. The over-
all yield for this process is 21%.
Although we considered the high potency of 4 as clearly
promising, we realized that this class of monocylic cyclo-
hexadienes still shows a remaining trend to 3,6/3,5
double bond isomerization, prohibitive for further
development.
A coupling reaction of N-butyl substituted aminocroton
esters IV with benzylidene-type Michael acceptors was
used to synthesize hetero-substituted bicyclic cyclohexa-
dienes. In this case the large butyl amine substituent af-
Table 1. IKCa-channel inhibition of cyclohexadiene derivatives 1–11
R3
R4
R2
R1
2
1
Compound
R1
R2
R3
R4
IC50 (nM)
1
Me
Me
COOMe
COOMe
COOMe
COOMe
COO-t-Bu
COOH
C6H11
3-Pyridyl
3,5-Cl2Ph
COOMe
COOMe
COOMe
COOMe
COOMe
COOMe
COOMe
COOMe
COOMe
COOEt
>1000
>1000
100
2
3
Me
Me
4
4-Cl, 3-CF3Ph
4-Cl, 3-CF3Ph
4-Cl, 3-CF3Ph
4-Cl, 3-CF3Ph
4-CF3Ph
1.5
rac-5
rac-6
rac-7
rac-8
rac-9
rac-10
rac-11
Me
Me
>1000
>1000
30
Me
CONH2
COOMe
COOMe
COOEt
NHMe
NHMe
NHMe
NHMe
200
20
4-NO2Ph
3-NO2Ph
3-NO2Ph
70
25
COOMe
COOMe
Given is the mean of the IC50 (inhibition constant) of at least two experiments each performed in triplicates.

K. Urbahns et al. / Bioorg. Med. Chem. Lett. 15 (2005) 401–404
403
Table 2. IKCa-channel inhibition of bicyclic hexadiene lactones rac-
12–16
(A)
(B)
120
(C)
R3
%
O
R2
R1
100
80
60
40
20
0
-20%
*
O
-14%
-43%
-27%
*
-24%
*
-45%
*
Compound R1
R2
R3
IC50 [nM]
-46%
*
*
rac-12
HO
HO
COOEt
COOEt
2-ClPh
4-Cl, 3-CF3Ph
200
40
-60%
*
rac-13
rac-14
rac-15
rac-16
NHMe COOMe 2,3-Cl2Ph
NHMe COOMe 3-NO2Ph
400
200
8
Me
COOMe 4-Cl, 3-CF3Ph
Given is the mean of the IC50 (inhibition constant) of at least two
experiments each performed in triplicates.
0
1
3
10 30
0
3
10
0
40 400
rac-16 (µg/kg i.v.)
The bicyclic cyclohexadiene lactone series had the clear
advantage to show a lower tendency of isomerization,
possibly due to ring strain imposed by the lactone
moiety on the cyclohexadiene core. Comparing the
potency of rac-11/rac-15 and 4/rac-16 revealed that
bicyclic lactones are generally by one order of magni-
tude less potent than their monocyclic counterparts
(Table 2).
Figure 2. A and B: Efficacy of rac-16 against developing brain edema
in a 24h rat SDH model as shown by (A) intracranial pressure and (B)
brain water content. rac-16 was administered as a continuous iv
infusion for 4h directly after induction of SDH. Values are given as
percent change compared to controls (*P < .05, n = 8–12 per dose
group). C: Neuroprotective efficacy of rac-16 in a rat SDH model,
administered immediately after induction of SDH. rac-16 was admin-
istered as a continuous iv infusion for 4h. Infarct volumes were
determined 7days after SDH. Infarct volumes were calculated as
percentage of infarct volumes of the control group, which was set to
100%. Values above bars indicate the percent infarct volume reduction
compared to controls (*P < .05, n = 8–12 per dose group).
We identified rac-16 as a potent and stable cyclohexa-
diene and therefore selected this compound for further
pharmacological investigations.6–8 After iv administra-
tion to rats, rac-16 showed a more than 10-fold enrich-
ment in brain tissue, allowing for meaningful
investigations of rac-16 in CNS-related animal models
(Fig. 1). In a rat animal model of subdural hematoma
(SDH), rac-16 significantly and dose-dependently re-
duced intracranial pressure (ICP). Concomitantly, the
agent also reduced the content of water in brain tissue,
as determined 24h after SDH. In a separate experiment,
rac-16 also reduced the infarct volume as determined
7d post SDH, indicating a sustained therapeutic effect
(Fig. 2).
firms a potential role of IKCa blockade for the treatment
of traumatic brain injury.
References and notes
1. (a) Logsdon, N. J.; Kang, J.; Togo, J. A.; Christian, E. P.;
Aiyar, J. J. Biol. Chem. 1997, 272, 32723; (b) Joiner, W. J.;
Wang, L. Y.; Tang, M. D.; Kaczmarek, L. K. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 11013; (c) Gardos, G. Biochim.
Biophys. Acta 1958, 30, 147–158; (d) Schwab, A. Am. J.
Physiol. Renal Physiol. 2001, 280, F739; (e) Schwab, A.
Curr. Res. Ion Channel Mod. 1998, 3, 126; (f) Khanna, R.;
Roy, L.; Zhu, X.; Schlichter, L. C. Am. J. Physiol. Cell.
Physiol. 2001, 280, C796; (g) Chandy, K. G.; Wulff, H.;
Beeton, C.; Pennington, M.; Gutman, G. A.; Cahalan, M.
D. Trends Pharm. Sci. 2004, 25(5), 280.
In conclusion, we have demonstrated the development
of novel cyclohexadienes as potent low molecular weight
IKCa-channel blockers starting from the dihydropyri-
dine Nifedipine as a weaklead. The in vivo activity of
rac-16, as a novel, pyran-unrelated structural class con-
2. (a) Urbahns, K.; Horvath, E.; Stasch, J.-P.; Mauler, F.
Bioorg. Med. Chem. Lett. 2003, 13, 2637; (b) Ellory, J. C.;
Culiford, S. J.; Smith, P. A.; Wolowyk, M.; Knaus, E. E.
Br. J. Pharmacol. 1994, 111, 903.
3. (a) Urbahns, K.; Kieck, D.; Goehrt, A. Synthesis 1998,
1807; (b) Urbahns, K.; Mauler, F.; DE 19612645, 1997;
Chem. Abstr. 1997, 127, 307213.
4. Urbahns, K.; Heine, H.-G.; Junge, B.; Schohe-Loop, R.;
Wollweber, H.; Sommermeyer, H.; Glaser, T.; Wittka, R.;
De Vry, J. M. V. EP 698597, 1996; Chem. Abstr. 1996, 124,
342867.
5. Goldmann, S.; Schramm, M.; Thomas, G.; Gross, R. DE
3416293, 1985; Chem. Abstr. 1986, 105, 42631.
1000
c [µg/l]
100
10
1
6. Synthesis of rac-16: In an argon atmosphere at À78ꢁC a
solution of NaHMDS (22.0mL, 22.0mmol, 1M solution in
THF) was added to a suspension of VII (10.0g, 22.0mmol)
in 250mL of dry toluene and it was stirred for 60min at
À78ꢁC. Subsequently, a solution of VIII (12.5g, 32.9mmol)
in 250mL toluene was added, the reaction mixture was
allowed to reach room temperature over night and was
0.1
t [h]
0
1
Figure 1. Blood (Á Á Á) and brain (—) concentrations of rac-16 after iv
administration of 0.2mg/kg to rats (5% ethanol/ 5% Solutol/ 90%
saline).

404
K. Urbahns et al. / Bioorg. Med. Chem. Lett. 15 (2005) 401–404
maintained at this temperature for 4days. At 0ꢁC a
A; 4–6min: 90% A, 0.5mL/min, 40ꢁC): Rf = 3.34min, peak
area (UV detection at 210nm) = 100%, [M+H] = 387.
Elemental anal. for C18H14ClF3O4 (386.75). Calcd C,
55.90; H, 3.65; Cl, 9.17; O, 16.55. Found C, 55.82; H,
3.50; Cl, 9.23; O, 16.82. Solubilities of rac-16: (solvent)
14mg/L (buffer, pH6.5), 12mg/L (3mM Taurocholat), and
70mg/L (15mM Taurocholat).
solution of NaOMe in methanol (2.5mL of a 25% solution,
11.0mmol) was added and the mixture turned deep red
instantly. After 35min another portion of NaOMe (2.51mL
of a 25% solution, 11.0mmol) was added, the cooling bath
was removed, and it was stirred at room temperature for
85min. A saturated aqueous solution of NH4Cl was added
(250mL) followed by four time extraction with ethyl acetate
(250mL each). The combined organic layers were dried
(Na2SO4) and the solvents were distilled off in vacuo. The
crude product was purified using a C18-HPLC column
(Kromasil 100, eluent: acetonitrile/water 15:85–90:10) to
yield 1.78g (21% yield) of rac-16. 1H NMR (300MHz,
DMSO-d6): d = 2.10 (s, 3H), 3.21 (m, 1H), 3.52 (s, 3H), 3.62
(m, 1H), 4.70 (m, 1H), 4.96 (m, 2H), 7.43 (dd, 1H), 7.60 (m,
2H). HPLC–MS (parameters: column symmetry, 50 ·
2.1mm, C18, 3.5lm; eluent A: MeCN + 0.1% HCOOH;
B: 0.3g 30% aq HCl in 1L H2O; 0–4min: from 10% to 90%
7. Neuroprotection model: (a) Mauler, F.; Mittendorf, J.;
´
Horvath, E.; De Vry, J. J. Pharmacol. Exp. Ther. 2002, 302,
359; Brain edema model: (b) Mauler, F.; Hinz, V.;
´
Augstein, K.-H.; Faßbender, M.; Horvath, E. Brain Res.
2003, 989, 99.
8. rac-16 has been separated into its enantiomers (Chiralpak
AD-H; 250 · 20mm; 5lm; UV-detection (254nm); Eluent:
i-hexane/i-propanol 10:1 (v/v); 15mL/min). The enantio-
mers were tested in the primary assay and the eudismic
ratio is 10. The absolute configuration of the eutomer has
not been determined.