Further evaluation of the tropane analogs of haloperidol

Article abstract of DOI:10.1016/j.bmcl.2014.07.018

Previous work from our labs has indicated that a tropane analog of haloperidol with potent D₂ binding but designed to avoid the formation of MPP⁺-like metabolites, such as 4-(4-chlorophenyl)-1-(4- (4-fluorophenyl)-4-oxobutyl)pyridin-

Full text of DOI:10.1016/j.bmcl.2014.07.018

Bioorganic & Medicinal Chemistry Letters 24 (2014) 4294–4297
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Further evaluation of the tropane analogs of haloperidol
Dinithia Sampson a, Barbara Bricker a, Xue Y. Zhu a, Kwakye Peprah a, Nazarius S. Lamango a,
Vincent Setola b, Bryan L. Roth b, Seth Y. Ablordeppey a,
⇑
a Division of Basic Pharmaceutical Sciences, Florida A&M University, College of Pharmacy and Pharmaceutical Sciences, Tallahassee, FL 32307, USA
b Department of Pharmacology, Medicinal Chemistry and Psychiatry, University of North Carolina at Chapel Hill, School of Medicine, NC 27599, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Previous work from our labs has indicated that a tropane analog of haloperidol with potent D2 binding
but designed to avoid the formation of MPP+-like metabolites, such as 4-(4-chlorophenyl)-1-(4-(4-fluo-
rophenyl)-4-oxobutyl)pyridin-1-ium (BCPP+) still produced catalepsy, suggesting a strong role for the
D2 receptor in the production of catalepsy in rats, and hence EPS in humans. This study tested the hypoth-
esis that further modifications of the tropane analog to produce compounds with less potent binding to
the D2 receptor than haloperidol, would produce less catalepsy. These tests have now revealed that while
haloperidol produced maximum catalepsy, these compounds produced moderate to low levels of cata-
lepsy. Compound 9, with the least binding affinity to the D2R, produced the least catalepsy and highest
Minimum Adverse Effective Dose (MAED) of the analogs tested regardless of their affinities at other
receptors including the 5-HT1AR. These observations support the hypothesis that moderation of the D2
binding of the tropane analogs could reduce catalepsy potential in rats and consequently EPS in man.
Published by Elsevier Ltd.
Received 1 May 2014
Revised 5 July 2014
Accepted 8 July 2014
Available online 14 July 2014
Keywords:
Haloperidol
Tropane analogs
Antipsychotic agents
Catalepsy
Pyridinium metabolite
Haloperidol is an effective antipsychotic medication that is used
to treat schizophrenia; however, it is associated with producing
short and long term extrapyramidal side effects (EPS), that is,
Parkinsonism-like symptoms. These adverse effects may be due
to its potent binding at the D2 receptor and/or its metabolism into
quaternary pyridinium species, such as BCPP+ and RHPP+1–4
(Fig. 1).
Previous studies in our laboratories have shown that modifica-
tions in the butyrophenone and the piperidine moieties of
haloperidol significantly altered its binding affinities at the D2,
D4, and 5-HT1A receptors.1,5,4 Modifications of the piperidine ring
resulting in D2 receptor binding profiles that are similar or better
than haloperidol and which could not form toxic pyridinium
metabolites, were also identified.4,6,7 These studies have indicated
that adding an ethylene bridge to the piperidine moiety to form a
tropane analog, maintained or increased affinity for the D2 receptor
subtype. It was also observed that replacement of the carbonyl
functional group in haloperidol with an oxygen or a sulfur atom
retains D2 binding and increases affinity at the 5-HT1A receptor,
indicating that the carbonyl group is not essential for binding at
the D2R but can be replaced with other groups to produce com-
pounds that bind with high affinity to the receptors associated
with antipsychotic properties.8
While the tropane analog of haloperidol (5) appears to have a
better binding profile than haloperidol, in vivo studies in our hands
have confirmed that its similar potent antagonism at the D2R sub-
type resulted in very significant catalepsy in rats.5,9 Thus, it was of
interest in this study, to further test the hypothesis that modera-
tion of the binding affinity at the D2 receptor could attenuate cat-
alepsy in the tropane analogs of haloperidol. To test this
hypothesis, we modified the fluorobutyrophenone moiety while
replacing the piperidinol with 3-tropanol to obtain compounds
2–10 as shown in Figure 2 and then evaluated their binding affin-
ities at receptors associated with antipsychotic activity, that is, D1
and D2-like receptors, 5-HT1AR, 5-HT2AR and 5-HT7R (Table 1).
Subsequently, binding affinities at off-target receptors were also
evaluated and reported in Table 2.
The syntheses of compounds 2-7 were previously reported.8–10
The alkylating agent, (3-chloropropyl)(4-fluorophenyl)sulfane, was
prepared using the procedure in our previously published paper,8
with slight modification, by refluxing 4-fluorobenzenethiol and
1-bromo-3-chloropropane in the presence of K2CO3 in iPrOH. The
other two alkylating agents were similarly prepared by using the
corresponding n-fluorobenzenethiol. The target compounds, 8–10
were obtained by alkylating 3-(4-chlorophenyl)-8-azabicy-
clo[3.2.1]octan-3-ol, in the presence of K2CO3, KI and DME with
the appropriate alkylating agent as shown in Scheme 1.
The first modifications to moderate the D2R binding involve
replacement of the carbonyl group in haloperidol and the results
⇑
Corresponding author.
E-mail address: seth.ablordeppey@famu.edu (S.Y. Ablordeppey).
http://dx.doi.org/10.1016/j.bmcl.2014.07.018
0960-894X/Published by Elsevier Ltd.

D. Sampson et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4294–4297
4295
O
F
C
(CH2)3
N
Cl
BCPP+ (HPP+)
RHPP+
O
C
OH
F
(CH2)3
N
Cl
OH
F
C
(CH2)3
N
Cl
H
Figure 1. Quaternary pyridinium metabolites of haloperidol.
receptors remained about the same except for the deoxygenated
compound (2) which had a 27-fold decrease at the D2 receptor
(Ki = 24 nM, haloperidol Ki = 0.89 nM). The next compounds (5–8),
the tropane analogs of haloperidol and compounds 2–4, were also
screened at the same receptors and the results are also presented
in Table 1. The methylene (6) and the oxygen (7) analogs had sim-
ilar binding across all 7 receptors of interest (Table 1); however,
their affinities were significantly lower than those of the previously
reported carbonyl analog (5). The binding affinities of compound 8,
the sulfur analog of 6 and 7, turned out to be generally more potent
at the primary receptors of interest, that is, D2, 5-HT1A, 5-HT2A and
5-HT7 receptors. Thus, compound 8 was selected for further
investigation.
O
C
N
HO
Cl
OH
F
N
(CH2)3
Cl
N
A
B
Cl
Haloperidol (1)
HO
S
Y
F
X
(CH2)3
F
X
(CH2)3
B
A
(CH2)3
B
2. X = CH2
3. X = O
4. X = S
5. X = CO
8. Y = 4-F
6.
7.
X = CH2
X = O
9
. Y = 3-F
10. Y = 2-F
Figure 2. Structures of compounds evaluated.
are reported in Table 1. Deoxygenation (2), replacement with
oxygen (3) or sulfur (4) resulted in significant improvements in
binding to the 5-HT1A, D1 and D4 receptors. Affinity however, was
decreased at the 5-HT7R while binding to the D2 and 5-HT2A
Having been aware that replacement of the fluorine atom on the
butyrophenone moiety with other substituents resulted in
significant changes in binding affinities,8 we synthesized the other
Table 1
Binding affinities expressed as mean Ki SEM (nM) of the analogs at CNS receptors22
Compound
Binding affinities (Ki SEM) in nM at relevant CNS receptors
5-HT1A
5-HT2A
5-HT7
D1
D2
D3
D4
D5
Haldol (1)
3600
120
ND
120
0.89
2.5
10
ND
2
3
658 50
317 29
74.0 6.0
27.7 8.0
102 0.7
302 23
72 7
93 10
30.9 6.0
1042 45
656 29
1044 52
ND
130 11.0
22.0 1.0
27.0 2.0
ND
>10K
>10K
24 2.0
8.8 1.0
5.3 0.3
0.31
33 0.26
22 0.17
2.9 0.31
27 0.36
4.8 0.04
20
668 39
49.0 1.7
182 22
0.71
9.7 0.1
4.10 0.03
0.70 0.00
11.0 0.1
1.3 0.01
45
1.5 0.1
4.3 0.2
2.4 0.1
12.0
9.4 0.1
49.0 0.5
8.9 0.04
28.0 0.3
27.0 0.2
50
414 34
103
113
ND
5
9
4
5a
6
7
8
9
317
236
2
1
472
420
5
4
>10K
>10K
MT
MT
MT
123
2
20.0 0.2
13.0 0.1
8.2 0.1
2100
89.0 1.4
97.0 1.0
917 11
347
MT
5
509
115
3.3
7
1
10
882
NA
9
4093 68
52
Olanzapineb
NA
ND = not determined; NA = not available; MT = missed assay threshold of 50% inhibition.
a
Previously reported in Ref. 9 except at 5-HT1AR and 5-HT2AR.
Data from Ref. 12.
b
Table 2
Binding affinities expressed as mean Ki SEM (nM) for the compounds at off-target receptors22
Compound
Binding affinities (Ki SEM) in nM
5-HT2B
5-HT2C
H1
M1–M3
DAT
NET
SERT
1800
1867 146
2228 187
160 24
ND
414
255
316
227
Haldol (1)
2
3
4
5
6
7
8
NA
4700
MT
>10K
MT
872 178
2434 44
MT
1498 26
1861 20
1122 15
440
ND
MT
MT
MT
ND
MT
MT
MT
MT
MT
ND
MT
>10K
497 28
ND
778 10
620
1081 11
285 4.38
354 11
5500
723 65
1622 134
790 68
ND
1381 23
898
128.5 17.5
698.6 91.7
115.0 9.7
8780 1625
1719 30
4149 14
977 16
1225 87
3959 471
2249 239
ND
MT
1791 28
1409 19
6
2
4
3
9
1057 14
9
10
270
520
2
5
1207 16
1009 10
510
462
5
5
102 0.6
ND = not determined; MT = missed assay threshold of 50% inhibition.
Ki values without the associated SEM, are within 20% of the mean value.
F
F
F
i
ii
SH
Br
S
S
Cl
B
+
(CH2)2CH2 Cl
(CH2)2CH2
(CH2)2CH2
Scheme 1. Reagents and conditions: (i) iPrOH, K2CO3, reflux 12 h; (ii) 3-(4-chlorophenyl)-8-azabicyclo[3.2.1]octan-3-ol (B), DME, KI, K2CO3, 100 °C, reflux, 12 h.

4296
D. Sampson et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4294–4297
Catalepsy CLP Test
Catalepsy Bar Test
A
B
C
100
90
80
70
60
50
40
30
20
10
0
30
20
10
0
30
****
****
****
***
20
*
*
10
0
VEH 0.193 0.578 0.964 1.35 1.74
VEH 0.193 0.578 0.964 1.35 1.74
mg/kg CMPD 8
0.01
0.1
1
10
mg/kg CMPD 8
mg/kg CMPD 8
Catalepsy CLP Test
D
F
Catalepsy Bar Test
E
100
30
20
10
0
90
80
70
60
50
40
30
20
10
0
30
20
****
***
*
*
10
0
0.1
1
10
VEH 3.17 6.34 9.50 12.7 15.8
mg/kg CMPD 9
VEH 3.17 6.34 9.50 12.7 15.8
mg/kg CMPD 9
mg/kg CMPD 9
G
Catalepsy CLP Test
100
90
80
70
60
50
40
30
20
10
0
Catalepsy Bar Test
I
H
****
****
****
30
20
10
0
****
30
20
10
0
****
***
*
0.001
0.01
0.1
1
VEH 0.01 0.12 0.24 0.48 1.2 1.8
mg/kg Haloperidol
VEH 0.01 0.12 0.24 0.48 1.2 1.8
mg/kg Haloperidol
mg/kg Haloperidol
Figure 3. Inhibition of APO-induced climbing behavior in Swiss-Webster mice, n = 5/dose, by compound 8, compound 9, and haloperidol in graph 1A, 1D, and 1G,
respectively. Graphs B, C, E, F, and I display mean time in seconds on the bar or in the CLP stance for Sprague-Dawley rats, n = 5. Graphs 1B and 1C are compound 8. Graphs 2E
and 2F are compound 9. Graphs 2H and 2I are haloperidol. Significant difference between vehicle and dose of compound indicated by (⁄p <0.05), (⁄⁄⁄p <0.001), and
(
⁄⁄⁄⁄p <0.0001).
two fluorinated analogs with the fluorine atom located at the meta
(9) and the ortho (10) positions on the butyrophenone moiety. The
analogs were subsequently screened and the results are recorded
in Table 1. Compounds 8, 9 and 10 bind to the 5-HT1AR with similar
affinities (Ki = 20.0, 13.0 and 8.2 nM, respectively) as expected. At
the 5-HT2AR, the binding Ki of the para- (8) and ortho- (10) substi-
tuted analogs were similar, and 4–6 times more potent that of the
meta-substituted analog (9). Apart from compound 8 which binds
with a Ki of 97.0 nM at the 5-HT7R, both 9 and 10 have about 9-fold
weaker affinity. All three analogs, 8–10, have potent binding at the
D2-like receptors although the meta-substituted analog, 9, dis-
played an 8-fold lower affinity (Ki = 27.0 nM) at the D2R subtype.
This D2 binding affinity turned out to be similar to that of the liter-
ature reported12 value for olanzapine (Ki = 20.0 nM). Spurred on by
this observation, (albeit superficial since the binding studies were
reported from different laboratories) and the potent binding to the
5-HT1AR, purportedly associated with attenuating catalepsy in ani-
mal models,11 we selected compounds 9 and 8 (for comparison) for
evaluation in animal models predictive of antipsychotic efficacy
and EPS side effects.
results reported in Table 2, indicate that all the compounds have
relatively low affinities at these receptors and hence are expected
to have low potentials to induce weight gain, sedation and other
potential side effects associated with these receptors.13
To test for a compound’s ability to display antipsychotic proper-
ties in humans, the inhibition of apomorphine (APO)-induced
climbing behavior in mice is evaluated.14,15 Thus, in this study,
the reversal of APO-induced climbing was assessed in Swiss Web-
ster mice given 1.5 mg/kg of APO in a manner consistent with pre-
vious reports.1,5 Doses were expressed as the free base for
haloperidol and apomorphine, and as the HCl salt for compounds
8 and 9 and were given in a volume of 10 mL/kg for the highest
dose given by intraperitoneal (ip) injection, except for APO, which
was given by subcutaneous (sc) injection. Five mice were each
dosed, at various concentrations with compounds 8 or 9, then
30 min later with APO (1.5 mg/kg). Each mouse was subsequently
placed in a cylindrical wire cage to assess the climbing behavior.
Compound 8, the para-fluoro substituted analog had significant
inhibition of the APO-climbing behavior, with an ED50 value of
0.165 mg/kg (R2 = 0.9511) compared to its meta-substituted coun-
terpart, 9 (ED50 = 3.14 mg/kg; R2 = 0.9132). These results are con-
sistent with the fact that the para analog is more potent than the
meta analog and possesses APO-induced climbing reversal proper-
ties similar to haloperidol (ED50 = 0.122 mg/kg) as expected, while
9 is less potent at the D2 receptor and displayed an equally lower
The compounds were also evaluated for off-target interactions
at receptors that may produce side effects such as 5-HT2B, (valvular
heart disease) 5-HT2C, (induction of weight gain), H1, (induction of
weight gain and/sedation) M1–M3, (cognitive symptoms) DAT, NET
and SERT which might interfere with in vivo observations. The

D. Sampson et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4294–4297
4297
Table 3
In vivo data on haloperidol, compounds 8 and 9
Compound
APO-reversal ED50
mole/kg
Bar test/catalepsy MAED
CLP/catalepsy MAED
mg/kg
l
mg/kg
l
mole/kg
mg/kg
lmole/kg
Haloperidol
Compound 8
Compound 9
0.122
0.165
3.14
0.325
0.373
7.10
0.240
0.578
12.7
0.640
1.31
28.7
0.480
0.964
12.7
1.28
2.18
28.7
MAED = Minimum Adverse Effective Dose.
potency in the in vivo model. The results are reported in Figure 3
and Table 3.
Acknowledgements
The potential for compounds 8 and 9 to induce catalepsy in rats
and hence extrapyramidal side effects in human beings16,11 was
also assessed using the catalepsy bar test as previously
reported.17,18 A secondary evaluation was also made using the
crossed-legs position (CLP) test, which is reported to be more sen-
sitive to the anticataleptic actions of 5-HT1A receptor agonists.19,11-
Five rats at each dose point, for 8 (ED50-9 ED50) or 9 (ED50-5 ED50),
were injected ip and tested in the CLP test for 30 s followed by the
bar test for 30 s. A righting test20,21 was performed immediately
after, to assess the potential of the compounds to induce sedation.
This was repeated at 3 and 6 min after the start of the first trial. The
mean of the three trials was used. The results are reported in Fig-
ure 3 and Table 3.
Compound 8 induced catalepsy in both the bar and CLP tests.
One-way ANOVA with Bonferroni post tests with p <0.05 indicated
that there was a significant difference between vehicle (1% lactic
acid) and compound 8. The Minimum Adverse Effective Dose
(MAED) for the bar test was observed at 0.578 mg/kg or 3.5 times
its ED50 value for reversal of APO-induced climbing behavior. Since
all animals ‘righted’ themselves after the catalepsy tests, catalepsy
displayed by compound 8 was not associated with a sedative effect
by the compound, which is consistent with its poor binding at the
5-HT2C and H1 receptors (Ki = 1500 and 977 nM, respectively;
Table 2). Compound 9 also demonstrated catalepsy with a MAED
of 12.7 mg/kg or 4.0 times the ED50 of its reversal of APO-induced
climbing behavior. Given that the MAED of haloperidol was only
2.0 times its ED50 value of reversal of the APO-induced climbing
behavior, the observed trend supports the hypothesis that potent
binding at the D2 receptor is associated with the potential of a
compound to induce catalepsy and that its moderation would
result in attenuation of catalepsy induction.
In conclusion, initial modifications of haloperidol have pro-
duced analogs with antipsychotic potential as demonstrated by
inhibition of apomorphine-induced climbing behavior. Catalepsy
testing revealed that a tropane analog with potent binding to the
D2 receptor, still had potent catalepsy, suggesting that binding to
the D2R plays a key role in EPS in humans. Further modifications
of the tropane analog produced compounds with relatively lower
binding affinities to the D2 receptor compared to haloperidol.
While haloperidol produced maximum catalepsy, these com-
pounds produced moderate to low levels of catalepsy. Compound
9, with the least binding to the D2R, produced the least catalepsy
and the highest MAED of the analogs tested even though the Ki val-
ues for the 5-HT1AR were similar and moderately potent. These
observations are consistent with the hypothesis that further mod-
eration of the binding affinity of the tropane analogs at the D2R,
would attenuate their catalepsy potential in rats.
The authors acknowledge financial support through NIH Grant
# 5SC1GM088451-01-04 and a Title III Grant to Florida A&M
University. The work was also supported in part by the RCMI Grant
# 5 G12 MD007582-30 and the Pharmaceutical Research Center
NIH/NCRR Grant # 1C06-RR12512-01.
References and notes
1. Ablordeppey, S. Y.; Lyles-Eggleston, M.; Bricker, B.; Zhang, W.; Zhu, X.;
Goodman, C.; Roth, B. L. Bioorg. Med. Chem. Lett. 2006, 16, 3219.
2. Eyles, D. W.; McLennan, H. R.; Jones, A.; McGrath, J. J.; Stedman, T. J.; Pond, S. M.
Clin. Pharmacol. Ther. 1994, 56, 512.
3. Igarashi, K.; Kasuya, F.; Fukui, M.; Usuki, E.; Castagnoli, N., Jr. Life Sci. 1995, 57,
2439.
4. Lyles-Eggleston, M.; Altundas, R.; Xia, J.; Sikazwe, D. M.; Fan, P.; Yang, Q.; Li, S.;
Zhang, W.; Zhu, X.; Schmidt, A. W.; Vanase-Frawley, M.; Shrihkande, A.;
Villalobos, A.; Borne, R. F.; Ablordeppey, S. Y. J. Med. Chem. 2004, 47, 497.
5. Ablordeppey, S. Y.; Altundas, R.; Bricker, B.; Zhu, X. Y.; Kumar, E. V.; Jackson, T.;
Khan, A.; Roth, B. L. Bioorg. Med. Chem. 2008, 16, 7291.
6. Sikazwe, D. M.; Li, S.; Mardenborough, L.; Cody, V.; Roth, B. L.; Ablordeppey, S.
Y. Bioorg. Med. Chem. Lett. 2004, 14, 5739.
7. Sikazwe, D. M.; Nkansah, N. T.; Altundas, R.; Zhu, X. Y.; Setola, V.; Roth, B. L.;
Ablordeppey, S. Y. Bioorg. Med. Chem. 2009, 17, 1716.
8. Peprah, K.; Zhu, X. Y.; Eyunni, S. V.; Setola, V.; Roth, B. L.; Ablordeppey, S. Y.
Bioorg. Med. Chem. 2012, 20, 1291.
9. Sikazwe, D. M.; Li, S.; Lyles-Eggleston, M.; Ablordeppey, S. Y. Bioorg. Med. Chem.
Lett. 2003, 13, 3779.
10. Peprah, K.; Zhu, X. Y.; Eyunni, S. V.; Etukala, J. R.; Setola, V.; Roth, B. L.;
Ablordeppey, S. Y. Bioorg. Med. Chem. 2012, 20, 1671–1678.
11. Kleven, M. S.; Barret-Grevoz, C.; Slot, L. B.; Newman-Tancredi, A.
Neuropharmacology 2005, 49, 135.
12. Schmidt, A. W.; Lebel, L. A.; Howard, H. R., Jr.; Zorn, S. H. Eur. J. Pharmacol. 2001,
425, 197.
13. Coccurello, R.; Moles, A. Pharmacol. Ther. 2010, 127, 210.
14. Needham, P. L.; Atkinson, J.; Skill, M. J.; Heal, D. J. Psychopharmacol. Bull. 1996,
32, 123.
15. Protais, P.; Costentin, J.; Schwartz, J. C. Psychopharmacology 1976, 50, 1.
16. Hoffman, D. C.; Donovan, H. Psychopharmacology 1995, 120, 128.
17. Bricker, B.; Jackson, T.; Boateng, B.; Zhu, X. Y.; Ablordeppey, S. Y. Pharmacol.
Biochem. Behav. 2012, 102, 294.
18. Bricker, B.; Sampson, D.; Ablordeppey, S. Y. Pharmacol. Biochem. Behav. 2014,
120, 109.
19. Depoortere, R.; Bardin, L.; Auclair, A. L.; Kleven, M. S.; Prinssen, E.; Colpaert, F.;
Vacher, B.; Newman-Tancredi, A. Br. J. Pharmacol. 2007, 151, 253.
20. McCreary, A. C.; Glennon, J. C.; Ashby, C. R., Jr.; Meltzer, H. Y.; Li, Z.; Reinders, J.
H.; Hesselink, M. B.; Long, S. K.; Herremans, A. H.; Stuivenberg, H. V.; Feenstra,
R. W.; Kruse, C. G. Neuropsychopharmacology 2007, 32, 78.
21. Reeve, B.; Dingwall, B.; Darlington, C. L.; Scott, S. J.; Sansom, A. J.; Smith, P. F.
Pharmacol. Biochem. Behav. 1992, 42, 183.
22. Binding affinities reported in Tables 1 and 2 were conducted by the National
Institute of Mental Health Psychoactive Drug Screening Program (NIMH-PDSP).
Details of the methods and radioligands used for the binding assays were
previously reported in Shapiro, D. A.; Renock, S.; Arrington, E.; Chiodo, L. A.;
Liu, L. X.; Sibley, D. R.; Roth, B. L.; Mailman, R. Neuropsychopharmacology 2003,
28, 1400. In the initial screening assays, test compounds were tested at a
concentration of 10 mM in quadruplicate at selected GPCRs. For molecular
targets at which >50% inhibition was measured, Ki determinations were
obtained using at least six concentrations of the test compound; Ki values were
calculated in quadruplicate using GraphPad Prism.