4-Aminopyridine derivatives with antiamnesic activity

Article abstract of DOI:10.1016/S0223-5234(00)00103-3

Acetylcholine (Ach) enhancement, useful in the treatment of Alzheimer's disease (AD), may be obtained by means of ion channel modulators such as 4- aminopyridine (4-AP). 4-AP is also the central ring of tacrine, the first drug approved for the treatment of AD. The synthesis and pharmacological activity of three 4-AP derivatives, prepared with the aim of improving their antiamnesic activity, is here described. In two of these compounds 4-AP is connected to 4-aminobutyric acid (GABA), whereas in the third it is connected to 2-indolinone, i.e., the skeleton of linopirdine, another Ach enhancing agent. The new compounds showed potent antiamnesic activity in comparison with piracetam. (C) 2000 Edition scientifiques et medicales Elsevier SAS.

Full text of DOI:10.1016/S0223-5234(00)00103-3

Eur. J. Med. Chem. 35 (2000) 77−82
© 2000 Édition scientifiques et médicales Elsevier SAS. All rights reserved
S0223523400001033/FLA
77
Original article
4-Aminopyridine derivatives with antiamnesic activity#
Aldo Andreania*, Alberto Leonia, Alessandra Locatellia, Rita Morigia, Mirella Rambaldia,
Claudio Pietrab, Gino Villettib
aDipartimento di Scienze Farmaceutiche, Universita’ di Bologna, Via Belmeloro 6, 40126 Bologna, Italy
bChiesi Farmaceutici S.p.A., Via Palermo 26/A, 43100 Parma, Italy
Received 25 May 1999; revised 25 August 1999; accepted 7 September 1999
Abstract – Acetylcholine (Ach) enhancement, useful in the treatment of Alzheimer’s disease (AD), may be obtained by means of ion channel
modulators such as 4-aminopyridine (4-AP). 4-AP is also the central ring of tacrine, the first drug approved for the treatment of AD. The
synthesis and pharmacological activity of three 4-AP derivatives, prepared with the aim of improving their antiamnesic activity, is here
described. In two of these compounds 4-AP is connected to 4-aminobutyric acid (GABA), whereas in the third it is connected to 2-indolinone,
i.e., the skeleton of linopirdine, another Ach enhancing agent. The new compounds showed potent antiamnesic activity in comparison with
piracetam. © 2000 Édition scientifiques et médicales Elsevier SAS
4-aminopyridine / 4-aminobutyric acid / 2-indolinone / tacrine / linopirdine / antiamnesic activity / Alzheimer’s disease
1. Introduction
patients, 4-AP, an A-type K+ channel blocker, elicited
inconsistent and unremarkable effects [3, 4].
The improvement of cholinergic transmission is a
rational and well documented approach to the treatment
of Alzheimer’s disease (AD), which is considered the
most common form of dementia. This target may be
achieved with acetylcholine (Ach) precursors, muscarinic
agonists or acetylcholinesterase inhibitors. The enhance-
ment of Ach release leads to an increase in neuronal
activity, thus restoring central cholinergic tone and im-
proving attention and cognition. Besides acetylcholinest-
erase inhibitors, Ach enhancement may be obtained by
modulating both ligand and voltage-gated (Ca++ and K+)
ion channels. Agents with this latter activity, such as the
potassium channel modulators 4-aminopyridine (4-AP)
[1] and linopirdine [2] (figure 1) have been clinically
evaluated as Ach release enhancing agents. In AD
On the other hand, tacrine (figure 1), the first drug
approved by the FDA for the symptomatic treatment to
mild cognitive disturbances associated with AD, may be
considered a 4-AP derivative.
The disappointing clinical results of linopirdine, whose
enhancement of Ach release correlates with its ability to
block M-type K+ channels, are probably related to its
suboptimal pharmacokinetic profile.
On the basis of these considerations we designed and
synthesized two 4-AP derivatives as potential antiamne-
sic agents. For the first one we planned to bind 4-AP to a
molecule which could improve the diffusion into the
brain and for this purpose we chose 4-aminobutyric acid
(GABA). For the second one we selected 2-indolinone as
the supporting moiety, since it is the skeleton of linopir-
dine. These two molecules and one precursor were
subjected to an antiamnesic test (CO2-induced amnesia in
mice) in comparison to piracetam, the compound selected
as a positive reference, whose antiamnesic activity in
rodent species is well documented [5].
*Correspondence and reprints: aldoandr@alma.unibo.it
#Presented in part at the Italian-Hungarian-Polish Joint Meeting
on Medicinal Chemistry, Giardini Naxos-Taormina, September
28th–October 1st 1999.

78
Figure 1. Structure of 4-aminopyridine, linopirdine and tacrine.
2. Chemistry
chloric acid. Even with repeated attempts to optimize
every step, the overall yield of the aldehyde 12 was very
low. For this reason we studied another approach de-
scribed in figure 4. We prepared dimethyl-[2-(4-nitro-1-
oxypyridin-3-yl)vinyl]amine 14 by reaction of the acti-
vated methyl group of 3-methyl-4-nitropyridine 1-oxide 5
with N,N-dimethylformamide dimethylacetal. The enam-
ine 14 was subjected to oxidation with NaIO4 [10] to
produce 4-nitro-1-oxypyridine-3-carbaldehyde 15 (fig-
ure 4). After hydrogenation of the nitro compound we
obtained the expected aminoaldehyde 12 in good yield.
Compounds bearing 4-AP and GABA moieties (2–3,
figure 2) were prepared from GABA, protected according
to the literature method [6]. The reaction between 4-tert-
butoxycarbonylamino-butyric acid
1 and 4-amino-
pyridine in the presence of ethyl chloroformate gave
[3-(pyridin-4-ylcarbamoyl)-propyl]carbamic acid tert-
butyl ester 2, which was easily deprotected under mild
conditions to afford 4-amino-N-pyridin-4-yl-butyramide
3. For the synthesis of 3-(4-aminopyridin-3-yl-
methylene)1,3-dihydroindol-2-one 13, 4-aminopyridine-
3-carbaldehyde 12 was the starting material (figure 3). It
is reported in the literature at least twice [7, 8], starting
from 4-aminonicotinic acid hydrazide 11 which, in turn,
may be prepared from 3-picoline-N-oxide 4 [7, 9]. These
methods are very similar, we checked both of them and
we modified the procedure according to figure 3. The
most relevant difference from the reported methods is the
oxidation of N-(3-methylpyridin-4-yl)acetamide 7: we
did not obtain directly the expected 4-aminonicotinic acid
9, as reported, but its acetyl derivative 8 which could be
converted into 9 by treatment with concentrated hydro-
The reaction of 12 with 2-indolinone in the presence of
piperidine gave only one isomer of 13 which was
subjected to a series of NOE experiments. In the first one,
the NH2 group was irradiated (6.43 ppm) and NOE was
observed at 6.66 ppm (py-5) and 7.48 ppm (CH). When
py-2 was irradiated (8.35 ppm) the effect was evident at
7.33 ppm (ind-4) and a final irradiation at 7.33 ppm
(ind-4) gave NOE at 6.85 ppm (ind-5) and 8.35 ppm
(py-2). On the basis of these results we may conclude that
compound 13 in our hands belongs to the E configuration.
Figure 2. Synthesis of 4-amino-N-pyridin-4-yl-butyramide hydrochloride.

79
Figure 3. Synthesis of 3-(4-aminopyridin-3-ylmethylene)1,3-dihydroindol-2-one.
3. Pharmacological results and discussion
All 4-AP derivatives, compared to piracetam, showed
an antiamnesic effect, but with different potencies. A
significant (P < 0.01 vs. controls) activity was observed
from 0.4, 107.4 and 42.1 µmol/kg p.o. after administra-
tion of 2, 3 and 13, respectively. Therefore, 2 was almost
The effects of the synthesized compounds in antago-
nizing the CO2-induced amnesia in mice are reported in
table I. For comparison purposes, LD50 values and thera-
peutic indexes are also reported.
Figure 4. Alternative synthesis of the intermediate 4-aminopyridine-3-carbaldehyde.

80
Table I. Antiamnesic activity of compounds 2, 3 and 13.
Compound
Passive avoidance test
LD50 (µmol/kg p.o.)
Therapeutic index*
Dose (µmol/kg p.o.)
Latency (% of control)
2
0.4
1.2
4.0
10.7
35.8
107.4
12.6
42.1
126.4
2 110
+ 841**
+ 3 748**
+ 2 823**
+ 234
87
3 150
73.10
29.33
–
3
+ 350
+ 2 386**
+ 114
13
> 4 215
> 7 034
+ 1 503**
+ 2 517**
+ 1 723**
Piracetam
–
*: Ratio between LD50 and the dose inducing the maximal antiamnesic effect. **: P < 0.01 in comparison with the amnesic control group.
100–300-fold more potent than other compounds. The
efficacy of the antiamnesic effects shown by 3 and 13 at
their maximal dose was similar.
of this type need to be performed in order to clarify
whether compounds 2, 3 and 13 also have a similar
activity.
In acute toxicity studies, at doses higher than those
endowed with antiamnesic activity, treatment of mice
with 2 and 3 resulted in a typical behaviour of hyperex-
citability followed by agitation, facial clonus, stretching
and tonic convulsions followed by death. By contrast, the
administration of 13 at 4 215 µmol/kg determined, in
only 1 out of 5 mice, the appearance of the following
behaviour: sedation, hypoactivity, palpebral closure,
ataxia and dyspnea followed by death. Therefore, 13 was
distinctly less toxic than other compounds. Due to this
property it was impossible to calculate the LD50 value for
this compound and, consequently, an appropriate thera-
peutic index.
The reference compound (piracetam at 2 110 µmol/kg)
exhibited a significant antiamnesic effect, but the latency
increase was lower than maximal values obtained after
the administration of compounds 2, 3 and 13. Piracetam
did not show a behavioural profile of hypolocomotion or
ataxia. Also for this compound endowed with low toxic-
ity, it was impossible to calculate an appropriate thera-
peutic index.
In conclusion, the new synthesized compounds showed
a significant antiamnesic activity when compared to the
reference drug piracetam. According to the pharmacology
of the 4-aminopyridine derivatives, the activity of the
compounds could also be due to an increase of the
cerebral Ach, the primary neurotransmitter involved in
cognitive processes. In fact, derivatives of the K+ channel
blocker linopirdine were reported to be active in increas-
ing the rat brain Ach levels after oral administration when
tested in microdialysis experiments [11]. Further studies
4. Experimental protocols
4.1. Chemistry
The melting points are uncorrected. Analyses (C, H, N)
were within ± 0.4% of the theoretical values. TLC was
performed on Bakerflex plates (Silica gel IB2-F): the
eluent was a mixture of petroleum ether/acetone in
various proportions. Kieselgel 60 (Merck) was used for
column chromatography. The IR spectra were recorded in
nujol on a Perkin-Elmer 683. The 1H-NMR spectra were
recorded in d6-DMSO on a Varian Gemini (300 MHz),
and were referenced to solvent signals. Chemical shifts
are expressed in δ and J in Hz (py = pyridine, ind =
indole). The mass spectra were recorded on a VG 7070E.
4.1.1. Synthesis of [3-(pyridin-4-ylcarbamoyl)-
propyl] carbamic acid tert-butyl ester 2
A mixture of 4-tert-butoxycarbonylamino-butyric acid
1 (1.5 g, 7.4 mmol) in THF (50 mL) and Et3N (1 mL,
7.4 mmol) was cooled to –10 °C. Ice-cooled ClCOOEt
(0.7 mL, 7.4 mmol) was added and stirring continued for
20 min at –10 °C. A solution of 4-aminopyridine (0.9 g,
9.6 mmol) and THF (25 mL) was then added and the
heterogeneous mixture was allowed to gradually warm to
room temperature and was stirred overnight. The crude
product, obtained by evaporation under reduced pressure,
was purified by column chromatography. Eluent: acetone/
petroleum ether 1/1. A white solid was obtained (2.0 g,
yield 97%). C14H21N3O3 (279.3), m.p. 145–149 °C. IR:

81
1
3 300–3 100, 1 660, 1 585, 1 290, 1 160 cm–1. H-NMR:
was evaporated under reduced pressure. Chromatography
of the residue on a silica gel column using methanol as
the eluent led, after evaporation of the appropriate frac-
tion, to the isolation of a white solid (0.3 g, yield 59%)
m.p. 108–110 °C: lit. 113–114 °C [7], 110–112 °C [8].
1.38 (9H, s, t-Bu), 1.70 (2H, m, CH2–CH2–CH2), 2.36
(2H, m, CH2–CO), 2.98 (2H, m, CH2–N), 6.82 (1H, t,
NH–CH2), 7.57 (2H, d, py, J = 5), 8.41 (2H, d, py, J = 5),
10.29 (1H, s, CONH).
4.1.2. Synthesis of 4-amino-N-pyridin-
4-yl-butyramide hydrochloride 3
4.1.6. Synthesis of 3-(4-aminopyridin-3-
ylmethylene)1,3-dihydroindol-2-one 13
A solution of compound 2 (0.5 g, 1.8 mmol) in aceto-
nitrile (20 mL) and HCl 3 N (5 mL) was stirred at room
temperature for 36 h. The suspension was filtered and the
precipitate crystallized (methanol/acetonitrile) to obtain a
white solid (0.25 g, yield 63%). C9H14ClN3O (215.7),
m.p. 267–270 °C dec. IR: 3 500–3 000, 1 700, 1 600,
1 510, 1 175 cm–1. 1H-NMR: 1.92 (2H, m, CH2–
CH2–CH2), 2.69 (2H, m, CH2CO), 2.83 (2H, m, CH2–N),
A mixture of 2-indolinone (0.1 g, 0.75 mmol), com-
pound 12 (0.092 g, 0.7 mmol), methanol (20 mL) and
piperidine (0.1 mL) was refluxed for 5 h. After cooling at
room temperature the precipitate thus formed was re-
moved by filtration to obtain a yellow solid (0.1 g, yield
55%). C14H11N3O (237.2), m.p. 229–231 °C dec. IR:
3 500–2 500, 1 685, 1 600, 810, 720 cm–1. 1H-NMR:
6.43 (2H, s, NH2), 6.66 (1H, d, py-5, J = 5.8), 6.85 (1H,
t, ind-5, J = 7), 6.87 (1H, d, ind-7, J = 7), 7.21 (1H, t,
ind-6, J = 7), 7.33 (1H, d, ind-4, J = 7), 7.48 (1H, s, CH),
8.04 (1H, d, py-6, J = 5.8), 8.35 (1H, s, py-2), 10.57 (1H,
s, NH).
+
8.15 (3H, m, NH3 ), 8.18 (2H, d, py, J = 6.9), 8.70 (2H,
d, py, J = 6.9), 12.15 (1H, s, NH).
4.1.3. Synthesis of 4-acetylamino-nicotinic acid 8
To a suspension of N-(3-methylpyridin-4-yl)acetamide
7 (10 g, 66.6 mmol) in water (200 mL) was added
KMnO4 (15 g, 94.9 mmol). The addition was made in
small portions at 0 °C. The suspension was stirred for 2 h
at room temperature, heated for 5 h at 70 °C and then
filtered. The precipitate was washed with hot water. The
solution was concentrated to a small volume (100 mL)
and treated at 0 °C with HCl 12 N until pH 2 was reached.
After 10 min the suspension was filtered to obtain a white
solid (2.5 g, yield 20%). C8H8N2O3 (180.2), m.p.
247–250 °C dec. IR: 3 470, 1 700, 1 645, 1 365,
4.1.7. Synthesis of dimethyl-[2-(4-nitro-
1-oxypyridin-3-yl)vinyl]amine 14
A solution of 3-methyl-4-nitropyridine-1-oxide 5 (6 g,
38.9 mmol), dry DMF (30 mL) and N,N-dimethyl-
formamide dimethylacetal (4 mL, 46.7 mmol), was
warmed to 150 °C for 3 h under nitrogen. After cooling at
room temperature the suspension was diluted with water
(20 mL). A bright brown solid was obtained by filtration
(4 g, yield 49%). C9H11N3O3 (209.2), m.p. 207–210 °C
dec. IR: 1 585, 1 530, 1 220, 1 090, 1 060 cm–1. 1H-
NMR: 2.97 (6H, s, CH3), 5.84 (1H, d, CH, J = 13), 7.60
(1H, dd, py-6, J = 1.8, J = 7.2), 7.85 (1H, d, CH, J = 13),
7.89 (1H, d, py-5, J = 7.2), 8.67 (1H, d, py-2, J = 1.8).
MS: 209 (M+, 85), 86 (66), 53 (67), 42 (100).
1
1 210 cm–1. H-NMR: 2.19 (3H, s, CH3), 8.46 (1H, d,
py-5), 8.58 (1H, m, py-6), 9.00 (1H, s, py-2), 11.92 (1H,
broad s, NH). MS: 180 (M+, 62), 138 (82), 120 (100), 43
(84).
4.1.4. Synthesis of 4-aminonicotinic acid 9
4.1.8. Synthesis of 4-nitro-1-oxypyridine-
3-carbaldehyde 15
A suspension of compound 8 (3.7 g, 20.5 mmol) in
water (60 mL) was treated with HCl 37% (3 mL) and
heated to reflux for 6 h. After cooling, the solution was
concentrated to a small volume (30 mL) and NH4OH
(30%) was added at 0 °C until pH 5–6 was reached. After
30 min the white solid obtained was filtered (1.8 g, yield
63%, m.p. 330–333 °C dec.: lit. [7] m.p. 335–336 °C
dec.).
Dimethyl-[2-(4-nitro-1-oxypyridin-3-yl)vinyl]amine 14
(4 g, 19.1 mmol) and NaIO4 (12 g, 56.1 mmol) were
mechanically stirred in 50% aqueous THF (200 mL) at
room temperature until the reaction was judged complete
by TLC. The insolubles were removed and the filtrate was
extracted with ethyl acetate. The organic layer was dried
over Na2SO4 and concentrated. The crude product was
purified by column chromatography (eluent: acetone/
petroleum ether 2/8). A yellow solid was obtained (2 g,
yield 62%). C6H4N2O4 (168.1), m.p. 154–155 °C dec. IR:
4.1.5. Synthesis of 4-aminopyridine-3-carbaldehyde 12
Methanol (40 mL) was mixed with 10% palladium on
charcoal (0.03 g) and 4-nitro-1-oxypyridine-3-carbalde-
hyde 15 (0.7 g, 4.16 mmol) was added. The mixture was
hydrogenated at atmospheric pressure and room tempera-
ture. After the calculated amount of hydrogen was taken
up, the catalyst was removed by filtration and methanol
1 690, 1 505, 1 290, 1 245, 1 080 cm–1. H-NMR: 8.24
(1H, d, py-5, J = 7.1), 8.45 (1H, d, py-2, J = 2), 8.58 (1H,
dd, py-6, J = 2, J = 7.1), 10.26 (1H, s, CHO). MS: 168
(M+, 75), 121 (7), 110 (34), 63 (100).
1

82
4.2. Pharmacology
performed. During this trial mice received a footshock of
0.25 mA for 10 s, beginning 10 s after entering the
chamber. Immediately after application of the footshock,
mice were put in a box filled with carbon dioxide until
respiratory arrest occurred; mice were then revived by
artificial respiration using oxygen. Twenty-four hours
after the acquisition trial, a single retrieval trial was given
to each mouse and the step-through latency determined. If
a mouse did not enter the chamber within 180 s it was
removed from the runway and a score of 3 min was
recorded. The retrieval times (the times on day 3 of the
experiment) were submitted to a Kruskall-Wallis analysis
of ranks. Other relevant comparisons between treatments
and controls were made using Student’s t distribution.
4.2.1. Acute toxicity
For acute toxicity in mice, groups of 5 male Crl:CD-1
(ICR)BR mice (Charles River, Italy) weighing 21–30 g
were used. Mice fasted for 16 h were treated orally with
various doses of the compounds and observed for 1 week
after treatment. Deaths were recorded daily. The maxi-
mum dose administered, for compounds not showing
clear signs of toxicity, was referred to the value of
1 000 mg/kg p.o.
The median lethal doses (LD50) were then calculated
according to the method described by Litchfield and
Wilcoxon [12].
4.2.2. Effects on CO2-induced memory impairment of
passive avoidance task in mice
Acknowledgements
Hypercapnia, as well as hypoxia or ischaemia, is a
condition in which the reduction of the energy supply to
the brain results in an amnesia or in a learning defi-
cit [13]. It has been reported that the CO2-induced
memory impairment might be a manifestation of a
transient cerebral ischaemia [14]. Therefore, the above
method can be considered in evaluating new compounds
that might improve cognitive deficiencies due to cerebral
insufficiencies.
This work was supported by the University of Bologna
(funds for selected research topics).
References
[1] Thesleff S., Neuroscience 5 (1980) 1413–1419.
[2] Maciag C.M., Logue A.R., Tinker W.J., Saydoff J.A., Tam S.W.,
Zaczek R., J. Pharmacol. Exp. Ther. 271 (1994) 891–897.
[3] Wesseling H., Agoston S., Van Dam G.B., Pasma J., DeWit D.J.,
Havinga H., N. Engl. J. Med. 310 (1984) 988–989.
In our experimental conditions we used the method
described by Van Eys et al. [15] with some modifications.
Groups of 10 male Crl:CD-1 (ICR)BR mice (Charles
River, UK) weighing 19–25 g were used. The one-trial
passive avoidance apparatus consisted of a 16 × 15 ×
15 cm chamber with opaque walls and a metal grid floor.
A 4 cm wide, 11.5 cm long runway protruded from the
front wall of the chamber. The runway was illuminated,
whereas the chamber was dark. When placed on the
runway, a mouse could enter the chamber through a 3 ×
3 cm opening. A scrambled footshock could be delivered
through the metal grid floor of the chamber. On day 1 of
the experiment each mouse was subjected to three train-
ing trials and could visit the chamber without receiving
the footshock. Mice were fasted overnight prior to the
experiment with water available ad libitum. On day 2 of
the experiment mice were dosed orally with vehicle or
tested compounds and 1 h after the acquisition trial was
[4] Davidson M., Zemishlany Z., Mohs R.C., Horvath T.B., Powchik P.,
Blass J.P., Davis K.L., Biol. Psychiatry 23 (1988) 485–490.
[5] Lenegre A., Chermat R., Avril I., Steru L., Porsolt R.D., Pharmacol.
Biochem. Behav. 29 (1988) 625–629.
[6] Houssin R., Bernier J.L., Hénichart J.P., Synthesis (1988) 259–261.
[7] Armarego W.L.F., J. Chem. Soc. (1962) 4094–4103.
[8] Hawes E.M., Gorecki D.K.J., J. Heterocycl. Chem. 9 (1972)
703–706.
[9] Taylor E.C., Crovetti A.J., J. Org. Chem. 19 (1954) 1633–1640.
[10] Vetelino M.G., Coe J.W., Tetrahedron Lett. 35 (1994) 219–222.
[11] Earl R.A., Zaczek R., Teleha C.A., Fisher B.N., Maciag C.M.,
Marynowski M.E. et al., J. Med. Chem. 41 (1998) 4615–4622.
[12] Litchfield Jr. J.T., Wilcoxon F., J. Pharmacol. Exp. Ther. 96 (1949)
99–113.
[13] Lewis J.D., Psycol. Rev. 76 (1969) 461–472.
[14] Jensen T.S., De Fine Olivarius B., Acta Neurol. Scand. 61 (1980)
115–124.
[15] Van Eys G., Rigter H., Leonard B.E., Pharmacol. Biochem. Behav.
3 (1975) 787–793.