Synthesis of conformationally constrained analogues of KN62, a potent antagonist of the P2X7-receptor

Article abstract of DOI:10.1016/S0960-894X(00)00083-4

Conformationally constrained analogues of KN62 containing 1,2,3,4-tetrahydro-7-hydroxyisoquinoline-3-carboxylic acid with S configuration in position 3 were synthesized and their antagonist activities were tested on human macrophage cells. While KN62 is a potent antagonist of the P2X₇ receptor, these analogues were inactive as antagonists and only one compound showed appreciable activity as P2X₇ antagonist, which was 30 times weaker than that reported for KN62. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00083-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 681±684
Synthesis of Conformationally Constrained Analogues of KN62,
a Potent Antagonist of the P2X7-Receptor
Pier Giovanni Baraldi, a,* Romeo Romagnoli, a Mojgan Aghazadeh Tabrizi, a
Simonetta Falzoni b and Francesco Di Virgilio b
a
Á
Dipartimento di Scienze Farmaceutiche, Universita di Ferrara, I-44100 Ferrara, Italy
b
Á
Dipartimento di Medicina Diagnostica e Sperimentale, Sezione di Patologia Generale, Universita di Ferrara, I-44100 Ferrara, Italy
Received 16 November 1999; accepted 3 February 2000
AbstractÐConformationally constrained analogues of KN62 containing 1,2,3,4-tetrahydro-7-hydroxyisoquinoline-3-carboxylic
acid with S con®guration in position 3 were synthesized and their antagonist activities were tested on human macrophage cells.
While KN62 is a potent antagonist of the P2X7 receptor, these analogues were inactive as antagonists and only one compound
showed appreciable activity as P2X7 antagonist, which was 30 times weaker than that reported for KN62. # 2000 Elsevier Science
Ltd. All rights reserved.
Introduction
been proven very useful for the evaluation of P2X7
agonists and antagonists, and among these latter the
KN62 (1-(N,0-bis(1,5-isoquinolinesulfonyl)-N-methyl-l-
tyrosil)-4-phenylpiperazine) compound is, to the best of
our knowledge, the most potent with IC50, for inhibition
of ATP-stimulated Ca2+ in¯ux into fura-2 loaded
human macrophage cells, of 13.4 nM and complete
inhibition at a concentration of 500 nM.5 The same
compound selectively inhibited Ca2+/calmodulin-
dependent protein kinase II (CaMK II)6 activity and,
at non-cytotoxic concentrations (2 mM), it enhanced
etoposide cytotoxicity in Adriamycin-resistant cells
(HL-60).7
It is increasingly recognized that extracellular ATP acts
as an extracellular messenger and mediates a wide range
of e€ects by stimulating purinergic P2 receptors,1 which
are considered a promising new target for anti-
in¯ammatory drug development. P2 receptor subtypes
comprise an ionotropic (P2X1 7) and a metabotropic
(P2Y1 11) family, and increasing evidence suggests that
they are involved in the modulation of the immune and
in¯ammatory responses. Unfortunately, characteriza-
tion of native and recombinant P2 receptor continues to
be hindered by the lack of speci®c and subtype-selective
antagonists.
In an attempt to improve the activity and to study the
structure±activity relationship of KN62 inhibition on
P2X7 receptor, tethering the N-methyl of the tyrosine
backbone to the ortho-position of the phenyl ring has
resulted in a series of conformationally constrained
KN62 analogues with the formula 1±5. These deriva-
tives have been synthesized in order to restrict con-
formational freedom and to stabilize the desired
bioactive conformation of KN62. The derivatives 1±5
possess 1,2,3,4-tetrahydro-isoquinoline (Tic) moiety,
that can be considered as a cyclic tyrosine in which the
side chain orientation has been ®xed by the methylene
unit which bridges the 20 position in the aromatic ring
and the a-nitrogen. While the compound 1 contains the
isoquinoline-5-sulfonyl moiety which acts as ``ATP
mimic'',8 in the derivatives 2 and 3 this function has
been substituted with two isomers corresponding to
One of the most interesting members of the ionotropic
P2X family is P2X7. This receptor is mainly, if not
exclusively, expressed by mononuclear phagocytes,
where it mediates cytotoxic responses, cytochine release
and cell fusion.2,3 The P2X7 receptor is formed by the
assembly of an unknown number of subunits each 595
AA long, and upon stimulation by extracellular ATP
generates a nonselective membrane pore permeable to
hydrophilic molecules with molecular weight up to 900
D.4 Due to its likely involvement in immunomodula-
tion, it would be of the most importance to develop
selective P2X7 antagonists. Human macrophages have
*Corresponding author. Tel.: +39-532-291293; fax: +39-532-291296;
e-mail: pgb@dns.unife.it
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00083-4

682
P. G. Baraldi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 681±684
quinoline-5 and 8-sulfonyl, respectively. With the aim to
study the binding capacity of the isoquinoline nitrogen
to the P2X7 receptor, we have considered the replace-
ment of one or both of these moieties with the same
number of naphtalenic rings to obtain the derivatives 4
and 5, respectively.
(fura-2-AM) as described previously.13 Results are
expressed as percentage inhibition of maximal response
to 1 mM ATP in the absence of inhibitor, which was
de®ned as 100% response.
The compounds 1±5 synthesized had no e€ect on the
P2X7 receptor and for the derivatives 1±3 of this series,
their inhibitory e€ects at this receptor was reversed
when their concentrations were increased. This phe-
nomenon could be explained from the appearance of
cytotoxic e€ects or precipitation at high concentrations.
The reference compound KN62 was considerably more
potent than 5 (IC50, 13.4 nM for KN62 versus 316 nM
for 5), whereas 1±4 are inactive. As can be seen in
Table 1, the compound 5 was the most active antagonist
in this series and blocked more markedly the P2X7
receptor than did the other compounds 1±4. It is note-
worthy that the constrained form of KN62, corre-
sponding to derivative 1,14 lost completely antagonist
properties of the parent compound. In fact, the com-
pound 1, at the concentration of 50 nM, only inhibited
the Ca2+ in¯ux by 1.66%, while at the same con-
centration, KN62 shows a 94% of inhibition.
Chemistry
The Pictet±Spengler reaction has been widely used for
the synthesis of a variety of alkaloids because of gen-
erally good yields and mild reaction conditions.9 This
reaction has been applied as a key step to access to the
tetrahydroisoquinoline moiety present in the con-
formationally constrained analogues 1±5 of KN62.
The preparation of 1±5 has been accomplished by a
synthetic sequence according to Scheme 1. The com-
pound 6 has been easily obtained by already published
procedure10 and transformed into the corresponding
benzyloxycarbonyl (Z or CBz) derivative 7 by action of
Z-OSu/TEA system in dry DMF. The protected pre-
cursor 7 was converted to the activated 1-hydroxy-1,
2,3-benzotriazole (HOBT) ester and condensed with the
N-phenylpiperazine to give rise the amide 8. This latter
compound has been dehalogenated by catalytic hydro-
genation with 10% Pd/C and in presence of TEA,
without removal of Z protecting group, furnishing the
N-CBz-protected 9. The compounds 10±14 were pre-
pared from the anion of 9 (generated with NaH) by
treatment with a CH2Cl2 solution of the appropriate
sulfonyl chloride.11 The protecting CBz group in 10±14
was conveniently cleaved o€ by the use of 3 equiv of
(CH3)3SiI in acetonitrile12 and furnished 15±19, respec-
tively, in good yields. Finally, the sulphonamides 1±5
were synthesized from the amines 15±19 by treatment
with the appropriate sulfonyl chloride.11
Concerning the derivative 2, where the isoquinoline-5-
sulfonyl moiety has been substituted with quinoline-5-
sulfonyl, at the concentration of 50 nM it shows a 59%
inhibition, i.e. a value slight lower than that reported for
KN62 at the same concentration. The compound 3,
which possess the quinoline-8-sulfonyl instead of iso-
quinoline-5-sulfonyl moiety, shows a behaviour com-
parable than that of 1, since its antagonist activities
decreased with the increasing of the concentration. In
the compound 1, the substitution of the 5-isoquino-
linesulfonyl residue linked to the nitrogen with a naph-
thalene lead to the derivative 5, which is, if compared to
the other derivatives 1±4, the most potent inhibitor in
this series. Nevertheless, the IC50 value of 5 was 316
nM, almost 30-fold higher than that reported for KN62.
Replacing the isoquinoline with naphthalene provides 4,
which shows no capability to inhibit the Ca2+in¯ux at
any concentration.
Results and Discussion
The antagonist activity of KN62 and derivatives 1±5
was determined by the measurements of the ATP-
stimulated Ca2+ in¯ux into macrophages loaded with
the ¯uorescent indicator fura-2-acetoxymethyl ester
Comparing the antagonist activities of 1 and KN62, the
results con®rm that an extended rather than folded
conformation of KN62 is preferred at P2X7 receptor.

P. G. Baraldi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 681±684
683
Scheme 1. Reagents: (a) Z-Osu, TEA, DMF, rt; (b) HOBt, DCC, DMF, 70ꢀC, 4 h; (c) 4-Pheylpiperazine, rt, 18 h; (d) H2, Pd/C, TEA, dioxane/
EtOH, rt, 6 h; (e) NaH, dioxane, rt, 15 min; (f ) appropriate solfonyl chloride (see ref 11), TEA, dioxane, rt, 18 h; (g) (CH3)3Sil, CH3CN, 1 h at 0ꢀC
and then 4 h at rt; (h) appropriate solfonyl chloride (see ref 11), TEA, dioxane, rt, 18 h.
Table 1. Activities of synthesized compounds 1±5 and KN62 on the
calcium in¯ux
References and Notes
1. For reviews see: (a) Burnstock, G. Neuropharmacology
1997, 36, 1127; (b) Fredholm, B. B.; Abbracchio, M. P.;
Burnstock, G.; Daly, J. M.; Harden, T. K.; Jacobson, K. A.;
Le€, P.; Williams, M. Pharmacol. Rev. 1996, 46, 143; (c) Di
Virgilio, F.; Chiozzi, P.; Falzoni, S.; Ferrari, D.; Sanz, M. J.;
Venketaraman, V.; Baricoldi, O. R. Cell Death and Di€er-
entiation 1998, 5, 191; (d) Fischer, B. Exp. Opin. Therap. Pat.
1999, 9, 385.
% Inhibition of Ca2+ in¯ux with
di€erent concentration of inhibitor
IC50
(nM)
a
Compound
50 nM
500 nM
1000 nM
1
2
3
4
1.66‹0.42 34.48‹5.26 25.80‹3.26
59.05‹8.04 72.36‹8.11
n.d.b
n.d.
n.d.
n.d.
66‹5.58
6.90‹1.11
0
3.40‹0.48
0
0
0
2. Di Virgilio, F.; Falzoni, S.; Mutini, C.; Sanz, J. M.; Chiozzi,
P. Drug Dev. Res. 1998, 45, 207.
5
KN62
17.63‹2.36 58.80‹7.18 58.80‹8.22
94.4‹4.7 100 100
316‹38.6
13.4‹1.4
3. (a) Dubyak, G. B.; Cli€ord, E. E.; Humphreys, B. D.; Ker-
tesy, S. B.; Martin, K. A. Drug. Dev. Res. 1996, 39, 269; (b)
Ferrari, D.; Los, M.; Bauer, M. K. A.; Vandenabeele, P.;
Wesselborg, S.; Schulzeloho€, K. FEBS Lett. 1999, 447, 71;
(c) Schulzeloho€, E.; Hugo, C.; Rost, S.; Arnold, S.; Gruber,
A.; Brune, B.; Sterzel, R. B. Am. J. Physiol. 1998,44, 962;
(d) Coutinho Silva, R.; Persechini, P. M.; Bisaggio, R. D.;
Perfettini, J. L.; Neto, A. C. T. D.; Kanellopoulos, J. M.;
Motta Ly, I.; Dautry Varsat, A.; Ojcius, D. M. Am. J. Physiol.
1999, 45, 1139.
4. Suprenant, A.; Rassendren, F.; Kawashima, E.; North, R.
A.; Buell, G. Science 1996, 272, 735.
5. Gargett, C. E.; Wiley, J. S. Br. J. Pharmacol. 1997, 120,
1483.
6. Hidaka, H.; Yokokura, H. Adv. Pharmacol. 1996, 36, 193.
7. Kawamura, K.; Grabowski, D.; Krivacic, K.; Hidakan, H.;
Ganapathi, R. Biochem. Pharmacol. 1996, 12, 1903.
8. Ricouart, A.; Gesquire, J. C.; Tartar, A.; Sergheraert, C. J.
Med. Chem. 1991, 34, 73.
aIC50=50% inhibitory concentration represents the mean from dose±
response curves of at least three experiments. All experiments were
repeated three times.
bn.d.=not determined.
Therefore in the synthesized compounds 1±5, the incor-
porated conformational constraint probably completely
inhibit the interaction with the P2X7 receptor. The
resulting compounds 1±5 were shown not to be P2X7
antagonists and constraining the relatively mobile
phenyl ethyl group of the tyrosine onto the framework
of TIC did not yield the desired results.
Acknowledgement
9. (a) Whaley, W. M.; Govindachari, T. R. Org. React. 1951,
6, 151; (b) Cox, E. D.; Cook, J. M. Chem. Rev. 1995, 6, 1797.
10. Verschueren, K.; Toth, G.; Tourwe, D.; Lebl, M.; Van
Binst, G.; Hruby, V. Synthesis 1991, 458.
We thank Ministero Universita e Ricerca Scienti®ca
(MURST) (60%) for generous ®nancial support of this
work.

684
P. G. Baraldi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 681±684
11. For the synthesis of the isoquinoline-5-sulfonyl chloride
see: Morikawa, A.; Sone, T.; Asano, T. J. Med. Chem. 1989,
32, 42; The quinoline-8-sulfonyl chloride has been prepared
following the procedure reported by Edinger, A. Chem. Ber.
1908, 41, 937; while for the preparation of quinoline-5-sulfonyl
chloride see: Albert, A.; Garlin, G. B. J. Chem. Soc. 1959,
2384. The naphtalene-1-sulfonyl chloride is commercially
available.
12. (a) Jung, M. E.; Lyster, M. A. J. Chem. Soc., Chem.
Commun. 1978, 315; (b) Olah, G. A.; Narang, S. C.; Gupta, B.
G. B.; Malhotra, R. J. Org. Chem. 1979, 44, 1247.
13. Di Virgilio, F.; Fasolato, C.; Steinberg, T. H. Biochem. J.
1988, 256, 959. Brie¯y, washed cells (106ÂmL 1) suspended in
HEPES bu€ered saline were loaded with 4 mM fura-2-acetoxy-
methyl ester and incubated in a thermostat-controlled (37 ꢀC)
for 20 min in the dark and magnetically stirred ¯uorometer
cuvette. Cell suspension was incubated with KN62 or com-
pounds 1±5 (25±500 nM) for 5 min at 37 ꢀC before ¯uorimetric
analysis in a stirred cuvette at 37 ꢀC. Intracellular Ca2+ con-
centration was determined with the 340:380 nm excitation
ratio at an emission wavelenght of 500 nm. All experiments
were repeated three times.
14. Selected data for the compound 1. O€ white solid; mp
(uncorrected): 142±146 ꢀC (dec.); [a]2d5 (dioxane): +27.7 (c=0.6).
1H NMR (DMSO-d6) d: 2.96 (m, 2H), 3.22 (m, 4H), 3.76 (m,
4H), 4.49 (d, J=11 Hz, 1H), 4.60 (d, J=12.6 Hz, 1H), 5.40 (m,
1H), 6.46 (m, 2H), 6.83 (m, 4H), 7.23 (t, J=8 Hz, 2H), 7.80 (t,
J=7.2 Hz, 1H), 7.87 (t, J=7.2 Hz, 1H), 8.42 (m, 4H), 8.61 (t,
J=8 Hz, 2H), 8.82 (t, J=6.4 Hz, 2H), 9.42 (s, 1H), 9.58 (s,
1H). IR (KBr) cm 1: 3434, 1647, 1458, 1378, 1225, 1174, 828.
Elemental analysis (calcd): C. 63.40; H. 4.62; N. 9.73; S. 8.91
(found): C. 62.98; H. 4.43; N. 9.42; S. 8.71.