Structure-activity relationships for the Ca2+-releasing activity of 6-hydroxy-β-carboline analogues in skeletal muscle sarcoplasmic reticulum - The effects of halogen substitution at C-5 and C-7

Article abstract of DOI:10.1211/0022357001774309

This study of structure-activity relationships of 6-hydroxy-β-carboline analogues has been performed on the basis of quantitative measurement of Ca²⁺-releasing activity in the sarcoplasmic reticulum of skinned fibres of skeletal muscle. Substitution of halogens for hydrogens at the C-5 and C-7 positions and further introduction of a methyl group into the N-9 position of 6-hydroxy-β-carboline resulted in Ca²⁺-releasing activity. The 50% effective concentrations of 5,7-dibromoeudistomin D, 5,7-dichloroeudistomin D, 5,7-diiodoeudistomin D, 9-methyl-5,7-dibromoeudistomin D, 9-methyl-5,7-dichloroeudistomin D, 9-methyl-5,7-diiodoeudistomin D, and caffeine were 5.6 x 10^-6, 6.3 x 10^-6, 7.8 x 10^-6, 2.1 x 10^-6, 2.0 x 10^-5, 3.7 x 10^-5, and 4.7 x 10^-4 M, respectively, indicating that these analogues are 10-200 times more potent than caffeine. Substitution of bromine by chlorine or iodine at the C-5 and C-7 positions markedly reduced the activity of the analogues with a methyl group at the N-9 position. These results suggest that halogens at the C-5 and C-7 positions in the β-carboline skeleton are essential for Ca²⁺-releasing activity and that an N-9 methyl group also affects the activity of these analogues. Thus, these 6-hydroxy-β-carboline analogues might become powerful tools for studying the molecular mechanism of Ca²⁺ release in the sarcoplasmic reticulum.

Full text of DOI:10.1211/0022357001774309

J. Pharm. Pharmacol. 2000, 52: 517±521
Received August 16, 1999
Accepted January 9, 2000
# 2000 J. Pharm. Pharmacol.
Structure±Activity Relationships for the Ca2‡-releasing Activity
of 6-Hydroxy-b-carboline Analogues in Skeletal Muscle
Sarcoplasmic ReticulumÐThe Effects of Halogen Substitution at
C-5 and C-7
ASAMI SEINO-UMEDA*, MASAMI ISHIBASHI{, JUN'ICHI KOBAYASHI{ AND YASUSHI OHIZUMI
Department of Pharmaceutical Molecular Biology, Graduate School of Pharmaceutical Sciences, Tohoku
University, Aoba-ku, Sendai 980-8578, *Yokohama Research Centre, Mitsubishi Chemical Corporation,
Aoba-ku, Yokohama 227-8502, {Laboratory of Natural Product Chemistry, Faculty of Pharmaceutical
Sciences, Chiba University, Inage-ku, Chiba 263-8522 and {Department of Pharmacognosy, Graduate
School of Pharmaceutical Sciences, Hokkaido University, Sapporo 060-0812, Japan
Abstract
This study of structure±activity relationships of 6-hydroxy-b-carboline analogues has been
performed on the basis of quantitative measurement of Ca2‡-releasing activity in the
sarcoplasmic reticulum of skinned ®bres of skeletal muscle.
Substitution of halogens for hydrogens at the C-5 and C-7 positions and further
introduction of a methyl group into the N-9 position of 6-hydroxy-b-carboline resulted in
Ca2‡-releasing activity. The 50% effective concentrations of 5,7-dibromoeudistomin
D, 5,7-dichloroeudistomin D, 5,7-diiodoeudistomin D, 9-methyl-5,7-dibromoeudistomin
D, 9-methyl-5,7-dichloroeudistomin D, 9-methyl-5,7-diiodoeudistomin D, and caffeine
6
6
6
6
5
5
were 5Á6 6 10 , 6Á3 6 10 , 7Á8 6 10 , 2Á1 6 10 , 2Á0 6 10 , 3Á7 6 10 , and
4
4Á7 6 10 M, respectively, indicating that these analogues are 10±200 times more potent
than caffeine. Substitution of bromine by chlorine or iodine at the C-5 and C-7 positions
markedly reduced the activity of the analogues with a methyl group at the N-9 position.
These results suggest that halogens at the C-5 and C-7 positions in the b-carboline
skeleton are essential for Ca2‡-releasing activity and that an N-9 methyl group also affects
the activity of these analogues. Thus, these 6-hydroxy-b-carboline analogues might
become powerful tools for studying the molecular mechanism of Ca2‡ release in the
sarcoplasmic reticulum.
Contraction of skeletal muscle is triggered by the
rapid release of Ca2‡ from the ryanodine receptor
located in the terminal cisternae of the sarco-
plasmic reticulum (Meissner 1994; Zucchi &
Ronca-Testoni 1997). The puri®ed ryanodine
receptor is identical in morphology with the base
structures that span the transverse tubule-sarco-
plasmic reticulum junction and form caffeine-sen-
sitive Ca2‡ channels (Shoshan & Ashley 1998).
Numerous studies using fragmented sarcoplasmic
reticulum preparations and sarcoplasmic reticulum
from skinned ®bres of skeletal muscle have
revealed the presence of a caffeine-sensitive Ca2‡-
release pathway (Coronado et al 1994), although
characterization of the caffeine-binding site in the
Ca2‡-release channels of the sarcoplasmic reticu-
lum has not been possible because of its low
af®nity. The detailed molecular mechanism of
caffeine-induced Ca2‡ release from the sarco-
plasmic reticulum remains unresolved.
In the course of our survey of Ca2‡ release in the
sarcoplasmic reticulum, 6-hydroxy-b-carboline
analogues derived from eudistomin D, a marine
natural product isolated from a Caribbean tunicata
Eudistoma olivaceum, have been found to be
powerful Ca2‡ releasers in skeletal muscle sarco-
plasmic reticulum (Kobayashi et al 1984, 1989;
Nakamura et al 1986). It has been demonstrated
that 9-methyl-5,7-dibromoeudistomin D shares the
same binding site as caffeine in the skeletal muscle
Correspondence: Y. Ohizumi, Graduate School of Pharma-
ceutical Sciences, Tohoku University, Aoba-ku, Sendai 980-
8578, Japan.

518
ASAMI SEINO-UMEDA ET AL
sarcoplasmic reticulum and induces Ca2‡ release in
the same manner as caffeine (Seino et al 1990;
Fang et al 1993).
Materials and Methods
Synthesis of 6-hydroxy-b-carboline analogues
The 6-hydroxy-b-carboline analogues were syn-
thesized as previously described (Kobayashi et al
1988, 1989). 5,7-Dibromoeudistomin D was synthe-
sized by bromination of 6-methoxypyrido[3,4-b]
indole with N-bromosuccinimide in acetic acid then
demethylation with BBr3. 5,7-Dichloroeudistomin
D was prepared by the same procedure but with N-
chlorosuccinimide. 5,7-Diiodoeudistomin D was
synthesized by iodination of 6-methoxypyrido[3,4-
b]indole with iodine and HIO4 in acetic acid then
demethylation with BBr3. 5,7-Dibromoeudistomin
D, 5,7-dichloroeudistomin D, and 5,7-diiodoeudis-
tomin D were O-acetylated by treatment with acetic
anhydride; methylation with methyl iodide was
then followed by saponi®cation to give 9-methyl-
5,7-dibromoeudistomin D, 9-methyl-5,7-dichloro-
eudistomin D, and 9-methyl-5,7-diiodoeudistomin
D, respectively.
This paper reports a quantitative study of
structure±activity relationships for the Ca2‡-
releasing activity of 6-hydroxy-b-carboline analo-
gues (Figure 1) in the sarcoplasmic reticulum of
skinned muscle ®bres. We examined the effect on
Ca2‡-releasing activity of halogen substitution at
positions C-5 and C-7 (5,7-dibromoeudistomin D,
5,7-dichloroeudistomin D, and 5,7-diiodoeudisto-
min D) and further introduction of a methyl group
at the N-9 position (9-methyl-5,7-dibromoeudisto-
min D, 9-methyl-5,7-dichloroeudistomin D, and 9-
methyl-5,7-diiodoeudistomin D) of 6-hydroxy-b-
carboline, and report that these analogues are 10±
200 times more potent than caffeine.
Ca2‡ release from the sarcoplasmic reticulum of
skinned muscle ®bres
The isometric tension of chemically skinned ®bres
from guinea-pig psoas muscle was measured at
23ꢀC as previously described (Endo & Iino 1988;
Seino et al 1990). If the concentration of EGTA,
magnesium methanesulphonate, and calcium
methanesulphonate in a solution are denoted X,
Y and Z, respectively, the solution denoted
``GXMYCZ'' contained X mM EGTA, Y mM
magnesium methanesulphonate, Z mM calcium
methanesulphonate, 4 mM ATP, and 20 mM
PIPES±KOH (pH 7Á0). The concentration of
Mg2‡ATP was adjusted to 3Á5 mM (for G2M5C0,
G0M5C0, G10M5C7 and G0.2M5C0) or 1 mM (for
G2M1C0 and G0.2M1C0). In each solution the
ionic strength was ®xed at 0Á2 M by adding an
appropriate amount (90±130 mM) of potassium
methanesulphonate (Endo & Iino 1988).
A small bundle (0Á1 mm diam., 3 mm length,
approx.) of muscle ®bres was ®xed on the needles
of a force displacement transducer (AE 801, Sen-
soNor, Norway) in G2M5C0 solution ([Mg2‡]free
(free Mg2‡ ion) ˆ 1Á5 mM) and the resting sarco-
mere length was set to 2Á8 mm. The ®bre was treated
for 30 min with G2M5C0 solution containing
1
50 mg mL saponin (ICN Pharmaceuticals, Cleve-
land, OH), and then washed with a G2M5C0
solution.
Figure 1. The chemical structures of caffeine, 6-hydroxy-b-
carboline, 5,7-dibromoeudistomin D, 5,7-dichloroeudistomin
D, 5,7-diiodoeudistomin D, 9-methyl-5,7-dibromoeudistomin
D, 9-methyl-5,7-dichloroeudistomin D and 9-methyl-5,7-diio-
doeudistomin D.
It has been reported that ions or compounds can
immediately enter the intracellular contractile sys-
tem of skinned ®bres prepared by this procedure
(Endo & Iino 1988). Suf®cient perforation of the

CA2‡-RELEASING ACTIVITY OF 6-HYDROXY-b-CARBOLINE ANALOGUES
519
cell membrane was indicated by the rapid response
of the ®bres to changes of extracellular Ca2‡ con-
centrations and the rigour state of the ®bres on
removal of Mg2‡ATP; this could immediately be
reversed by re-introducing Mg2‡ATP (Endo & Iino
1980; Horiuti 1986). When the ®bres were
immersed in G0M5C0 solution ([Mg2‡]free
ˆ
1Á5 mM), spontaneous contractions occurred
repeatedly at constant intervals of 10 to 15 min.
The amount of Ca2‡ released by the drug was
then measured. After immersion in the G10M5C7
solution ([Ca2‡]free (free Ca2‡ ion) ˆ 1 mM,
[Mg2‡]free ˆ 1Á5 mM) for 2 min to load sarco-
plasmic reticulum with Ca2‡, loading was stopped
by washing with G2M5C0 solution for 0Á5 min. The
®bres were then treated for 1 min with G2M1C0
solution ([Mg2‡]free ˆ 0Á09 mM) containing the test
substance. After washing with G0.2M5C0 solution
([Mg2‡]free ˆ 1 mM) for 0Á5 min, G0.2M1C0
([Mg2‡]free ˆ 0Á09 mM) solution containing 40 mM
caffeine was applied to the ®bres to cause con-
tractions that resulted from the release of Ca2‡
remaining in the sarcoplasmic reticulum. The
amount of Ca2‡ released by the drugs in the
G2M1C0 solution was determined from the dif-
ference between the amounts of contraction in the
presence and absence of the drugs.
Results and Discussion
We have previously reported that 6-hydroxy-b-
carboline has no stimulatory effect on Ca2‡-
releasing activity in skeletal muscle sarcoplasmic
reticulum, whereas brominated derivatives of this
compound induce Ca2‡ release and have properties
similar to those of caffeine (Nakamura et al 1986;
Seino et al 1990; Ohizumi 1997).
Figure 2. Contractile response to 6-hydroxy-b-carboline ana-
logues of chemically skinned ®bres from skeletal muscle. The
5
6-hydroxy-b-carboline analogues (10 M) and procaine
(3 mM) were applied just after relaxation from the contraction.
A. 5,7-dibromoeudistomin D; B. 5,7-dichloroeudistomin D; C.
5,7-diiodoeudistomin D.
In this study the Ca2‡-releasing activity of six 6-
hydroxy-b-carboline analogues (5,7-dibromoeudis-
tomin D, 5,7-dichloroeudistomin D, 5,7-diiodoeu-
distomin D, 9-methyl-5,7-dibromoeudistomin D, 9-
methyl-5,7-dichloroeudistomin D, and 9-methyl-
5,7-diiodoeudistomin D) and of caffeine were
examined in chemically skinned muscle ®bres.
Spontaneous contractions, repeated at constant
intervals of 5 to 15 min, were recorded for skinned
®bres the sarcoplasmic reticulum of which had
been loaded with Ca2‡ (Figure 2). The addition of
5,7-dibromoeudistomin D, 5,7-dichloroeudistomin
of contractions. Similar contractions were observed
after addition of 9-methyl-5,7-dibromoeudistomin
D, 9-methyl-5,7-dichloroeudistomin D, 9-methyl-
5
5,7-diiodoeudistomin
(10 M) (data not shown).
D
(10 M) or caffeine
3
These results suggest that these analogues act on
Ca2‡-induced Ca2‡-release channels of the sarco-
plasmic reticulum in a similar manner to caffeine.
The Ca2‡-releasing activity of the 6-hydroxy-b-
carboline analogues on the sarcoplasmic reticulum
of skinned ®bres was examined by quantitative
measurement of the amount of Ca2‡ released.
Figure 3A shows typical recording traces from the
measurement of Ca2‡ release from the sarco-
plasmic reticulum of skinned ®bres. In the control,
5
D or 5,7-diiodoeudistomin D (10 M) just after
relaxation induced immediate tension then con-
tractions of higher frequency. The subsequent
application of procaine (3 mM), an inhibitor of
Ca2‡-induced Ca2‡ release, reversed the effect of
the analogues and markedly reduced the frequency

520
ASAMI SEINO-UMEDA ET AL
The 50% effective concentrations (EC50) of 5,7-
dibromoeudistomin D, 5,7-dichloroeudistomin D,
5,7-diiodoeudistomin D, 9-methyl-5,7-dibromoeu-
distomin D, 9-methyl-5,7-dichloroeudistomin D,
9-methyl-5,7-diiodoeudistomin D, and caffeine
in Ca2‡-releasing action are shown in Table 1.
The results indicate that the 6-hydroxy-b-carboline
analogues were 10±200 times (approx.) more
potent than caffeine in sarcoplasmic reticulum from
skinned muscle ®bres. Substitution of the bromine
of 5,7-dibromoeudistomin D by chlorine or iodine
slightly increased the EC50 value, but it was not
signi®cantly different from that of 5,7-dibromo-
eudistomin D. The EC50 value was increased 10
and 20 times (approx.) by the substitution of the
bromine of 9-methyl-5,7-dibromoeudistomin D by
chlorine or iodine.
These results suggest that the presence of halo-
gens at the C-5 and C-7 positions of the b-carboline
skeleton is essential for the development of the
Ca2‡-releasing activity. They also suggest that
substitution of bromine by chlorine or iodine at the
C-5 and C-7 positions reduced the activity of ana-
logues in which a methyl group has been intro-
duced into the N-9 position. In addition, the slopes
of the concentration±response curves for 9-methyl-
5,7-dichloroeudistomin D, 9-methyl-5,7-diiodoeu-
distomin D and caffeine are slightly different from
those for 5,7-dibromoeudistomin D, 5,7-dichlor-
oeudistomin D, 5,7-diiodoeudistomin D and 9-
methyl-5,7-dibromoeudistomin D (Figure 3B),
suggesting that the pharmacological behaviour
of 9-methyl-5,7-dichloroeudistomin D, 9-methyl-
5,7-diiodoeudistomin D or caffeine are different
from those of the other 6-hydroxy-b-carboline
analogues.
Figure 3. Stimulatory effect of 6-hydroxy-b-carboline ana-
logues or caffeine on Ca2‡ release from the sarcoplasmic
reticulum of skinned skeletal muscle ®bres. A. Typical
recording traces. After 2 min Ca2‡ loading in a Ca2‡-contain-
ing solution (G10M5C7), the ®bres were treated with a
G2M5C0 solution for 0Á5 min,
a
releasing solution
5
(G2M1C0) with or without 3 6 10 M 5,7-dibromoeudisto-
min D for 1 min, a G0.2M5C0 solution for 0Á5 min, and a
contraction-inducing solution containing 40 mM caffeine
(G0.2M1C0). B. Concentration-dependent stimulation of
Ca2‡ release from the sarcoplasmic reticulum of skinned
muscle ®bres: s, 5,7-dibromoeudistomin D; n, 5,7-dichloro-
eudistomin D; u, 5,7-diiodoeudistomin D; d, 9-methyl-5,7-
dibromoeudistomin D; m, 9-methyl-5,7-dichloroeudistomin D;
j, 9-methyl-5,7-diiodoeudistomin D; ., caffeine. Data are
meansÆ s.e.m., n ˆ 4±10.
It has been shown that the ryanodine receptor is
identical with the Ca2‡-induced Ca2‡-release
channels of the sarcoplasmic reticulum (Meissner
Table 1. The 50% effective concentration (EC50) values of
6-hydroxy-b-carboline analogues on Ca2‡-releasing activity in
skinned ®bre sarcoplasmic reticulum.
the addition of 40 mM caffeine to the ®bres induced
a transient contraction as a result of the rapid
release of Ca2‡ accumulated in the sarcoplas-
mic reticulum. Pretreatment of the ®bres with
Compound
n
EC50 (M)
6
5,7-Dibromoeudistomin D
5,7-Dichloroeudistomin D
5,7-Diiodoeudistomin D
6 5Á6 (3Á5±9Á1) 6 10
6 6Á3 (3Á7±10Á7) 6 10
6 7Á8 (3Á2±18Á2) 6 10
5
6
6
6
5,7-dibromoeudistomin D (3 6 10 M) markedly
reduced the caffeine (40 mM)-induced contraction,
suggesting that Ca2‡ released by 5,7-dibromo-
eudistomin D is chelated by EGTA, reducing the
amount of Ca2‡ remaining in the sarcoplasmic
reticulum. As shown in Figure 3B, the 6-hydroxy-
b-carboline analogues enhanced Ca2‡ release from
sarcoplasmic reticulum of skinned ®bres in a con-
centration-dependent manner.
9-Methyl-5,7-bromoeudistomin D 7 2Á1 (0Á79±5Á8) 6 10
5
9-Methyl-5,7-dichloro-
eudistomin D
4 2Á0 (0Á83±47Á9) 6 10
5
9-Methyl-5,7-diiodoeudistomin D 4 3Á7 (2Á7±5Á0) 6 10
4
Caffeine
10 4Á7 (4Á0±5Á4) 6 10
The EC50 values were calculated from the concentration±
response curves in Figure 3. Values are means (95% con-
®dence limits in parentheses).

CA2‡-RELEASING ACTIVITY OF 6-HYDROXY-b-CARBOLINE ANALOGUES
521
Horiuti, K. (1986) Some properties of the contractile system
and sarcoplasmic reticulum of skinned slow ®bers from
Xenopus muscle. J. Physiol. 373: 1±23
Kobayashi, J., Harbour, G. C., Gilmore, J., Rinehart, K. L.
(1984) Eudistomins A, D, G, H, I, J, M, N, O, P, and Q,
bromo-, hydroxy-, pyrrolyl-, and 1-pyrrolinyl-carbolines
from the antiviral Caribbean tunicate Eudistoma olivaceum.
J. Am. Chem. Soc. 106: 1526±1528
Kobayashi, J., Taniguchi, M., Hino, T., Ohizumi, Y. (1988)
Eudistomin derivatives, novel phosphodiesterase inhibitors:
synthesis and relative activity. J. Pharm. Pharmacol. 40:
62±63
1994; Zucchi & Ronca-Testoni 1997; Shoshan
& Ashley 1998). Although caffeine is a typical
activator of the ryanodine receptor=Ca2‡ release
channels, because of its low af®nity and multiple
effects it cannot be used as a tool for clarifying its
binding site on the ryanodine receptor (Sawynok &
Yaksh 1993). The pharmacological properties
of 9-methyl-5,7-dibromoeudistomin D have been
investigated and have provided useful information
(Ohizumi 1997). In this study we have shown that
6-hydroxy-b-carboline analogues act on Ca2‡-
induced Ca2‡-release channels of the sarcoplasmic
reticulum and that their Ca2‡-releasing activity in
skinned muscle ®bres is 10±200 times more potent
than that of caffeine.
Kobayashi, J., Ishibashi, M., Nagai, U., Ohizumi, Y. (1989) 9-
Methyl-7-bromoeudistomin D, a potent inducer of calcium
release from sarcoplasmic reticulum of skeletal muscle.
Experientia 45: 782±783
Meissner, G. (1994) Ryanodine receptor=Ca2‡-release chan-
nels and their regulation by endogenous effectors. Annu.
Rev. Physiol. 56: 485±508
Not only 9-methyl-5,7-dibromoeudistomin D but
also the other analogues described here might
provide promising tools for studying the Ca2‡-
release process in excitation±contraction coupling
of skeletal muscle.
Nakamura, Y., Kobayashi, J., Gilmore, J., Mascal, M., Line-
hart, K. L., Nakamura, H., Ohizumi, Y. (1986) Bromo-
eudistomin D, a novel inducer of calcium release from
fragmented sarcoplasmic reticulum that causes contraction
of skinned muscle ®bers. J. Biol. Chem. 261: 4139±4142
Ohizumi, Y. (1997) Application of physiological active sub-
stances isolated from natural resources to pharmacological
studies. Jpn. J. Pharmacol. 73: 263±289
Sawynok, J., Yaksh, T. L. (1993) Caffeine as an analgesic
adjuvant: a review of pharmacology and mechanism of
action. Pharmacol. Rev. 45: 43±85
References
Coronado, R., Morrissette, J., Sukhareva, M., Vaughan, D. M.
(1994) Structure and function of ryanodine receptors. Am. J.
Physiol. 266: C1485±C1504
Endo, M., Iino, M. (1980) Speci®c perforation of muscle cell
membranes with preserved sarcoplasmic reticulum functions
by saponin treatment. J. Muscle Res. Cell Motil. 1: 89±100
Endo, M., Iino, M. (1988) Measurement of Ca2‡ release in
skinned ®bers from skeletal muscle. Methods Enzymol. 157:
12±26
Seino, A., Kobayashi, M., Kobayashi, J., Fang, Y.-I., Isibashi,
M., Nakamura, H., Momose, K., Ohizumi, Y. (1990) 9-
Methyl-7-bromoeudistomin D, a powerful radio-labelable
Ca2‡ releaser having caffeine-like properties, acts on
Ca2‡-induced Ca2‡ release channels of sarcoplasmic reti-
culum. J. Pharmacol. Exp. Ther. 256: 861±867
Shoshan, B. V., Ashley, R. H. (1998) The structure, function,
and cellular regulation of ryanodine-sensitive Ca2‡ release
channels. Int. Rev. Cytol. 183: 185±270
Fang, Y.-I., Adachi, M., Kobayashi, J., Ohizumi, Y. (1993)
High af®nity binding of 9-[3H]methyl-7-bromoeudistomin D
to the caffeine-binding site of skeletal muscle sarcoplasmic
reticulum. J. Biol. Chem. 268: 18622±18625
Zucchi, R., Ronca-Testoni, S. (1997) The sarcoplasmic reticu-
lum Ca2‡ channel=ryanodine receptor: modulation by endo-
genous effectors, drugs and disease states. Pharmacol. Rev.
49: 1±51