Design and synthesis of furoxan-based nitric oxide-releasing glucocorticoid derivatives with potent anti-inflammatory activity and improved safety

Article abstract of DOI:10.1016/j.bmcl.2006.11.018

A series of furoxan-based nitric oxide-releasing glucocorticoid derivatives was synthesized. The pharmacological assays indicated that three compounds, including I₄, I₅, and I₆, had anti-inflammatory activity. Furthermore compared with the leading compound hydrocortisone the safety of I₆ was greatly improved. Due to releasing NO in vivo the side effects of glucocorticoids, including hypertension and osteoporosis, were effectively avoided.

Full text of DOI:10.1016/j.bmcl.2006.11.018

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1062–1066
Design and synthesis of furoxan-based nitric oxide-releasing
glucocorticoid derivatives with potent anti-inflammatory
activity and improved safety
a,b
a,
b,
a
c
c
*
*
Lei Fang, Yihua Zhang, Jochen Lehmann, You Wang, Hui Ji and Dayong Ding
a
Center of Drug Discovery, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, PR China
b
Institute of Pharmacy, Friedrich-Schiller-University of Jena, Philosophenweg 14, D-07743 Jena, Germany
Department of Pharmacology, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, PR China
c
Received 11 October 2006; revised 7 November 2006; accepted 8 November 2006
Available online 10 November 2006
Abstract—A series of furoxan-based nitric oxide-releasing glucocorticoid derivatives was synthesized. The pharmacological assays
indicated that three compounds, including I , I , and I , had anti-inflammatory activity. Furthermore compared with the leading
compound hydrocortisone the safety of I was greatly improved. Due to releasing NO in vivo the side effects of glucocorticoids,
including hypertension and osteoporosis, were effectively avoided.
2006 Elsevier Ltd. All rights reserved.
4
5
6
6
ꢀ
Glucocorticoids (GC) have been widely used for their
potent anti-inflammatory and immunosuppressive ac-
tions since the 1950s. However, the side effects, which
depend on the dose used and duration of treatment,
lar diseases for several decades, though its mechanism
of action was not elucidated until the 1980s when the
biology of NO was identified. Moreover, a great number
of encouraging research results for NO-releasing non-
steroidal anti-inflammatory drugs (NSAIDs) have dem-
onstrated that NO could exert strong anti-inflammatory
1
greatly limit the application of GC. The long-term
treatment with GC leads to an array of unwanted effects
ranging from alteration of the endocrine homeostasis to
a number of complications such as hypertension, osteo-
porosis, gastrointestinal damage, etc. Therefore, the
development of potent GC derivatives with improved
safety is crucial.
4
effects as well as reduce gastrointestinal damage.
Recently, it was found that NO could both inhibit bone
5
resorption and increase bone formation, which will be
very beneficial to osteoporosis sufferers.
In our study, we describe a group of NO-releasing deriv-
atives of hydrocortisone that act as a new class of im-
proved GC-related agents. Benzenesulfonyl-substituted
furoxans, a kind of NO donors which can donate two
molecules of NO in vivo under the action of thiol cofac-
Nitric Oxide (NO), which is naturally generated from
L-arginine by the action of NO synthase (NOS), is a key
signaling molecule involved in the regulation of many
physiological processes including vascular relaxation,
neurotransmission, and events of the immune system.
Dysfunction of NO formation has been implicated in
the pathogenesis of a number of disorders. Exogenous
NO sources constitute a powerful way to supplement
NO when the body cannot generate enough for normal
6
tors, had been incorporated to the 21 position of hydro-
cortisone through various spacers (as shown in Table 1).
In 2004, Pier and his co-workers reported a group of
NO-releasing derivatives of GC, which have proved that
the structure modification of the 21 position does not
2
,3
7
biological functions. NO-releasing agents, such as
glyceryl trinitrate, have been used to treat cardiovascu-
interfere with the molecular recognition of GC. In
our design, furoxans were substituted for organic ni-
trates acting as NO donors, because furoxans are sup-
posed to be able to release higher concentrations of
NO in vivo and can also effectively avoid ‘‘the nitrate
tolerance’’. By releasing NO in vivo, we hope to enhance
the anti-inflammatory activity of these derivatives and
particularly to minimize the side effects.
Keywords: Glucocorticoids; Hydrocortisone; Furoxan; NO-donors;
Anti-inflammatory activity; Safety.
*
Corresponding authors. Tel./fax: +86 25 86635503 (Y.Z); tel.: +49
641 94 9803; fax: +49 3641 94 9802 (J.L.); e-mail addresses:
zyhtgd@hotmail.com; j.lehmann@uni-jena.de
3
0
960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.018

L. Fang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1062–1066
Table 1. The structures of the target compounds and spacers
1063
In preparing compound 7 and compound 7 , we as-
7
8
sumed two isomers existed, because OH functions in
both phenol and alcohol reportedly react with the di-
O
O
SO C H
2 6 5
X
O
HO
OH
O
N
N
9
O
O
phenylsulfonylfuroxans (6). However, only one product
was gained. Its H NMR spectrum showed no phenol
1
10
O
–OH signal and when added to FeCl solution, no color
3
change was observed. Thus, we could confirm only that
phenol –OH reacted with the furoxans. The reason is
probably that under the basic conditions the phenol
Compound
X
/
Spacers2–8
/
I
1
I
2
I
3
I
4
I
5
I
6
I
7
I
8
–OH could be converted to relatively more stable
ph–O anion than alcohol –O anion, and then the
ꢀ
ꢀ
ꢀ
ph–O
anion subsequently substitutes for the
phenylsulfonyl group of the furoxans to form the
products (Scheme 2).
The anti-inflammatory activity of the target compounds
was at first evaluated using xylene-induced mouse ear
swelling as a model.
Three of the compounds, including I , I , and I , show
5
4
6
anti-inflammatory activity (Fig. 1). Particularly, the
activity of I was comparable to that of the control
6
Hyc. The different results of the tested compounds
may be due to the different spacers. Derivatives with
spacers of 5–7 atoms are more active than the com-
pounds with shorter or longer spacers. Moreover, the
aliphatic spacers were found to be better than the aro-
matic ones. The activity of the tested compounds is sup-
posed to depend on the release of Hyc and NO in vivo.
HOCH2
OH
HO(H C)2
2
OH
2 6 5
SO C H
C H O S
6
5
2
HOH
2
C
OH OH-
HOH
2
C
O-
N
N
Hydrocortisone (1,Hyc) was reacted with succinic anhy-
dride in refluxing pyridine to give ester (2) in 94% yield.
The substituted furoxans were prepared in a four-step
sequence. The starting material benzenethiol (3) was
converted to 2-(phenylthio)acetic acid (4) by treatment
with chloroacetic acid. Compound 4 was oxidized by
30% H O aqueous solution to give 2-(phenylsulfo-
2 2
nyl)acetic acid (5), followed by treatment with fuming
HNO to offer diphenylsulfonylfuroxan (6). It was then
3
O
6
O
HOH
2
C
O
2 6 5
SO C H
N
N
O
O
77
HO
CH
2
O
2 6 5
SO C H
×
SO
O
2
C
6
H
5
N
N
C
6
H
5
O
2
S
O
O
N
N
O
converted to various mono-phenylsulfonylfuroxans
(71–8) by treatment with corresponding diol, or amino
substituted alcohol (spacers2–8). Finally, the resulting
furoxans were reacted with 2 to give target compounds.
The synthetic route is shown in Scheme 1.
6
7
Scheme 2. The mechanism of the reaction of preparing compound 7 .
8
O
O
O
O
O
OH
OH
O
OH
HO
O
HO
OH
O
pyridne,reflux 5h
O
O
1
2
DCC
Cl COOH
H
2
O
2
I
1-8
SH
SCH
2
COOH
DMAP
2 2
SO CH COOH
CH
spacers
5%NaOH
3
CO
2
H
3
4
5
SO
2
C H
6 5
H R
2 6 5
SO C H
HNO
3 6 5 2
C H O S
N
N
N
N
O
O
O
O
2
reflux,5h
6
71-8
Figure 1. Effects of tested compounds on xylene-induced ear edema in
mice (n = 10, x ꢁ s). P < 0.05, **P < 0.01 versus CMC-Na.
*
Scheme 1. The synthetic route of the target compounds I1–8
.

1064
L. Fang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1062–1066
Different length and steric conformation of the spacers
can affect both the hydrolysis of the 21 position ester
bond, which needs catalysis by enzymes, and interaction
with thiol cofactors for releasing NO out of the furoxan.
Furthermore, I was selected to be evaluated for anti-
6
acute-inflammatory activity by carrageenan-induced
rat paw edema model (Table 2) and anti-subacute-in-
flammatory activity by cotton pellet-induced granular
tissue formation model (Fig. 2). And the anti-immune
inflammatory activity of I was also tested in the arthri-
6
tis model induced by complete Freund’s adjuvant
(
Figure 2. Effects of I
formation (n = 8, x ꢁ s).
6
on cotton pellet-induced granular tissue
P < 0.05, **P < 0.01, ***P < 0.001 versus
Fig. 3).
*
CMC-Na.
The results revealed that I significantly inhibited carra-
6
geenan-induced hind paw edema in acute inflammatory
models, and exhibited pronounced dose-related inhibi-
tory activity in the formation of granuloma in the suba-
cute inflammatory model. The anti-inflammatory effects
of an equimolar dose of I were comparable to Hyc and
6
no difference among the groups was observed. Both Hyc
and an equimolar dose of I inhibited the arthritis in-
6
duced by complete Freund’s adjuvant significantly. In
the early phase (at the 17th day) Hyc had a similar effect
on the secondary arthritis, whereas in the later phase (at
the 21th and the 24th day) I had a more potent effect
6
than Hyc.
Osteoporosis (OP) is one of the primary adverse effects
of GC. NO is supposed to prevent OP by both inhibiting
Figure 3. Effect of I
6
on paw swelling in AA rats (n = 10, x ꢁ s).
#
1
1
bone resorption and promoting bone growth. Thus to
determine the bone formation activity of I , osteoblasts
*P < 0.05, **P < 0.01 versus AA model group; P < 0.05 versus Hyc
group; P < 0.01 versus normal group.
6
were cultured from the parietal bone of newborn SD
1
mice in vitro using a similar method to that described
2
1
3
by Garett. Then the osteoblasts were treated with I6.
The numbers of osteoblasts (Fig. 4) as well as the activ-
ity of alkaline phosphatase (ALP) (Table 3) were
measured.
In addition, in order to assess the bone resorption activ-
ity of I osteoclasts were also obtained from the extreme
6
1
4
bones of newborn rats in vitro. After the treatment of
I6, the numbers of osteoclasts (Fig. 5) and tartrate-resis-
tant acid phosphatase cells (TRAP) (Fig. 6) were
measured.
The osteoblast assays demonstrated that there was no
significant difference among I and Hyc group in the
6
Figure 4. Effects of I and Hyc on proliferation of osteoblasts (n = 6,
6
proliferation of osteoblasts. Both I and Hyc could stim-
6
x ꢁ s).
Table 2. Effects of I
Group
6
on carrageenan-induced rat paw edema (n = 8, x ꢁ s)
CMC-Na
Hyc
I
6
I
6
I
6
Dose (mg/kg)
Edema volume (ml)
—
21.7
23.2
46.4
92.9
*
*
*
*
*
**
*
**
***
***
***
***
*
1 h
0.24 ± 0.04
0.62 ± 0.13
0.93 ± 0.13
1.14 ± 0.14
1.09 ± 0.25
1.04 ± 0.17
0.15 ± 0.07
0.25 ± 0.10
0.54 ± 0.19
0.85 ± 0.21
0.98 ± 0.20
1.02 ± 0.23
0.10 ± 0.10
0.39 ± 0.25
0.74 ± 0.32
0.99 ± 0.21
1.12 ± 0.17
1.07 ± 0.17
0.13 ± 0.06
0.24 ± 0.10
0.56 ± 0.11
0.94 ± 0.13
0.95 ± 0.17
0.92 ± 0.18
0.12 ± 0.06
0.22 ± 0.13
0.41 ± 0.21
0.65 ± 0.19
0.80 ± 0.21
0.81 ± 0.19
**
**
*
***
***
*
2
3
4
5
6
h
h
h
h
h
*
*
**
P < 0.05. P < 0.01.
***
P < 0.001 versus CMC-Na.

L. Fang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1062–1066
1065
Table 3. Effects of I
(
6
and Hyc on the activity of ALP in osteoblasts
ulate bone formation by increasing the proliferation of
osteoblasts as well as the activity of ALP. The reason
is perhaps that compared with osteoclasts, osteoblasts
are reportedly less sensitive to the change of NO concen-
tration, especially when the concentration of NO is not
n = 4, x ꢁ s)
Group (mol/L)
King’s U/100 ml
I
6
Hyc
DMSO
—
2.13 ± 0.26
15
high. However, in the osteoclast assays, Hyc appeared
to have no obvious effect on osteoclasts. Only at
ꢀ13
ꢀ12
ꢀ11
ꢀ10
ꢀ9
ꢀ
ꢀ
#
10
10
10
10
10
10
10
1.89 ± 0.08
2.24 ± 0.26
2.30 ± 0.08
1.70 ± 0.97
2.97 ± 0.23
2.23 ± 0.19
1.54 ± 0.28
2.07 ± 0.06
2.69 ± 0.10
2.68 ± 0.19
3.16 ± 0.51
2.66 ± 0.43
2.50 ±0.58
2.26 ± 0.19
#
**
*
ꢀ
7
#
10 mol/L concentration could Hyc inhibit the survival
rate of osteoclasts (P < 0.05), while I affected osteo-
#
*
6
ꢀ
13
ꢀ11
**
clasts potently. At 10 –10 mol/L concentrations,
I significantly increased the survival rate of osteoclast
cells cultured in vitro. However, at 10 –10 mol/L
8
6
7
*
ꢀ11
ꢀ7
*
**
P < 0.05. P < 0.01 versus DMSO. P < 0.05 versus Hyc.
#
concentration, I significantly inhibited the survival rate
6
of osteoclast cells cultured in vitro (P < 0.01). What’s
more, the higher the concentration, the stronger the
inhibition. This result may be due to the release of NO
and is consistent with the previous research of
1
6
Wimalawansa.
Hypertension is another adverse effect of GC. In our as-
says, I (92.9 mg/kg and 31.0 mg/mg), Hyc, and CMC-
6
Na were given to SD mice for 3 weeks. The blood pres-
sure was measured by the BESN system of multi-chan-
1
7
nel tail-artery blood pressure measurement. Then
plasma samples were collected and serum NO was deter-
mined by nitrate reductase assay (Table 4).
Compared with the CMC-Na group, the blood pressure
of the Hyc group was obviously stimulated to a higher
level, while that of I group did not vary significantly.
6
Figure 5. Effect of control vehicle, Hyc and I
6
at different concentra-
*
Furthermore, in the high-concentration group
(92.9 mg/kg) the blood pressure varied even less than
that in the low-concentration group (31.0 mg/kg). We
tions on numbers of TRAP (+) cell (n = 4, x ꢁ s). P < 0.05, **P < 0.01
versus control.
hypothesized that this activity of I is due to the release
6
of NO in vivo, so serum NO was measured. As the result
indicated, serum NO of I group was evidently higher
6
than that of either the Hyc group or the CMC-Na
group. And the higher the concentration of NO was,
the more potent the anti-hypertension activity was,
which was consistent with our hypothesis.
In summary, we have designed and synthesized a series
of furoxan-based NO-releasing GC derivatives, which
are of interest as a new class of improved GC-related
agents. Three of the derivatives maintain the anti-in-
flammatory activity and meanwhile compared with GC
the safety is greatly improved. By releasing NO in vivo,
the tested compound could effectively avoid stimulating
hypertension; moreover, it was also effective in prevent-
ing bone loss in the cell assays. Poor water solubility is
one of the disadvantages of the tested compound and
Figure 6. Effect of control vehicle, Hyc, and I
6
at different concentra-
tions on TRAP activity in osteoclasts (n = 4, x ꢁ s).
Table 4. Effect of I
6
on blood pressure (BP) and heart rate (HR) in rats (n = 10, x ꢁ s)
Dose (mg/kg)
CMC-Na
Hyc (43.4)
I
6
(92.9)
6
I (31.0)
*
*
*
*
##
##
*,#
#
SAP(mmHg)
DAP(mmHg)
HR
131.6 ± 10.5
89.5 ± 11.8
363.0 ± 30.6
27.5 ± 5.4
#
156.7 ± 8.5
109.6 ± 7.4
137.8 ± 11.8
144.5 ± 11.3
97.8 ± 14.5
360.3 ± 17.8
31.0 ± 2.7
86.9 ± 7.3
365.2 ± 24.9
28.0 ± 6.0
379.3 ± 22.9
*
40.9 ± 6.7
*,#
Serum NO (lmol/L)
**
P < 0.05. P < 0.01 versus Normal group. P < 0.05.
*
##
P < 0.01 versus Hyc group.

1066
L. Fang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1062–1066
further structural modification to improve the bioavail-
ability of the compound is in progress.
containing the culture medium. This was repeated three
times. The third passage osteoblasts were used in the later
measurement. To measure the proliferation of osteoblasts,
the osteoblasts were removed from the dish to be
4
incubated in 96-well plates at a density of 5 · 10 cells/
References and notes
6
ml for 24 h. Then I and Hyc at the concentrations of from
ꢀ
13
ꢀ7
1
0
to 10 mol/L were added to the medium; each
1
2
3
4
5
. Sch a¨ cke, H.; D o¨ cke, W. D.; Asadullah, K. Pharmacol.
Ther. 2002, 96, 23.
. Moncada, S.; Palmer, R. M.; Higgs, E. A. Pharmacol. Rev.
concentration repeated in 6 wells. One vehicle control
group was also performed. Forty-eight hours later 20 ll of
5 mg/ml MTT was added to the medium and the cells
continued be incubated for another 4 h. Then the medium
was removed and 100 ll DMSO was added to each well.
Then OD was measured at k = 490 nm. To measure the
1991, 43, 109.
. Wallace, J. L.; Ignarro, L. J.; Fiorucci, S. Nat. Rev. Drug
Disc. 2002, 1, 375.
. Fiorucci, S.; Santucci, L.; Gresele, P. Gastroenterology
activity of ALP, the osteoblasts were incubated in 24-well
5
2003, 124, 600.
plates at a density of 1 · 10 cells/ml for 24 h. I
6
and Hyc
to 10 mol/L were added
ꢀ
at concentrations of from 10
13
ꢀ7
. Loretta, L.; Barbara, R.; Marco, L. L.; Gian, C. T.;
Roberta, F.; Alberto, G.; Guido, D.; Harald, L. G. J.
Med. Chem. 2005, 48, 1322.
to the medium; each concentration repeated in 4 wells.
One vehicle control group was also performed. 48 h later
the medium was removed and 200 lL of 1% aqueous
solution of Triton X-100 was supplemented. The cells were
kept in a refrigerator at 4 ꢁC for 1 h. Thus the activity of
ALP was measured by the commercially available kits
using the method described by the manufacturer.
6
. Medana, C.; Di Stilo, A.; Visentin, S.; Fruttero, R.; Gasco,
A.; Ghigo, D.; Bosia, A. Pharm. Res. 1999, 16, 956.
. Pier, G. B.; Romeo, R.; Maria, C. N.; Mauro, P.; Mark, J.
P.; Massimiliano, F.; Mirco, G.; Francesca, B.; Ennio, O.
J. Med. Chem. 2004, 47, 711.
7
ꢀ
1
8
. Data for I
618vs, 1552vs. ESI-MS: m/z [M+Na] = 797. H NMR
400.1 MHz, CDCl ): d = 0.96 (s, 3H, C(19)–CH ), 1.00–
6
. mp = 69–71ꢁC. IR(m,cm ): 2937vs, 1730vs,
13. Garrett, R.I. WO 2003105752, 2003, 12–24.
+
1
1
14. Osteoclast assay: Extremities bones obtained from new-
born SD mice by separating mechanically were cultured in
MEM supplemented with 25 mmol/L Hepes, 10% fetal calf
serum, 100 U/ml penicillin, and 100 lg/ml phytomycin for
2 min. The bones were removed to 24-well plates cultured
(
3
3
1
.05 (m, 1H, C(9)–H), 1.10–1.26 (m, 2H, C(6)–H , C(15)–
b
H
H
b
), 1.44 (s, 3H, C(18)–CH
3
), 1.47–1.61 (m, 4H, C(12)–
), 1.66–1.81 (m, 2H,
, C(14)–H, C(15)–H , C(16)–H
b a
b
6
C(12)–H , C(1)–H ), 1.83–1.95 (m, 3H, C(1)–H , C(1)–H,
a
at a density of 1 · 10 cells/ml at 37 ꢁC in a humidified
b
a
C(6)–H
C(16)–H
2
3
C(28)–OCH
(
a
), 1.91–2.15 (m, 4H, C(7)–H, C(2)–H
), 2.16–2.27 (m, 2H, C(5)–H
.39 (m, 2H, C(24)–CH ), 2.42–2.60 (m, 2H, C(23)–CH
.47–3.52 (m, 1H, C(11)–H), 3.80 (t, J = 4.5 Hz, 2H,
), 3.92 (t, J = 4.5 Hz, 2H, C(29)–CH O), 4.30
t, J = 4.5 Hz, 2H, C(27)–CH O), 4.46 (s, 1H, C(11)–OH),
b
, C(2)–H
a
,
2
atmosphere of 5% CO for 1 h. The medium was changed
, C(5)–H
b
), 2.27–
),
and the bones were incubated for another 24 h. I
6
and Hyc
to 10 mol/L were added
a
a
ꢀ
at concentrations of from 10
13
ꢀ7
2
2
to the medium and at each concentration repeated in 4
wells. One vehicle control group was also incubated. To
measure the numbers of tartrate-resistant acid phosphatase
(+) (TRAP) cells, the cells were incubated with tested
compounds for 48 h. Cells were then stained with the
osteoclast-associated enzyme TRAP. Five milligrams of
naphthol AS-MX phosphate was resolved in 0.5 ml of
N,N-dimethylformamide, after which 30 mg of fast red
violet LB salt and 50 ml of 0.1 M sodium acetate buffer (pH
5.0) containing 50 mM sodium tartrate were added to the
TRAP-staining solution. The coverslips were fixed with
10% formalin for 10 min and then incubated with TRAP-
staining solution for 1 h at 37 ꢁC. The number of TRAP-
positive cells was counted to estimate the degree of
macrophageosteoclast differentiation.
2
2
2
4
1
.57 (t, J = 4.5 Hz, 2H, C(26)–OCH ), 4.85 (d, J = 9.0 Hz,
2
H, C(21)–H ), 5.06 (d, J = 9.0 Hz, 1H, C(21)–H ), 5.68
b
a
0
0
(
s, 1H, C(4)–H), 7.60–7.65 (m, 2H, C(3 )–H, C(5 )–H),
0 0
.73–7.76 (m, 1H, C(4 )–H], 8.05–8.08 (m, 2H, C(2 )–H,
7
C(6 )–H). Anal. Calcd for (C H N O S): C, H, N.
0
3
7
46
2
14
9
. Sorba, G.; Ermondi, G.; Fruttero, R.; Galli, U.; Gasco, A.
J. Heterocycl. Chem. 1996, 33, 327.
ꢀ1
1
0. Data for 7 . IR (m,cm ): 3390 versus, 1629 versus, 1547
7
+ 1
versus, 1020s. ESI-MS: m/z [M+H] = 349. H NMR
400.1 M, CDCl ): d = 1.62 (s, 1H, C–OH), 4.73 (s, 2H,
OCH ), 7.26–7.30 (m, 2H, C(3)–H, C(5)–H), 7.43–7.46 (m,
H, C(2)–H, C(6)–H), 7.62–7.65 (m, 2H, C(3 )–H, C(5 )–
(
3
2
0
0
2
0
0
H), 7.68–7.76 (m, 1H, C(4 )–H), 8.09–8.12 (m, 2H, C(2 )–H,
0
15. Das, U. N. Exp. Biol. Med. 2002, 227, 88.
C(6 )–H).
1. Wimalawansa, S.; Chapa, T.; Fang, L. J. Bone Miner. Res.
16. Wimalawansa, S. J.; De Marco, G.; Gangula, P.; Yallam-
palli, C. Bone 1996, 18, 301.
17. Blood pressure assay: 40 SD mice were divided into 4
1
1
2000, 15, 1119.
2. Osteoblast assay: Parietal bones were obtained from 3-
day-old SD mice and dissected free of adhering tissue.
After the treatment with the trypsin the bones were cut off
and cultured in DMEM supplemented with 10% fetal calf
serum. The incubation was conducted at 37 ꢁC in a
6 6
groups and administered I (92.9 mg/kg), I (31.0 mg/kg),
Hyc (43.4 mg/kg) and CMC-Na, respectively, for 3 weeks.
Then after 1 h since the last treatment of the drugs, the
blood pressure of each mouse was determined by the
BESN system of multi-channel tail-artery blood pressure
measurement. Plasma samples were then collected from
the femoral artery and serum NO (lmol/L) was measured
using a nitrate reductase assay.
2
humidified atmosphere of 5% CO for 24 h. Half of the
medium was changed once after 2 days. Proliferated
osteoblasts were harvested and transferred to a new dish