Bis-N-nitroso-caged nitric oxides: Photochemistry and biological performance test by rat aorta vasorelaxation

Article abstract of DOI:10.1016/S0968-0896(99)00084-X

Three new caged nitric oxides (NOs)-BNN3, BNN5Na, and BNN5M-were tested for biological use. BNNs have a strong ultraviolet (UV) absorption band (λ(max): 300 nm, ε: 13.5mM^-1cm^-1) extended to 420nm and produce NO upon irradiation with

Full text of DOI:10.1016/S0968-0896(99)00084-X

Bioorganic & Medicinal Chemistry 7 (1999) 1695±1702
Bis-N-nitroso-caged Nitric Oxides: Photochemistry and Biological
Performance Test by Rat Aorta Vasorelaxation
a
a
a
a
Shigeyuki Namiki, Fusako Kaneda, Masashi Ikegami, Tatsuo Arai,
a,
b
b
b
Ken Fujimori, * Sachie Asada, Hiroshi Hama, Yoshitoshi Kasuya
and Katsutoshi Goto b
aDepartment of Chemistry, University of Tsukuba, Tsukuba, Ibaraki, 305-0006 Japan
bDepartment of Pharmacology, University of Tsukuba, Tsukuba, Ibaraki, 305-0006 Japan
Received 27 July 1998; accepted 1 April 1999
AbstractÐThree new caged nitric oxides (NOs)ÐBNN3, BNN5Na, and BNN5MÐwere tested for biological use. BNNs have a
�
1
� 1
strong ultraviolet (UV) absorption band (lmax: 300 nm, e: 13.5 mM cm ) extended to 420 nm and produce NO upon irradiation
with 300±360 nm light in quantum yields about 2. A photoexcited BNN molecule yields two NOs with time constants of less than
ꢀ
1
0 ns for phase 1 and less than 20 ms for phase 2 at 37 C, suggesting usefulness of BNNs for measuring in vivo and in vitro fast NO
reactions. Upon irradiating with UV light, caged nitric oxides-loaded rat aortic strips maintained in a state of active tonic con-
traction e€ectively relaxed (<3 mM BNN5M loading solution concentration). BNN3 is incorporated in the lipid membrane.
BNN5Na, insoluble in organic solvents but water soluble, localizes in the water phase. BNN5M, is muscle-cell-permeable and
hydrolysed to BNN5Na to remain in cytosol. BNNs were thermally stable and demonstrated no observable toxicity. # 1999 Elsevier
Science Ltd. All rights reserved.
Introduction
precursors (caged compounds) is already generally used
9
in the biosciences. Although new types of photo-remo-
most
Nitric oxide (NO) plays diverse roles in mammals, e.g.
in blood pressure control, platelet aggregation inhibi-
tion, neurotransmission, immune regulation, and penile
erection. NO is also considered to relate to redox
signaling via S-nitrosothiol formation and decomposi-
vable protective groups are being developed,1
0±14
caged compounds thus far are o-nitrobenzyl esters of
bioactive target molecules.9
,15
1
±7
Ortho-nitrobenzyl caged NO (CNO1-4) developed by
15
8
tion. To explore these biological activities, it would be
Makings and Tsien was used to study the NO role as a
messenger initiating long-term depression in cere-
bellum,16 and long-term potentiation in hippocampus
desirable to create a high-performance caged NO that,
when light irradiated, generated desired NO at speci®c
target-tissue sites in a speci®c time. Photolytically gen-
erating bioactive molecules (uncaging) from photolabile
1
7
18
CA1 region and cultured hippocampal neurons.
20
Ru(NO)Cl519 and Rousin's salt, an iron±sulfur nitro-
21±23
syl cluster, were used similarly. S-nitrosothiols
4,25
and
were also reported to
2
-methyl-2-nitrosopropane2
produce NO upon irradiation with light. However,
0
0
Abbreviations: NO, nitric oxide; BNN3, N,N -dinitroso-N,N -dime-
0
0
thylphenylenediamine; BNN5, N,N -dinitrosophenylenediamine-N,N -
0
0
quantum yields of uncaging for these known caged NOs
26,27
diacetic acid; BNN5Na, N,N -dinitrosophenylenediamine-N,N -di-
0
acetic acid sodium salt; BNN5M, dimethyl N,N -dinitrosophenyl-
0
are quite poor. Both Rousin's salt
28
and S-nitro-
sothiols thermally produce NO without light, and
enediamine-N,N -diacetate; LFP, Laser ¯ash photolysis; mp, melting
point; FBS, fetal bovine serum; PMSF, phenylmethylsulfonyl ¯uoride;
DMEM, Dulbecco's modi®ed Eagle's medium; EGTA, ethyleneglycol-
further trace amounts of transition metal ion e€ectively
9±31
catalyze NO extrusion from S-nitrosothiols.2
biological use of these two is limited.
So, the
0
0
bis(b-aminoethyl ether)-N,N,N ,N -tetraacetic acid; PBS, phosphate
bu€ered saline solution; DMSO, dimethyl sulfoxide; TPPCo , meso-
II
tetraphenylporphinatocobalt(II); ODQ, 1H-[1,2,4]oxazolo[4,3-a]quin-
0
ozalin-1-one; carboxy-PTIO, 2-(4 -carboxyphenyl)-3,3,4,4-tetra-meth-
ylimidazoline-1-oxyl-3-oxide; PTIO, 2-phenyl-3,3,4,4-tetramethyl-
imidazoline-1-oxyl-3-oxide.
Thus, we synthesized three di€erent caged NOs (bis-N-
nitroso compounds) based on a new molecular design
0 0
Fig. 1), i.e. N,N -dinitroso-N,N -dimethylphenylenedi-
amine (BNN3), N,N -dinitrosophenylenediamine-N,N -
(
* Corresponding author. Tel.: +81-298-53-5241; fax: 81-298-53-5273;
e-mail: kfujimor@sta€.chem.tsukuba.ac.jp
0
0
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00084-X

1
696
S. Namiki et al. / Bioorg. Med. Chem. 7 (1999) 1695±1702
Figure 1. Uncaging of caged nitric oxides.
diacetic acid sodium salt (BNN5Na) and a BNN5
dimethyl ester (BNN5M). BNN3 is soluble in organic
Given that caged NOs would be used in living systems,
ꢀ
LFP experimental data at 37 C would be important. A
3
2
solvents but water-insoluble.
BNN5Na is water
500 mM BNN5M solution has been laser-photolyzed
with a 20 ns pulse width and 308 nm wavelength at 37 C.
3
2
ꢀ
soluble but insoluble in organic solvents. BNN5M
dissolves well in organic solvents and is slightly water-
soluble. Quantum yields of BNN3 and BNN5Na are
almost 2, i.e. absorption of one photon by one BNN3
or BNN5Na molecule yields two NO molecules through
an N-radical intermediate.32 BNNs are thus high-per-
formance in a chemical sense, but must be evaluated
biologically before general biological application can be
attempted.
Because BNN5M is not suciently for water-soluble
experiments, benzene solution was used. An absorbance
decay curve was observed at 400 nm due to intermediary
N-radicals (lmax 400 nm). Absorbance recorded during
the reaction from 5 to 50 ms after irradiaton ®ts ®rst-
order kinetics (simulation curve with rate constant,
5
� 1
kuc2 ˆ 1:74  10 s ; Fig. 2(A)). During the initial 5 ms,
N-radicals disappear, however, faster than the calcu-
lated rate based on the ®rst order decay with the above
Our purposes are to: (a) introduce a new BNN member,
N,N-dinitrosophenylenediamine-N,N-diacetate (BNN5M),
(
test caged NOs performance for biological applications
by rat aortic strips photorelaxation, and (d) detect the
localisation of BNNs compartment. Vasorelaxation
assay is suitable for testing caged NO performance,
because a large number of vasorelaxation experiments
with NO and NO-producing systems have been repor-
kuc2 indicating a second-order reaction of the N-radical
32
0
0
b) obtain kinetic data on LFP of BNNs at 37 C, (c)
and NO to regenerate BNN5M.
The N-radical
ꢀ
appeared within 10 ns after the laser ¯ash, indicating
that the excited state of BNN5M splits to yield the N-
radical intermediate and the ®rst NO molecule within
10 ns. The uncaging of BNN5M occurs with the tuc1
<10 ns for the ®rst NO-generating phase and
tuc2=5.7 ms for the second phase. The same experiments
ted in an attempt to identify the chemical feature of
endothelial-derived relaxant factor,1
±7,33
ever since
Furchgott and Zawadzki suggested its presence.34
Results
Stability of caged NO
BNN5Na was found to be stable in 100 mM phosphate
bu€er of pH 7.4 at 37 C for over 1 day in the dark.
ꢀ
When BNN5Na was incubated with 10 mM cysteine in
ꢀ
able reaction occurred. When benzene solutions of
phosphate bu€er of pH 7.4 at 37 C for 2 h, no notice-
ꢀ
BNN3 and BNN5M were kept at 37 C in the dark for
2
h, no noticeable reaction occurred.
Laser ¯ash photolysis (LFP) of caged NOs
tuc is an important photochemical constant in uncaging
kinetics, because tuc is the time constant for the uni-
molecular reaction of the photoexcited caged molecule
ꢀ
Figure 2. Laser ¯ash photolysis of BNN5M in benzene at 37 C. (A):
BNN5M alone; [BNN5M]=100 mM; absorbance at 400 nm of N-radi-
cal intermediate traced. (B): Mixture of BNN5M and TPPCo ;
with a ®rst-order rate constant, k , yielding the target
uc
II
bioactive molecule, i.e. the reciprocal of k . Uncaging
uc
II
[
BNN5M]=12.3 mM; [TPPCo ]=52.1 mM; absorbance at 546 nm of
TPPCo NO complex traced. Solid lines are theoretical ®rst-order
kinetic curves.
ꢀ
kinetics of BNN3 and BNN5Na at 23 C were pre-
II
viously studied using laser ¯ash photolysis (LFP).32

S. Namiki et al. / Bioorg. Med. Chem. 7 (1999) 1695±1702
1697
ꢀ
for BNN5Na in an aqueous bu€er solution at 37 C
showed that tuc1<10 ns for the ®rst phase and
tuc2=19 ms for the second phase. Similarly for BNN3,
tuc1<10 ns for the ®rst phase and tuc2=5.8 ms for the
ꢀ
second phase in benzene at 37 C.
To demonstrate the very small tuc values of BNN5M
and its usefulness in exploring fast NO reaction kinet-
ics, LFP of 12.9 mM BNN5M solution was conducted
in the presence of 65.8 mM mesotetraphenylporph-
II
inatocobalt(II) (TPPCo ), a model of soluble guanylate
ꢀ
at 546 nm is due to TPPCo NO (Fig. 2(B)). Experi-
cyclase (Fig. 3), in benzene at 37 C. Raising absorbance
II
mental points (Fig. 2) ®t well to the ®rst-order kinetic
equation to obtain a pseudo-®rst-order rate constant,
5
� 1
ꢀ
were conducted using di€erent TPPCo concentrations
1.32‹0.03Â10
s
, at 37 C. The same experiments
II
Figure 4. Plot of kobs of NO binding to TPPCoII against TPPCoII
concentration. LFP experiments were carried out with [BNN5M]=
(
(
40±95 mM) at a constant BNN5M concentration
12.9 mM). From the slope of the linear plot of pseudo-
II
ꢀ
12.9 mM and [TPPCo ]=40.1±95.1 mM in benzene under Ar at 37 C.
®
rst-order rate constants against TPPCoII concentra-
tions (Fig. 4), the second-order rate constant for com-
as the light-path shutter for uncaging was opened for
photons to reach the aorta strip, quick relaxation
occurred (Fig. 5). Reduction of the relaxation response
during repeated uncaging is presumably due to the con-
sumption of both incorporated caged NO and unknown
II
ꢀ
bination of TPPCo and NO in benzene at 37 C was
9
� 1 � 1
s .
found to be 1.99‹0.03Â10 M
Caged NO photovasorelaxation assay
photo-sensitive NO-donor inherently contained in
smooth muscle cells.3
6,42
Endothelial cells-denuded rat aortic strips were incu-
bated with one of the BNN reagents in an organ bath
The photoinduced vasorelaxa-
tion time course of the caged NO-loaded strips was
similar to that caused by adding aqueous NO solution
to the organ bath (data not shown). When the light-
path shutter was closed, tension returned to the steady
state initially induced by methoxamine (Fig. 5).
ꢀ
maintained at 37 C. After strips with BNN were loaded
into the organ bath, the organ bath solution was
replaced with fresh Krebs±Ringer's solution to remove
free BNN molecules. After repetition of the same
washing procedure, aortic strips were then irradiated
using 300 or 360 nm UV light while recording strip
tension. Furchgott et al. reported that UV light irradia-
tion induces relaxation of the rabbit aorta, maintained
Oxyhemoglobin, a typical NO quencher in biological
systems, and 1H-[1,2,4]oxazolo[4,3-a]quinoxalin-1-one
(ODQ),43 a guanylate cyclase inhibitor, inhibited the
BNN5Na-dependent photoinduced vasorelaxation (Fig.
5(B) and (C)). Results of these inhibition tests reveal
that the BNN5Na-dependent photoinduced vasorelaxa-
tion is caused by activating guanylate cyclase with NO
formed by photolytic uncaging of BNN5Na.
3
5±37
in a state of active tonic contraction.
Some photo-
sensitive NO donor seems to be inherently present in
smooth muscles, but mechanisms remain con-
3
8±42
troversial.
Taking this phenomenon into account,
carefully controlled experiments have been conducted
without using caged NOs, and irradiation of strips was
found to cause substantial relaxation (Table 1, entries
Under all assay conditions, photoinduced vasorelaxa-
tion of BNN reagent-loaded strips was reversible and
repeatable. Thus, at least for the contractile apparatus,
BNN reagents show no cytotoxicity.
1
and 6). Net BNN-induced photorelaxation was
obtained by subtracting relaxation value of control
experiments from that for BNN-loaded strips. As soon
Figure 3. Uncaging of BNN5M in heme-model presence.

1
698
S. Namiki et al. / Bioorg. Med. Chem. 7 (1999) 1695±1702
Table 1. Photorelaxation of caged NO-loaded rat aorta strips with 1 mM BNN for 30 min
Entry
1a
Caged NO
Wavelength (nm)c
Total % relaxationd
BNN-induced % relaxationa
f
28.3‹2.6e
49.6‹3.1
35.4‹6.0
60.6‹8.7
69.7‹2.5
46.8‹2.4e
60.0‹3.9
Control
BNN3
300
300
300
300
0
a
f
2
30.1‹9.0
10.4‹12.3
45.7‹17.4
58.1‹9.2
0
a
f
3
BNN5Na
BNN5Na
BNN5M
Control
BNN5M
b
f
4
a
f
5
a
300
360g
6
a
f
7
360
25.4‹13.0
a
Photolysis was conducted in BNN-free Krebs±Ringer's solution after washing strips.
Photolysis was conducted in Krebs±Ringer's solution containing 1 mM of BNN5Na.
Wavelength of uncaging light (light path slitwidth, ‹2 nm).
n=4.
b
c
d
e
f
n=6. Strips were not treated with caged NO.
0
.26 mW.
.45 mW.
g
0
In assay results for photoinduced vasorelaxation of the
three caged compounds (Table 1), except for entries 1,
BNN5Na present in the medium between the strip
and the light-delivering optical ®ber top pene-
trated tissue to induce vasorelaxation.44
4, and 6, strips were rinsed with caged NO-free Krebs±
Ringer's solution after loading of caged NOs. Data
demonstrate three points;
b. Since BNN3 and BNN5M induced vasorelaxation
upon uncaging even after strips were rinsed with
caged NO-free Krebs±Ringer's solution to remove
BNN5M from the medium (entries 2 and 3),
BNN3 and BNN5M were cell-membrane perme-
able.
c. Control vasorelaxation induced by 360 nm light is
1.7-fold greater than that by 300 nm light (entries 1
and 6). This is because 360 nm light is 1.9-fold
stronger than 300 nm light (see Experimental). The
net BNN5M dependent vasorelaxaton caused by
300 nm light is 23-fold greater than that by 360 nm
light (entries 5 and 7). Shorter wavelength light
(300 nm) is somewhat e€ective for uncaging caged
NO in rat aortic strips.
a. In irradiation of rat aortic strips in the organ
solution containing 1 mM BNN5Na, substantial
relaxation was observed (entry 4). However,
water-soluble BNN5Na did not cause vasorelaxa-
tion, when strips were rinsed with BNN5Na-free
Krebs±Ringer's solution after treatment with
BNN5Na (entry 3). Combining these two and the
previous observation that the BNN5Na-dependent
photoinduced vasorelaxation is inhibited by haem-
oglobin, one can say that BNN5Na does not
penetrate smooth muscle cells through the plasma
membrane but NO formed by the uncaging of
Fixing the BNN5M concentration for the loading solu-
tion at 1 mM, we studied the e€ect of loading time on
photorelaxation. Results (Fig. 6) showed that reagent
incorporation into the cell proceeds very e€ectively. As
Figure 5. Some selected traces. (A) Typical photoinduced BNN5M-
loaded rat aorta vasorelaxation ([BNN5M]: 1 mM; loading time:
3
0 min). After BNN5M doping, the strips were rinsed by displacing it
twice in an organ bath solution with BNN5M-free Krebs±Ringer's
solution. (B), (C) Inhibition of photoinduced vasorelaxation in 1 mM
BNN5Na-containing Krebs±Ringer's solution with (B) ODQ; (C)
Figure 6. E€ect of BNN5M-loading time on rat aortic strip photo-
relaxation. Incubation solution contained 1 mM BNN5M. After incu-
bation, the strips were rinsed with fresh Krebs±Ringer's solution and
irradiated with 300 nm wavelength light. n=4. Error bars: SEM.
5
2
0 mM oxyhemoglobin (HbO ).

S. Namiki et al. / Bioorg. Med. Chem. 7 (1999) 1695±1702
1699
can be seen in the BNN5M concentration±vasorelaxa-
tion relationship (Fig. 7), strips treated with even
each caged compound. e is the molar absorption coe-
cient of the caged compound at the wavelength of
uncaging light, and Èuc is the quantum yield of unca-
ging. eÈ value is thus a parameter representing the
relative amount of bioactive molecules generated by
uncaging with light of a unit strength from unit
amounts of caged compounds at a unit time interval.
tuc is the time constant for the uncaging process of
the photoexcited caged molecule yielding the target
bioactive molecule. Thus, a chemically high-perfor-
mance caged compound possesses a large eÈuc and a
small tuc.
100 nM of BNN5M underwent 25% relaxation. The
relaxation response increased concentration-depen-
dently, and peaked (ca. 80%) to level o€ at about 3 mM
BNN5M (Fig. 7).
BNN5M hydrolysis with muscle cells
Incubation of 0.75 mL solution containing 9.4 mM
BNN5M and cytosolic fraction obtained from ca.
6
ꢀ
5
Â10 cultured muscle cells for 20 min at 37 C resulted
in the hydrolysis of BNN5M to yield 5.5 mM BNN5Na.
The result encouraged us to examine the possibility that
BNN5M may enter cells and be hydrolyzed to yield
BNN5Na (Fig. 8). A10 cells were incubated in Dulbec-
In addition to these chemical requirements, high-per-
formance caged compounds, which are useful in bios-
cienti®c application, should have the following
biological properties: (a) the caged compound is itself
bioinactive and (b) the caged compound is localized at
the speci®c compartment at which the researcher wants
to generate target bioactive molecules.
co's modi®ed Eagle's medium (DMEM) containing
ꢀ
100 mM BNN5M at 37 C. After incubation at di€erent
time intervals, the cytosolic fraction was isolated and
the BNN5Na content was measured (Fig. 9). The
BNN5Na concentration in the cytosolic fraction
increases with incubation time until about 40 min, then
"
Èuc and vasorelaxation
4
5
plateaued.
Three BNN reagents have a strong UV absorption band
extended to 425 nm wavelength (lmax: 292±300 nm, e:
�
1
� 1
cm ) and the chemical yields of
NO in uncaging of both BNN5Na in water and BNN3
1
3,500±13,700 M
Discussion
The chemical quality of a caged compound is expressed
by physicochemical constants, eÈuc and t , speci®c to
uc
Figure 7. In¯uence of BNN5M loading solution concentration on rat
ꢀ
Figure 9. Time course for BNN5Na formation in smooth muscle cell
cytosol fraction during BNN5M incubation with cultured smooth
muscle cells. n=4. Error bars: SEM.
aorta strip relaxation (n=4; loading time: 30 min at 37 C). Error bars:
SEM.
Figure 8. Enzymatic BNN5M hydrolysis.

1
700
S. Namiki et al. / Bioorg. Med. Chem. 7 (1999) 1695±1702
in organic solvents are quantitative. When 2-(4-carboxy-
phenyl)-3,3,4,4-tetramethylimidazoline-1-oxyl-3-oxide
proceeds with the second-order rate constant on the
9
� 1 � 1 32
s .
order of 10 M
However, 3 mM is maximum con-
(
carboxy-PTIO) and 2-phenyl-3,3,4,4-tetramethylimid-
centration required in vasorelaxation assay, so the mM
level of BNN appears sucient in biological applica-
tions. Under such conditions the bimolecular process of
the recombination of the N-radical intermediate and
NO cannot compete with the unimolecular process of
the second NO extrusion reaction of the N-radical.
azoline-1-oxyl-3-oxide (PTIO) are used as NO sca-
vengers, the quantum yield for BNN5Na uncaging in
water is 1.87 and that for BNN3 in organic solvents is 2,
i.e. a single photon produces 1.6±2 NO molecules from
one BNN molecule. Values of Èuc for already known
caged NOs are rather poor: 0.02±0.05 for o-nitro-
3
2
1
5
benzyl ester caged NOs (CNO-1-4) and 0.05 for
1
Localization in aortic tissue
9
4
� 1
K Ru(NO)Cl . The eÈuc of BNN5Na is 2.7Â10 M
2
5
�
1
3
� 1
� 1
cm
3
with 300 nm light and 6Â10 M cm with
50 nm light; this is much greater than for previously
A third BNN advantage is that three di€erent BNN
solubilities are available. (a) BNN5Na is only soluble in
aqueous compartments and is membrane-impermeant
(Table 1, entries 3 and 4). If BNN5Na is introduced into
an outer cellular aqueous medium, the reagent remains
there and does not permeate into the cell. (b) BNN3 is
incorporated into vascular smooth muscle cells. Since
BNN3 is water-insoluble, it localizes at lipids in cells. (c)
BNN5M is also membrane-permeant, but it is only
slightly water-soluble and is hydrolysed by cytosolic
esterases to BNN5Na, which is membrane-impermeant
and hence remains in the cell.
�
1
� 1
reported CNOs (eÈ : 15±70 M cm with 360‹20 nm
light).
uc
1
5
The large eÈuc of BNNs is consistent with photo-
vasorelaxation assay results. The irradiaton of rat aortic
tissues loaded with BNN3 or BNN5M with only nM
level concentrations for 10 min resulted in NO release
sucient to cause a physiological response in the tissue
(
Figs. 6 and 7). Further, the photorelaxation response
of rat aortic strips was found to attain its maximum
with only 3 mM BNN5M loading solution (Fig. 7).
Either less of BNN or a weaker uncaging light source
may deliver NO at the target tissue, compared to caged
NOs having smaller eÈuc.
Conclusion
Choosing a suitable BNN and uncaging light source of
3
00±360 nm, the NO can be delivered at a precise time
ꢀuc
and space resolutions at a speci®c compartment in a
living system. At least for the contractile apparatus and
in hydrolysis experiments of 100 mM BNN5M using
cultured smooth muscle cells, we observed no noticeable
BNN cytotoxicity.
When a caged compound is used to explore very fast
biological events, the researcher should use a caged
compound having a very small tuc value. In the stepwise
release of two NO molecules from a photoexcited BNN
molecule, the ®rst tuc for the three BNNs is less than
1
0 ns and the second is 5.6 ms for BNN3 and 5.8 ms for
Experimental
Materials
BNN5M in benzene, and 19 ms for BNN5Na in water at
ꢀ
than that of o-nitrobenzyl ester caged compounds,
3
7 C. These t are several orders of magnitude smaller
uc1
4
6,47
which are usually at ms±ms level.
Caged NOs having
BNN3 and BNN5 were prepared as described else-
where.32 BNN5Na was prepared by treating BNN5
methanol solution with two equimolar amounts of 2 M
NaOH aqueous solution. Crystals liberated were col-
lected and dissolved in a minimum volume of water.
Pure pale yellow BNNNa crystals were precipitated by
a very small tuc are useful for kinetically investigating
very fast NO events in biological systems, e.g. NO
di€usion kinetics in tissue; sudden NO delivery for the
time-resolved X-ray crystallographic analysis of NO-
requiring enzymes such as soluble guanylate cyclase,
cytochrome P-450nor49 and nitrilehydratase;
determination of second-order rate constants for NO-
binding to metalloenzymes and its models, and kinetics
of biological NO signaling. This was exempli®ed by the
successful determination of the second-order rate con-
48
50
1
the
adding methanol to the solution. BNN5Na: H NMR
(D O, 200 MHz, in d): 7.77 (s, 4H, aromatic), 4.69 (s,
2
�
1
4H, CH ), UV: lmax, 300 nm (e: 13,500 M cm ) in
2
0.1 M sodium phosphate bu€er, pH 7.4. BNN5M was
prepared by slowly adding a diazomethane ether solu-
tion to a BNN5 methanol solution until yellow diazo-
methane no longer faded at room temperature. Solvents
were evaporated to dryness in a vacuum, leaving a pale
yellow solid that, upon alumina chromatography (elu-
ent: dichloromethane) and recrystallization from a mix-
ture of dichloromethane and ethyl acetate yielded pale
II
stant for the coordination of NO with TPPCo and NO
II
by LFP of a solution of TPPCo and BNN5M (Figs. 1
� 1 � 1
ꢀ
in benzene at 37 C,
9
and 2), i.e. 1.99‹0.03Â10 M
s
revealing that BNNs are promising for use in studying
fast NO reactions in vivo and in vitro.
ꢀ
An experimental result shown in Fig. 2(A) for LFP of
yellow pure BNN5M crystals. BNN5M mp.144±145 c
(decomp.), Anal. calcd. for C H N O : C, 46.45; H,
100 mM BNN5M solution was obtained by using rather
12
14
4
6
1
a higher concentration than that applied in actual
biological chemistry. As mentioned above, 3 mM is
maximum concentration in vasorelaxation assay. The
second-order rate constant for the combination of the
N-radical intermediate and NO to reproduce BNN5M
4.52; N, 18.06, found C, 46.36; H, 4.54; N, 17.85, H
NMR (CDCl , 200 MHz, in d):7.27 (s, 4H, aromatic),
3
4.71 (s, 4H, CH ), 3.81 (s, 6H, -OCH ); IR (nujol, in
2
3
�
1
cm ): 1738, 1143, 1116, 1520. UV: lmax, 292 nm (e:
13,700 M cm ) in methanol.
�
1

S. Namiki et al. / Bioorg. Med. Chem. 7 (1999) 1695±1702
1701
Methoxamine, fetal bovine serum (FBS), DMEM,
ethyleneglycol-bis(b-aminoethyl ether)-N,N,N ,N -tetra-
for 20 s. The solution was diluted with 10% aqueous
Cremophor-EL solution so that, when 40 mL of the
solution was added to a 20 mL Krebs±Ringer's solution
organ bath containing aorta strips, the ®nal BNN5M
concentration became the desired value. BNN5M was
loaded by adding 40 mL of the BNN5 solution to the
organ bath and equilibrated for predetermined time
intervals. After washing two times with normal Krebs±
Ringer's solution, 10 mM methoxamine was applied to
strips, which caused tension development that plateaued
within 20 min. Tissue was irradiated with 300 or 360 nm
light introduced in the solution through a 0.8 mm quartz
optical ®ber from a JASCO high-power monochrome
light source equipped with a 500 W xenon lamp and a
light-path shutter. The light source power was 0.26 mW
at 300 nm and 045 mW at 360 nm from the top of the
®ber, which was continuously moved by hand so that
isostrength light reached all tissue. BNN3 and BNN5Na
were assayed in the same way as BNN5M; while a
BNN5Na solution was prepared by dissolving BNN5Na
in water.
0
acetic acid (EGTA), phosphate bu€ered saline solution
0
(
PBS), fetal bovine serum (FBS), and phenyl-
methylsulfonyl ¯uoride (PMSF) were purchased from
Sigma, St. Louis, MO and Cremophore-EL from
Aldrich, Milwaukee, WI. Acetylcholine and other spe-
cial grade reagents were obtained from Wako Pure
Chemicals, Osaka, Japan.
Krebs±Ringer's solution consisted of (in mM): NaCl
(
113); KCl (4.8); NaHCO (25); KH PO (1.4) CaCl
3 2 4 2
(
2.2); MgSO 7H O (1.4); and glucose (5.5). Depolariz-
2
4
ing KCl solution consisted of (in mM): NaCl (63.2);
KCl (48); NaHCO3 (25); KH PO (1.4) CaCl (2.2);
MgSO 7H O (1.4). Homogenization bu€er solution for
2
4
2
4
2
cultured A10 cells consisted of (in mM): Tris±HCl buf-
fer of pH 7.5 (20); sucrose (250); EGTA (0.5); PMSF
(
1); 20 mg/mL aprotinin.
Oxyhemoglobin freshly isolated from human blood was
a generous gift from Professor Yasuhiko Yamamoto,
University of Tsukuba.
BNN5M Hydrolysis assay in cultured smooth muscle cells
A10 cells were cultured in DMEM with 10% FBS,
washed two times with fresh DMEM, and then 10 mL
of DMEM and 100 mL of BNN5M/DMSO solution
([BNN5M]=10 mM) were added to attain a ®nal
BNN5M concentration of 100 mM. The mixture was
Laser ¯ash photolysis (LFP)
A BNN solution in 10Â10 mm UV cuvette with a three
way cock was irradiated with a 308 nm wavelength laser
pulse having a 20 ns width from an excimer laser
ꢀ
(
Lambda Physik LPX) under argon. Absorbance time
incubated at 37 C. After incubation, cells were rinsed
courses of transient species in the UV±VIS region
formed by laser ¯ash photolysis were measured using a
system consisting of a xenon ¯ash lamp (Wacom KXL-
151, 150 W), a photomultiplyer (Hamamatsu Photonics
R 928), and a storage scope (Iwatsu TS-8123).
two times with ice-cold PBS until BNN5M was no
longer detected in the culture solution, and then lysed in
500 mL of homogenization bu€er and homogenized at
ꢀ
4 C. After centrifugation at 150,000 g for 30 min at
ꢀ
4 C, the supernatant (cytosolic fraction) was ultra-
®ltered using a Milipore Ultrafree-MC 30,000 NMWL
Filter Unit. The ®ltrate was freeze-dried to an appro-
priate volume and subjected to HPLC analysis. The
HPLC facility consisted of a Hitachi L 6200 intelligent
pump, TOSO ODS 80Ts column (4.6Â250 mm), and a
Hitachi L4200 UV±VIS detector. The eluent consisted
of (volume %) water (70), methanol (30), and phos-
phoric acid (0.1). In the course of enzymatic hydrolysis
of BNN5M, a partially hydrolysed product, monomethyl
ester of BNN5, may be present. Since we could not
prepare the authentic sample of the half ester, we did
not observe measurable peak at longer retention time
region of the liquid chromatogram. If the rate deter-
mining step of the conversion of BNN5M into
BNN5Na in smooth muscle cells is BNN5M hydrolysis
to the monoester, the half ester intermediate level is low
in cytoplasm.
Preparation of rat aortic strips and tension readings
Aortic segments were obtained from 12-week-old male
Wister rats (400±450 g). After washing inside with
Krebs±Ringer solution, segments were cut into
3
Â13 mm helical strips. The endothelium was denuded
by gentle rubbing with small swabs, and the removal
was con®rmed by loss of a relaxant response to acet-
ylcholine. Arterial strips were placed by means of thread
in a jacketed drop-away organ bath containing 20 mL
Krebs±Ringer's solution maintained at 37 C and aerated
with 95% O ±5%CO . Arterial strips were equilibrated
ꢀ
2
2
at a passive tension of 0.4 g until the contractile
+
51
response caused by 50 mM K became a steady state.
The upper thread of each strip was connected to a force-
displacement transducer (Nihon Kohden TB-611T),
while the lower thread was anchored at the bottom.
Contractile responses were recorded using a thermal pen
recorder (Nihon Kohden WT-647G).
Acknowledgements
We are grateful to Professor Yasuhiko Yamamoto for
supplying us with human haemoglobin.
Caged NO assay
All procedures were done in a dark room under red
light. BNN5M solution (1 mM) was prepared by adding
a 2.33 mg BNN5M solution in 50 mL of dimethyl sulf-
oxide (DMSO) to 1450 mL of 10% aqueous Cremophor-
EL solution followed by gentle ultrasound sonication
References and Notes
1. Furchgott, R. F. In Vasodilation; Vanhoute, P. D., Ed.;
Paven Press: New York, 1988; pp 401±414.

1
702
S. Namiki et al. / Bioorg. Med. Chem. 7 (1999) 1695±1702
2
3
. Ignarro, L. J. Circulation Res. 1989, 65, 1±21.
. Moncada, S.; Palmer, R. M. J.; Higgs, E. A. Pharmacol.
Rev. 1991, 43, 109±142.
31. Singh, R. J.; Hogg, K.; Joseph, J.; Kalyanaraman, B. J.
Biol. Chem. 1996, 271, 18596±18603.
32. Namiki, S.; Arai, T.; Fujimori, K. J. Am. Chem. Soc. 1997,
119, 3840±3841.
33. Feelisch, M.; te Poel, M.; Zamora, R.; Deussen, K.; Mon-
cada, S. Nature 1994, 368, 26±38.
4. Stamler, J. S.; Singel, D. J.; Loscalzo, J. Science 1992, 258,
1898±1902.
5. Feldman, P. F., Grith, O. W.; Stuehr, D. J. Chem. Ind.
News, 1993, 20 December, 26±38.
34. Furchgott, R. F.; Zawadzki, J. Nature, 1980, 288, 373±376.
35. Furchgott, R. F.; Sleator, W. Jr.; McCaman, M. W.;
Elchelepp, J. J. Pharmacol. Exp. Ther. 1955, 113, 22±23.
36. Furchgott, R. F.; Ehrreich, S. J.; Greenblatt, E. J. Gen.
Physiol. 1961, 44, 499±519.
6
2
7
8
9
7
1
. Butler, A. R.; Williams, D. H. R. Chem. Soc. Rev., 1993,
33±241
. Rand, M. J.; Li, C. G. Ann. Rev. Physiol. 1995, 57, 659±682.
. Stamler, J. S. Cell 1994, 78, 931±936.
. Adams, S. R.; Tsien, R. Y. Ann. Rev. Physiol. 1993, 55, 755±
84.
37. Ehrreich, S. J.; Furchgott, R. F. Nature 1968, 218, 684±
685.
0. Furuta, T.; Torigai, K.; Sugimoto, M.; Iwamura, M. J.
38. Steg, P. G.; Rongione, A. J.; Gal, D.; DeJesus, S. T.;
Clark, R. H.; Isner, M. Circulation 1989, 80, 189±197.
39. Chang, K. C.; Chong, W. S.; Park, B. K.; Seung, B. K.;
Chun, G. K.; Lee, I. J.; Park, T. S. Biochem. Biophys. Res.
Comm. 1993, 191, 509±510.
40. Goud, C.; Watts, S. K.; Webb, R. C. J. Vasc. Res. 1996,
33, 299±307.
41. Chaudhry, K.; Lynch, M.; Schomaker, K.; Birngruber, R.;
Gregory, K.; Kochevar, I. Photochem. Photobiol. 1993, 58,
661±669.
42. Venturini, C. M.; Palmer, R. M. J.; Moncada, S. J. Phar-
macol. Exp. Ther. 1993, 266, 1497±1500.
43. Garthwaite, J.; Southam, E.; Boulton, C. L.; Nielsen, E.
B.; Schmidt, K.; Mayer, B. Mol. Pharmacol. 1995, 48, 184±
188.
Org. Chem. 1995, 60, 3953±3956.
1. Okada, S.; Yamashita, S.; Furuta, T.; Iwamura, M. Pho-
tochem. Photobiol. 1993, 61, 431±434.
2. Ramesh, D.; Wiebolt, R.; Niu, L.; Carpenter, B. K.; Hess,
G. P. Proc. Natl. Acad. Sci. USA, 1993, 90, 11074±11078.
3. Niu, L.; Ramesh, D.; Wiebolt, R.; Niu, L.; Carpenter, B.
K.; Hess, G. P. Biochemistry 1996, 35, 8136±8149.
1
1
1
1
2
1
6
1
4. Park, C.-H.; Givens, R. S. J. Am. Chem. Soc. 1997, 119,
435±2463.
5. Makings, L. R.; Tsien, R. Y. J. Biol. Chem. 1994, 269,
282±6285.
6. Lev-Ram, V.; Makings, L. R.; Keitz, P. F.; Kao, J. P. Y.;
Tsien, R. Neuron 1995, 15, 40±415.
7. Son, H.; Hawkins, R. D.; Martin, K.; Kiebler, M.; Huang,
P. L.; Fishman, M. C.; Kandel, E. R. Cell, 1996, 87, 1015±1023.
8. Arancio, O.; Kiebler, M.; Lee, C. J.; Lev-Ram, V.; Tsien,
R.; Kandel, E. R.; Hawkins, E. R. Cell 1996, 87, 1025±1035.
9. Murphy, K. P.; Williams, S. J.; Bettache, N.; Bliss, T. V. P.
Neuropharmacol. 1994, 33, 1375±1385.
0. Black, M. D.; Matthews, E. K.; Humphrey, P. P. A. Neu-
ropharmacol. 1994, 33, 1357±1365.
1. Boulton, C. L.; Irving, A. J.; Collingridge, G. L.;
Garthwaite, J. J. Phyisiol. 1993, 467, 181 p.
2. Sexton, D. J.; Muruganandam, K.; McKenney, D. J.;
Mutus, B. Photochem. Photobiol. 1994, 59, 463±467.
1
44. One of the referees pointed out the possibility that
BNN5Na enters the smooth muscle cells during the loading
but is simply washed out of the tissue during the Krebs±Ringer
rinse. The fact that BNN5Na formed by hydrolysis of
BNN5M in cells remained in cells after washing with fresh
Krebs±Ringer solution (Fig. 9) clearly demonstrates that
BNN5Na in cells cannot be washed out of cells with Krebs±
Ringer rinse.
1
1
2
2
3
45. Assuming the single cell volume to be 3000 mm , con-
centration of BNN5 formed from BNN5M incorporated into
2
cells were estimated to be about 12 mM.
2
1
2
3. Singh, R. J.; Hogg, K.; Kalyanaraman, B. FEBS Lett.
995, 360, 47±51.
4. Makino, K.; Suzuki, K.; Moriya, F.; Rokushika, S.;
46. Barth, K.; Corrie, J. E. T.; Cradwell, M. J.; Maeda, Y.;
Mantele, W.; Meiner, T.; Trentham, D. R. J. Am. Chem. Soc.
1997, 119, 4149±4159.
Hatano, H. Radiat. Res. 1981, 86, 294±310.
5. Pou, S.; Anderson, D. E.; Surichamorn, K.; Keaton, L. L.;
Tod, M. L. Mol. Pharmacol. 1994, 46, 709±715.
6. Flitney, F. K.; Megson, I. L.; Flitney, D. E.; Butler, A. R.
Br. J. Pharmacol. 1992, 107, 842±848.
7. Matthews, E. K.; Seaton, E. D.; Forsyth, M. J.; Hum-
phrey, P. P. A. Br. J. Pharmacol. 1994, 113, 87±94.
47. Cohen, B. E.; Stoddard, B. L.; Koshland, D. E. Jr. Bio-
chemistry 1997, 36, 9035±9044.
48. Stone, J. R.; Marletta, M. A. Biochemistry 1994, 33, 5636±
5640.
49. Shiro, Y.; Fujii, M.; Lizuka, T.; Adachi, S.; Tsukamoto,
K.; Nakahara, K.; Shoun, H. J. Biol. Chem. 1995, 270, 1617±
1623.
2
2
2
2
1
2
8. Komiyama, T.; Fujimori, K. Bioorg. Med. Chem. Lett.
997, 7, 175±180.
9. Barnett, D. J., McAninly, J., Williams, D. L. H. J. Chem.
50. Okada, M.; Fujii, K.; Hoshino, M.; Noguchi, T.; Tsuji-
mura, M.; Nagashima, S.; Yohda, M.; Nagamune, T.; Inoue,
Y.; Endo, I. J. Am. Chem. Soc. 1997, 119, 3785±3791.
51. Kasuya, Y.; Ishikawa, T.; Yanagisawa, M.; Kimura, S.;
Goto, K.; Masaki, T. Am. J. Physol. 1989, 257, H1328±1335.
Soc. Perkin Trans. 2, 1994, 1131.
0. Barnett, D. J., Rios, A., Williams, D. L. H. J. Chem. Soc.
Perkin Trans. 2, 1995, 1279±1282.
3