Controlled photochemical release of nitric oxide from O2-benzyl-substituted diazeniumdiolates

Article abstract of DOI:10.1021/ja026900s

An investigation of potential photosensitive protecting groups for diazeniumdiolates (R₂N-N(O)=NO-) has been initiated, and here the effect of meta electron-donating groups on the photochemistry of O²-benzyl-substituted diazeniumdiolates (R₂N-N(O)=NOCH₂Ar) is reported. Photolysis of the parent benzyl derivative (Ar = Ph) results almost exclusively in undesired photochemistry-the formation of nitrosamine and an oxynitrene intermediate with very little, if any, photorelease of the diazeniumdiolate. We have been able to use meta substitution to tune the photochemistry of these benzylic systems. The desired diazeniumdiolate photorelease has been shown to become more substantial with stronger π-donating meta substituents. This effect has been verified by direct observation of the photoreleased diazeniumdiolate with ¹H NMR spectroscopy and by NO quantification measurements conducted in high- and low-ph solutions. In addition, the observed rates of NO release are consistent with that expected for normal thermal decomposition of the diazeniumdiolate in aqueous solutions and also show the same pH dependence.

Full text of DOI:10.1021/ja026900s

Published on Web 07/30/2002
Controlled Photochemical Release of Nitric Oxide from
O2-Benzyl-Substituted Diazeniumdiolates
Patrick H. Ruane, K. Mani Bushan, Christopher M. Pavlos, Raechelle A. D’Sa, and
John P. Toscano*
Contribution from the Department of Chemistry, Johns Hopkins UniVersity,
3400 North Charles Street, Baltimore, Maryland 21218
Received May 14, 2002
Abstract: An investigation of potential photosensitive protecting groups for diazeniumdiolates (R2NsN(O)dNO-)
has been initiated, and here the effect of meta electron-donating groups on the photochemistry of O2-
benzyl-substituted diazeniumdiolates (R2NsN(O)dNOCH2Ar) is reported. Photolysis of the parent benzyl
derivative (Ar ) Ph) results almost exclusively in undesired photochemistrysthe formation of nitrosamine
and an oxynitrene intermediate with very little, if any, photorelease of the diazeniumdiolate. We have been
able to use meta substitution to tune the photochemistry of these benzylic systems. The desired
diazeniumdiolate photorelease has been shown to become more substantial with stronger π-donating meta
substituents. This effect has been verified by direct observation of the photoreleased diazeniumdiolate
with 1H NMR spectroscopy and by NO quantification measurements conducted in high- and low-pH solutions.
In addition, the observed rates of NO release are consistent with that expected for normal thermal
decomposition of the diazeniumdiolate in aqueous solutions and also show the same pH dependence.
Scheme 1
Introduction
Compounds containing the diazeniumdiolate [N(O)dNO]-
functional group have proven useful as research tools in a variety
of applications requiring spontaneous release of the critical
bioregulatory molecule nitric oxide (NO).1 Anions such as
1-(N,N-dialkylamino)diazen-1-ium-1,2-diolates 1 are stable as
solid salts, but release up to 2 mol of NO when dissolved in
aqueous solution at physiologically relevant conditions (Scheme
1). Keefer and co-workers have developed anions 1 with half-
lives in aqueous buffer at pH 7.4 and 37 °C ranging from 2 s
to 20 h.2 These compounds have shown great potential in a
variety of medical applications3 requiring the rapid production
(e.g., to produce a fast but transient drop in blood pressure4) or
gradual release (e.g., to study the effects of prolonged cyto-
stasis on vascular smooth muscle cells5) of NO.
Recent efforts to make diazeniumdiolates more effective
pharmaceuticals have concentrated on using derivatives of such
compounds to deliver NO specifically to a targeted site.6 Given
the large number of biological phenomena now known to be
mediated by NO, such targeting will be important to the ultimate
success of most medical applications. One way to achieve
targeted release of NO is through the use of photochemistry.
Photosensitive precursors, usually called “caged compounds”
or “phototriggers”, have become important biomedical research
tools.7 Several direct photochemical precursors to NO (e.g.,
metal nitrosyl compounds,8 nitrosothiols,9 and bis-N-nitroso
* To whom correspondence should be addressed. Fax: (410) 516-8420.
E-mail: jtoscano@jhu.edu.
(1) For a recent review of the chemistry of diazeniumdiolate derivatives, see:
Hrabie, J. A.; Keefer, L. K. Chem. ReV. 2002, 102, 1135-1154.
(2) (a) Maragos, C. M.; Morley, D.; Wink, D. A.; Dunams, T. M.; Saavedra,
J. E.; Hoffman, A.; Bove, A. A.; Isaac, L.; Hrabie, J. A.; Keefer, L. K. J.
Med. Chem. 1991, 34, 3242-3247. (b) Keefer, L. K.; Nims, R. W.; Davies,
K. M.; Wink, D. A. Methods Enzymol. 1996, 268, 281-293.
(5) Mooradian, D. L.; Hutsell, T. C.; Keefer, L. K. J. CardioVasc. Pharmacol.
1995, 25, 674-678.
(3) (a) Morley, D.; Keefer, L. K. J. CardioVasc. Pharmacol. 1993, 22, S3-
S7. (b) Hanson, S. R.; Hutsell, T. C.; Keefer, L. K.; Mooradian, D. L.;
Smith, D. J. AdV. Pharmacol. (San Diego) 1995, 6, 383-398. (c) Saavedra,
J. E.; Southan, G. J.; Davies, K. M.; Lundell, A.; Markou, C.; Hanson, S.
R.; Adrie, C.; Hurford, W. E.; Zapol, W. M.; Keefer, L. K. J. Med. Chem.
1996, 39, 4361-4365. (d) Pulfer, S. K.; Ott, D.; Smith, D. J. J. Biomed.
Mater. Res. 1997, 37, 182-189. (e) Raulli, R. J. Pharm. Pharmacol. 1998,
50, 75-82. (f) Smith, D. J.; Simmons, M. L. J. Controlled Release 1998,
51, 153-159. (g) Jacobs, B. R.; Brilli, R. J.; Ballard, E. T.; Passerini, D.
J.; Smith, D. J. Am. J. Respir. Crit. Care Med. 1998, 158, 1536-1542. (h)
Fernandez-Tome, P.; Lizasoain, I.; Leza, J. C.; Lorenzo, P.; Mora, M. A.
Neuropharmacology 1999, 38, 1309-1315. (i) Saavedra, J. E.; Shami, P.
J.; Wang, L. Y.; Davies, K. M.; Booth, M. N.; Citro, M. L.; Keefer, L. K.
J. Med. Chem. 2000, 43, 261-269. (j) Fitzhugh, A. L.; Keefer, L. K. Free
Radical Biol. Med. 2000, 28, 1463-1469.
(6) (a) Saavedra, J. E.; Dunams, T. M.; Flippen-Anderson, J. L.; Keefer, L. K.
J. Org. Chem. 1992, 57, 6134-6138. (b) Smith, D. J.; Chakravarthy, D.;
Pulfer, S.; Simmons, M. L.; Hrabie, J. A.; Citro, M. L.; Saavedra, J. E.;
Davies, K. M.; Hutsell, T. C.; Mooradian, D. L.; Hanson, S. L.; Keefer, L.
K. J. Med. Chem. 1996, 39, 1148-1156. (c) Saavedra, J. E.; Billiar, T. R.;
Williams, D. L.; Kim, Y.-M.; Watkins, S. C.; Keefer, L. K. J. Med. Chem.
1997, 40, 1947-1954. (d) Tang, X.; Xian, M.; Trikha, M.; Honn, K. V.;
Wang, P. G. Tetrahedron Lett. 2001, 42, 2625-2629.
(7) For reviews, see: (a) McCray, J. A.; Trentham, D. R. Annu. ReV. Biophys.
Chem. 1989, 18, 239-270. (b) Adams, S. R.; Tsien, R. Y. Annu. ReV.
Physiol. 1993, 55, 755-784. (c) Givens, R. S.; Kueper, L. W., III. Chem.
ReV. 1993, 93, 55-66. (d) Corrie, J. E. T.; Trentham, D. R. In Biological
Applications of Photochemical Switches; Morrison, H., Ed.; Wiley: New
York, 1994; pp 243-305. (e) Methods in Enzymology; Marriott, G., Ed.;
Academic Press: New York; Vol. 291, 1998. (f) Zacharias, D. A.; Baird,
G. S.; Tsien, R. Y. Curr. Opin. Neurobiol. 2000, 10, 416-421. (g)
Pelliccioli, A. P.; Wirz, J. Photochem. Photobiol. Sci. 2002, 1, 441-458.
(4) Diodati, J. G.; Quyyumi, A. A.; Keefer, L. K. J. CardioVasc. Pharmacol.
1993, 22, 287-292.
9
9806
J. AM. CHEM. SOC. 2002, 124, 9806-9811
10.1021/ja026900s CCC: $22.00 © 2002 American Chemical Society

NO Photochemical Release from Diazeniumdiolates
A R T I C L E S
Scheme 2
Scheme 3
compounds10) have been developed and used in a variety of
applications.11 Photochemical precursors to diazeniumdiolate
anions 1 have the added advantage of ultimately releasing NO
at controlled, well-defined rates that can be varied by the
substituents R, pH, or temperature.
Scheme 4
The commonly employed 2-nitrobenzyl photosensitive pro-
tecting group7 has been used to develop potential photochemical
precursors 2 to diazeniumdiolates.12 These derivatives have been
solvolysis of 3,5-dimethoxybenzyl acetate (4a, Scheme 3) in
50% aqueous dioxane.18 Benzyl acetates lacking π-electron-
donating groups in the 3- or 5-position, on the other hand,
undergo a photohomolysis reaction leading to radical-derived
products. Barltrop19 and Chamberlin20 later described the use
of the 3,5-dimethoxybenzyl group as an efficient phototrigger
for the release of amino acids in aqueous ethanolic or dioxane
solutions (4b, Scheme 3).
On the basis of this previous work, we have prepared a series
of O2-benzyl-substituted diazeniumdiolates, 2a,b, 3b, and 5-9,
and have examined the effect of electron-donating meta sub-
stituents on the observed photochemistry and on the efficiency
of NO release.
used to inhibit thrombin-stimulated platelet aggregation,12 to
examine the induction of long-term depression in the cerebel-
lum,13 and to study long-term potentiation in cultured hippoc-
ampal neurons.14
We have found, however, that the photochemistry of 2 is
severely complicated by an efficient photochemical reaction that
produces potentially carcinogenic nitrosamines and oxygen-
substituted nitrenes.15 Indeed, the photochemistry of 2 is
exactly analogous to that of simple O2-alkylated derivatives 3
(Scheme 2).16
We have begun an investigation of alternate photosensitive
protecting groups for diazeniumdiolates. This study examines
the use of the well-established meta effect of electron-donating
groups in benzylic systems.17 Such substitution has been shown
to favor the formation of ionic products in photochemical
reactions. Zimmerman first demonstrated the efficient photo-
Results and Discussion
(8) (a) Bettache, N.; Carter, T.; Corrie, J. E. T.; Ogden, D.; Trentham, D. R.
In Methods in Enzymology; Packer, L., Ed.; Academic Press: New York,
1996; Vol. 268, pp 266-281. (b) Works, C. F.; Ford, P. C. J. Am. Chem.
Soc. 2000, 122, 7592-7593 and references therein.
(9) (a) Sexton, D. J.; Muruganandam, A.; McKenney, D. J.; Mutus, B.
Photochem. Photobiol. 1994, 59, 463-467. (b) Etchenique, R.; Furman,
M.; Olabe, J. A. J. Am. Chem. Soc. 2000, 122, 3967-3968.
(10) (a) Namiki, S.; Arai, T.; Fujimori, K. J. Am. Chem. Soc. 1997, 119, 3840-
3841. (b) Namiki, S.; Kaneda, F.; Ikegami, M.; Arai, T.; Fujimori, K.;
Asada, S.; Hama, H.; Kasuya, Y.; Goto, K. Bioorg. Med. Chem. 1999, 7,
1695-1702. (c) Yoshida, M.; Ikegami, M.; Namiki, S.; Arai, T.; Fujimori,
K. Chem. Lett. 2000 730-731.
Products of Photolysis. Analysis of the organic products
observed following photolysis (Rayonet, 300 nm) of compounds
3b and 5-8 (1 mM) in both argon- and oxygen-saturated 90%
aqueous acetonitrile is summarized in Scheme 4 and Table 1.
(Results observed with compound 9 will be discussed in more
detail below.)
In our previous study of the photoreactivity of O2-substituted
diazeniumdiolates 3, we reported that in acetonitrile the forma-
tion of oxime was completely quenched by molecular oxygen
concomitant with increased yields of aldehyde and newly formed
nitrate (Scheme 2).15 In the present study, we find that although
the yield of oxime is suppressed by oxygen, it is not completely
quenched and attribute this difference to the lower solubility of
oxygen in water vs acetonitrile.21
(11) For reviews of NO-donor chemistry, see: (a) Gasco, A.; Fruttero, R.; Sorba,
G. Il Farmaco 1996, 51, 617-635. (b) Hou, T.-C.; Janszuk, A.; Wang, P.
G. Curr. Pharm. Des. 1999, 5, 417-441. (c) Feelisch, M. Arch. Pharmacol.
1998, 358, 113-122. (d) Yamamoto, T.; Bing, R. J. Proc. Soc. Exp. Biol.
Med. 2000, 225, 200-206. (e) Janero, D. R. Free Radical Biol. Med. 2000,
28, 1495-1506. (f) Lehmann, J. Expert Opin. Ther. Pat. 2000, 10, 559-
574. (g) Megson, I. L. Drugs Future 2000, 25, 701-715. (g) Wang, P. G.;
Xian, M.; Tang, X.; Wu, X.; Wen, Z.; Cai, T.; Janczuk, A. J. Chem. ReV.
2002, 102, 1091-1134.
(12) Makings, L. R.; Tsien, R. Y. J. Biol. Chem. 1994, 269, 6282-6285.
(13) Lev-Ram, V.; Makings, L. R.; Keitz, P. F.; Kao, J. P.; Tsien, R. Y. Neuron
1995, 15, 407-415.
Quantification of NO released upon photolysis was performed
electrochemically with an inNO Nitric Oxide Measuring System
(14) Arancio, O.; Kiebler, M.; Lee, C. J.; Lev-Ram, V.; Tsien, R. Y.; Kandel,
E. R.; Hawkins, R. D. Cell 1996, 87, 1025-1035.
(15) Srinivasan, A.; Kebede, N.; Saavedra, J. E.; Nikolaitchik, A. V.; Brady,
D. A.; Yourd, E.; Davies, K. M.; Keefer, L. K.; Toscano, J. P. J. Am.
Chem. Soc. 2001 123, 5465-5472.
(18) Zimmerman, H. E.; Sandel, V. R. J. Am. Chem. Soc. 1963, 85, 915-922.
(19) Barltrop, J. A.; Schofield, P. J. Chem. Soc. 1965, 4758-4765.
(20) Chamberlin, J. W. J. Org. Chem. 1966, 31, 1658-1659.
(21) The concentration of oxygen in saturated water solutions is 1.3 mM, while
that in saturated acetonitrile solutions is 9.1 mM (Murov, S. L.; Carmichael,
I.; Hug, G. L. Handbook of Photochemistry, 2nd ed.; Marcel Dekker: New
York, 1993; pp 289 and 293).
(16) A very minor reaction pathway involving the extrusion of nitrous oxide
(N2O) is also observed upon photolysis of 3.15
(17) (a) Zimmerman, H. E. J. Am. Chem. Soc. 1995, 117, 8988-8991. (b)
Pincock, J. A. Acc. Chem. Res. 1997, 30, 43-49. (c) Zimmerman, H. E. J.
Phys. Chem. A 1998, 102, 5616-5621.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9807

A R T I C L E S
Ruane et al.
Table 1. Yields of Products Detected by HPLC Analysis Following
Table 2. Quantum Yields for Photodecomposition and Yields of
Photolysis of O2-Substitued Diazeniumdiolatesa
Diazeniumdiolate 15 Following Photolysis of O2-Substitued
Diazeniumdiolates
nitrosamine
oxime
aldehyde
nitrate
alcohol
solvent
photolysis
λ (nm)
quantum
yielda
% diazeniumdiolate
reactant
purge
10
12
13
14
17
reactant
%H O/CH CN
15b
2
3
3b
3b
5
Ar
O2
Ar
O2
Ar
O2
Ar
O2
Ar
O2
90
85
95
95
74
90
67
71
53
53
78
53
b
7
21
b
b
5
14
13
12
30
16
0
27
0
0
0
2a
2b
3b
5
99/1
300
300
300
300
300
254
254
300
300
254
254
300
300
254
254
c
c
c
c
5
3
trace
6
10
c
c
35
c
c
c
6
99/1
99/1
99/1
99/1
99/1
0/100
99/1
0/100
99/1
0/100
99/1
0/100
99/1
5
b
b
6
6
37
31
44
37
35
16
0
9
6
b
13
40
36
27
27
6
c
7
0
0.17
0.35
0.07
0.27
0.66
0.81
0.08
0.28
0.51
0.74
7
b
8
0
7
8
8
b
a Based on percent reactant converted (approximately 10%). Estimated
error (2%. b Not determined.
47
c
c
0/100
c
a Of reactant photodecomposition, determined with the potassium fer-
rioxalate actinometer.24 b Based on percent reactant converted (determined
by HPLC analysis), the yield of NO measured, and the observation that
diazeniumdiolate 15 gives 1.5 equiv of NO.22,23 Estimated error (5%. c Not
determined.
Scheme 5
Figure 1. Quantification of NO yields observed after 300 nm photolysis
of 7 (0.1 mM in 99/1 H2O/CH3CN) from 0 to 40% conversion (determined
by HPLC analysis). See the text for details.
Mechanistic Analysis. We have interpreted our results
concerning the photochemistry of O2-substituted diazeniumdi-
olates 5-8 in terms of path A (undesired) and path B (desired)
shown in Scheme 4. Other minor reaction pathways may
contribute slightly to the overall photochemistry, so each
individual product yield of Table 1 is not a quantitative measure
of the relative contributions of paths A and B, but the results
taken as a whole, along with the diazeniumdiolate yields of
Table 2, give a very good indication of the partitioning between
these two major reaction pathways.
Path A involves the production of carcinogenic25 nitrosamine
10 and oxynitrene 11 and has been discussed in detail for 3
previously.15 The oxynitrene either rearranges to an intermediate
C-nitroso compound that forms oxime 12 or is trapped by
oxygen to give aldehyde 13 (and HONO) and nitrate 14. Path
B involves the formation of ionic products, diazeniumdiolate
15 and benzyl cation 16. In water, cation 16 is trapped to provide
benzyl alcohol 17; diazeniumdiolate 15 dissociates to diethy-
lamine and NO.
using an amiNO-2000 probe (Innovative Instruments Inc.). The
amiNO-2000 probe was calibrated using ascorbic acid/sodium
nitrite before and after NO measurements were conducted. To
differentiate NO production arising from diazeniumdiolate 15
from that of non-diazeniumdiolate-derived NO, we utilized the
pH dependence of diazeniumdiolate stability. At room temper-
ature diazeniumdiolate 15 has a lifetime of several hours at pH
11, but only several seconds at pH 3. Aqueous solutions at pH
11 of each reactant were irradiated and then analyzed for NO.
Any NO detected under these conditions must arise from a non-
diazeniumdiolate pathway. For all reactants examined, the
amount of NO formed from such a pathway was negligible. To
liberate the NO associated with photoreleased diazeniumdiolate
15, solutions were acidified with 1.0 M H2SO4 to pH 2-3 and
NO analysis was continued. Substantial amounts of NO were
then observed (e.g., Figure 1). (Control experiments indicate
that there is no electrochemical response associated with this
large pH change.) In each case, the yield of NO was determined
from a series of measurements following photolysis to a range
of percent conversions. NO yields were found to be linear for
up to 20-40% conversion (e.g., Figure 1, inset). The NO yields
determined from these experiments, along with the observation
that diazeniumdiolate 15 dissociates to give 1.5 equiv of NO,22,23
were used to derive yields of photoreleased diazeniumdiolate
from the systems studied. These yields along with quantum
yields of photodecomposition are reported in Table 2. (Again,
results observed with compound 9 will be discussed in more
detail below.)
In addition to production via path A, nitrosamine 10 could
potentially be formed by way of the reaction pathway shown
in Scheme 5. Here, photoinduced homolytic cleavage gives
radicals 18 and 19, which would be expected to provide
(22) Quantification of NO yields by chemiluminescence measurements2 or
ultrafiltration experiments23 has indicated that diazeniumdiolate 15 dis-
sociates to 1.5 equiv of NO, in contrast to most other derivatives, which
provide 2 equiv of NO. We have independently confirmed these observa-
tions by electrochemical detection of NO. (See the Supporting Information.)
(23) Ramamurthi, A.; Lewis, R. S. Chem. Res. Toxicol. 1997, 10, 408-413.
(24) Calvert, N.; Pitts, N. J. Photochemistry; Wiley: New York, 1966; p 783.
(25) Lijinsky, W. Chemistry and Biology of N-Nitroso Compound; Cambridge
University Press: Cambridge, U.K., 1992; p 399.
9
9808 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

NO Photochemical Release from Diazeniumdiolates
A R T I C L E S
nitrosamine 10, NO, and bibenzyl 20. We observe no evidence
for this reaction pathway. We have not detected bibenzyl 20 or
any other products that might be expected to arise from benzyl
radical 19 for any of the O2-substituted diazeniumdiolates
studied. In addition, NO is not detected at high pH where
diazeniumdiolate 15 is stable, but radical 18 would still be
expected to dissociate to nitrosamine 10 and NO. Indeed, as
discussed above, we have not observed any NO that is produced
via a pathway other than that shown in Scheme 4, path B.
Triplet sensitization experiments indicate that path B is a
singlet-state reaction, consistent with expectations based on
previous studies of analogous benzyl acetates.17 Triplet-
sensitized photolysis of 7 or 8 results in significantly diminished
yields of the products of path B. For example, using 4,4′-
dihydroxybenzophenone as the triplet sensitizer in argon-
saturated 90% aqueous acetonitrile, only trace amounts of NO
and the corresponding benzyl alcohol 17 are detected, while
the yield of nitrosamine 10 increases dramatically.
The data of Tables 1 and 2 demonstrate that the relative
contribution of paths A and B is strongly dependent on the
aromatic ring substitution pattern. The yield of nitrosamine 10
decreases and that of NO and alcohol 17 increases with stronger
π-donating meta substitutents. (We have found in the case of 8
that the corresponding benzyl alcohol 17 (X ) Y ) NMe2) is
photosensitive and under the conditions of our experiments
undergoes secondary photolysis to aldehyde 13 (X ) Y )
NMe2). In addition, and as observed previously,15 oxime 12 is
photosensitive and can also be converted to aldehyde 13 under
the conditions of our experiments.) We estimate that the
photodecomposition of O2-substituted diazeniumdiolates 5-8
proceeds approximately 5%, 10%, 35%, and 50% through path
B (Scheme 4), respectively.
Figure 2. Comparison of NO yields and relative contributions to the
observed photochemistry from path B (Scheme 4) following 300 nm
photolysis of 5-9 (0.1 mM in 99/1 H2O/CH3CN). The photolyses were
carried out at pH 11.2 to approximately 10% conversion. See the text for
details.
O2-3,5-Dihydroxybenzyl-Substituted Diazeniumdiolate 9.
Given our observation that the desired path B is favored with
strong π-donating meta substituents, we examined dihydroxy-
benzyl derivative 9. We hoped to take advantage of the enhanced
acidity of phenolic protons in the excited state.26 Photoinduced
deprotonation to yield an oxyanionic meta-substituted derivative
was expected to result in significant enhancement of the relative
contribution of path B to the observed photochemistry.
Consistent with the expected large effect of oxyanionic meta
substitution, when 9 is irradiated (Rayonet, 300 nm) in a pH
11.2 aqueous solution (containing 1% acetonitrile) to 5-35%
conversion, the amount of NO detected corresponds to a yield
of photoreleased diazeniumdiolate 15 of approximately 92%.
A comparison of results observed with O2-substituted diazeni-
umdiolate derivatives 5-9 is shown in Figure 2.
Figure 3. (a) 1H NMR analysis of the products observed following
photolysis (Rayonet, 300 nm) of a 7 mM solution of 9 in D2O/NaOD (pD
11) to approximately 20% conversion. R ) benzyl alcohol 17 (X ) Y )
OH), â ) nitrosamine 10, γ ) dihydroxybenzyl derivative 9, δ )
diazeniumdiolate 15, and ꢀ ) diethylamine. (b) 1H NMR spectrum observed
after spiking with an authentic sample of diazeniumdiolate 15.
conditions, diazeniumdiolate 15 undergoes photolysis to di-
ethylamine (and NO) and nitrosamine 10 (and NO-).27 A full
characterization of the photochemistry of 15 and related
diazeniumdiolates will soon be reported.28
When 9 is irradiated (Rayonet, 300 nm) in pH 8.4 or 7.4
solutions, however, the yield of diazeniumdiolate 15 is sub-
stantially reduced (to 18% and 16%, respectively), while that
of nitrosamine 10 is significantly enhanced. Excited-state depro-
tonation must not be rapid enough to compete with path A under
these conditions. The pH 11.2 results are easily rationalized by
simply considering that the pKa of the phenolic protons of 9 is
The photoproduction of 15 from dihydroxybenzyl derivative
1
9 was further verified by H NMR spectroscopy (Figure 3). A
7 mM solution of 9 in D2O/NaOD (pD 11) was photolyzed
(Rayonet, 300 nm) to approximately 20% conversion. New
signals for benzyl alcohol 17 (X ) Y ) OH), diazeniumdiolate
15, nitrosamine 10, and diethylamine were observed and
characterized by spiking with authentic samples of each. Since
dissociation of diazeniumdiolate 15 to NO and diethylamine is
not expected under such basic conditions, we examined the
photolysis of 15 itself. We have found that, under identical
(27) The UV-vis absorption spectrum of diazeniumdiolate 15 in aqueous
solution has λmax ) 248 nm and tails out to A ) 0 at approximately 320
nm. Thus, to avoid secondary photolysis of photoreleased diazeniumdiolates,
O2-protected precursors should ideally be photolyzed with light of
wavelength greater than 320 nm.
(28) D’Sa, R. A.; Ruane, P. H.; Kumar, N. A.; Relyea, H. A.; Toscano, J. P.
Manuscript in preparation.
(26) (a) Bartok, W.; Lucchesi, P. J.; Snider, N. S. J. Am. Chem. Soc. 1962, 84,
1842-1844. (b) Tolbert, L. M.; Solntsev, K. M. Acc. Chem. Res. 2002,
35, 19-27.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9809

A R T I C L E S
Ruane et al.
Figure 4. HPLC-derived yield of nitrosamine 10 observed after 300 nm
photolysis of 9 (0.1 mM in 99/1 H2O/CH3CN) as a function of solution
pH.
Figure 5. NO release rates observed following photolysis of pH 3.8 and
6.9 solutions of 7 (0.1 mM in 99/1 H2O/CH3CN) compared with the
corresponding rates observed for the thermal dissociation of the sodium
salt of diazeniumdiolate 15.
approximately 10. Thus, at pH 11.2 this compound is already
deprotonated when the excited state is formed and path B
becomes dominant. Consistent with this interpretation, the yield
of nitrosamine is dependent on solution pH and shows a marked
decrease at pH values above 10 (Figure 4).29
The quantum yield for photodecomposition of 9 in aqueous
solution is high. For 254 nm photolysis, the quantum yield
(determined with the 1,3-cycloheptadiene actinometer30) is 0.87
at both pH 7 and pH 12; for 300 nm photolysis, the quantum
yield (determined with the Reinecke’s salt actinometer31) is 0.14
at pH 7 and 0.36 at pH 12.
NO Release Rate. The rate of NO release from diazenium-
diolates 1 can be controlled by factors such as the substituents
R, pH, and temperature.2 Thus, an advantageous feature of
photoprecursors 5-9 is that the flux of NO can be controlled
(and varied) by these factors. Since 2-nitrobenzyl derivatives 2
have been reported to release NO much faster than expected
(within 5 ms, rather than over the course of several minutes)
upon photolysis,12 we wished to confirm that the NO release
rate observed with photoreleased diazeniumdiolate 15 is indeed
identical to that observed upon normal thermal decomposition
of 15. As demonstrated in Figure 5, the NO release rate observed
following photolysis of dimethylamino derivative 7 compares
very favorably to that found for thermal dissociation of the
sodium salt of 15 and, moreover, shows the same pH depen-
dence. These release rates are consistent with those that have
been reported previously.2,23,32
direct observation of diazeniumdiolate 15 with 1H NMR
spectroscopy and by NO quantification measurements conducted
in high- and low-pH solutions. In addition, the observed rates
of NO release are consistent with that expected for normal
thermal decomposition of diazeniumdiolate 15 in aqueous
solutions and also show the same pH dependence.
We are currently extending the systems studied here to related
derivatives that show efficient photochemical release of diaz-
eniumdiolates at neutral pH upon photolysis with longer (g350
nm) wavelength light.33 These results will soon be reported.
Experimental Section
General Methods. Unless otherwise noted, materials were obtained
from Aldrich Chemical Co. and used without further purification.
Acetonitrile and dichloromethane were distilled from calcium hydride,
and tetrahydrofuran was distilled from sodium/benzophenone before
use. N,N-Dimethylformamide (DMF) was dried by azeotropic distil-
lation with benzene and then further distilled over neutral alumina under
vacuum. Melting points were determined on a Melt Temp II apparatus
and are uncorrected. Infrared spectra were recorded on a Bruker IFS
1
55 Fourier transform IR spectrometer at 4 cm-1 resolution. H NMR
and 13C NMR spectra were recorded on a Bruker AMX 300 (300 MHz)
or a Varian Unity Plus 400 (400 MHz) Fourier transform NMR
spectrometer. Resonances are reported in δ units downfield from that
of tetramethylsilane. Mass spectra were collected with a VG 70-S mass
spectrometer in the fast atom bombardment mode with sample
introduction via direct probe. HPLC analysis was performed on a Waters
Delta 600 system equipped with a model 6000A pump, a model 2487
dual-wavelength UV detector, and a model U6K injector with a 20 µL
injector loop (Rheodyne). A Waters C-18 symmetry analytical column
(3.9 × 150 mm) was used. Absorption spectra were obtained using a
Hewlett-Packard 8453 diode array spectrophotometer.
Conclusions
The effect of meta electron-donating groups on the photo-
chemistry of O2-benzyl-substituted diazeniumdiolates has been
examined. Photolysis of parent benzyl derivative 3b results
almost exclusively in undesired photochemistrysthe formation
of nitrosamine 10 and oxynitrene 11 with very little, if any,
photorelease of diazeniumdiolate 15. We have been able to use
meta substitution to tune the photochemistry of these benzylic
systems. The desired photorelease of diazeniumdiolate 15 has
been shown to become more substantial with stronger π-donat-
ing meta substituents. This photorelease has been verified by
General Procedure for Coupling 1-(N,N-Diethylamino)diazen-
1-ium-1,2-diolate (15) with Benzyl Bromides. A slurry of 0.39 g (2.52
mmol) of diazeniumdiolate 15, prepared as described previously,2,34 in
10 mL of dry tetrahydrofuran was cooled to -78 °C. To this mixture
was added 1 equiv of the appropriate benzyl bromide (prepared by
literature methods35) in 2 mL of DMF via cannula. The reaction mixture
was stirred under an inert atmosphere and allowed to warm to room
(29) The products observed upon photolysis of derivatives 5-8 are not dependent
on solution pH.
(33) Bushan, K. M.; Xu, H.; Ruane, P. H.; D’Sa, R. A.; Pavlos, C. M.; Smith,
J. A.; Toscano, J. P. Manuscript in preparation.
(34) Drago, R. S.; Karsetter, B. R. J. Am. Chem. Soc. 1961, 83, 1819-1822.
(35) (a) 3-Methoxybenzyl bromide: Vejdelek, Z. J.; Zemecek, O.; Musil, V.;
Simek, A. Collect. Czech. Chem. Commun. 1964, 29, 776-794. (b) 3,5-
Dimethoxybenzyl bromide: Takatori, K.; Nishihara, M.; Nishiyama, Y.;
Kagiwara, M. Tetrahedron 1998, 54, 1585-1588.
(30) Numao, N.; Hamada, T.; Yonemitsu, O. Tetrahedron Lett. 1977, 19, 1661-
1664.
(31) Wegner, E. E.; Adamson, A. W. J. Am. Chem. Soc. 1966, 88, 394-404.
(32) Horstmann, A.; Menzel, L.; Gabler, R.; Jentsch, A.; Urban, W.; Lehmann,
J. Nitric Oxide: Biol. Chem. 2002, 6, 135-141.
9
9810 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

NO Photochemical Release from Diazeniumdiolates
A R T I C L E S
temperature over 24 h. Diethyl ether (20 mL) was then added; the
sodium bromide was removed by gravity filtration, diethyl ether and
tetrahydrofuran were removed by rotary evaporation, and DMF was
removed under vacuum. The resulting yellow oil was dissolved in
dichloromethane and washed with water. The organic layer was dried
over magnesium sulfate and concentrated by rotary evaporation to yield
the desired product, which was purified on a silica column using
dichloromethane as the eluent.
(300 MHz, CDCl3) δ 1.07 (t, 6H, J ) 8 Hz), 2.97 (s, 6H), 3.06 (q, 4H,
J ) 8 Hz), 5.24 (s, 2H), 6.67-6.75 (m, 3H), 7.18-7.26 (m, 1H); 13C
NMR (400 MHz, CDCl3) δ 11.6, 40.69, 48.86, 76.4, 112.46, 112.82,
116.73, 129.28, 136.61, 150.85; MS (FAB) m/z (rel intens) 289 (M +
Na, 100), 265 (M - H, 5); UV-vis (CH3CN) λmax 306 nm (ꢀ ) 3058
M-1 cm-1).
Data for O2-(3,5-bis(dimethylamino)benzyl)-1-(N,N-diethylami-
no)diazen-1-ium-1,2-diolate (8): 20% yield as a yellow oil; 1H NMR
(300 MHz, CDCl3) δ 1.05 (t, 6H, J ) 8 Hz), 2.96 (s, 12H), 3.07 (q,
4H, J ) 8 Hz), 5.20 (s, 2H), 6.02-6.03 (s, 1H), 6.19-6.20 (s, 2H);
13C NMR (400 MHz, CDCl3) δ 11.81, 41.03, 48.95, 77.235, 97.78,
102.53, 137.24, 152.01; MS (FAB) m/z (rel intens) 332 (M + Na, 82),
309 (M+, 20); UV-vis (CH3CN) λmax 317 nm (ꢀ ) 3704 M-1 cm-1).
O2-(3,5-Dihydroxybenzyl)-1-(N,N-diethylamino)diazen-1-ium-1,2-
diolate (9). The commercially available 3,5-dihydroxybenzoic acid was
esterified in quantitative yield to afford the corresponding methyl ester.39
The phenolic hydroxy groups were converted to methoxymethyloxy
functionalities.40 Lithium aluminum hydride reduction gave the alco-
hol,39 which was brominated under nonacidic conditions.34 The bromide
was coupled with 15 as described above in high yield, and the
methoxymethyloxy groups were removed using aqueous acid to afford
9 in quantitative yield: 1H NMR (300 MHz, CDCl3), δ 1.15 (s, 6H, J
) 7 Hz), 3.09 (q, 4H, J ) 7 Hz), 5.12 (s, 2H), 6.30 (t, 1H, J ) 2 Hz),
6.44 (d, 2H, J ) 2 Hz), 6.80 (s br, 2H).
Data for O2-(3-methoxybenzyl)-1-(N,N-diethylamino)diazen-1-
1
ium-1,2-diolate (5): 80% yield as a yellow oil; H NMR (300 MHz,
CDCl3) δ 1.05 (t, 6H, J ) 8 Hz), 3.07 (q, 4H, J ) 8 Hz), 3.82 (s, 3H),
5.25 (s, 2H), 6.86-6.88 (d, 1H, J ) 8.5 Hz), 6.93-6.99 (m, 2H), 7.23-
7.28 (m, 1H); 13C NMR (400 MHz, CDCl3) δ 11.64, 48.89, 55.39,
75.71, 113.62, 114.56, 128.81, 129.71, 137.38, 159.86; MS (FAB) m/z
(rel intens) 276 (M + Na, 100), 254 (M + H, 3); UV-vis (CH3CN)
λmax 273 nm (ꢀ ) 3555 M-1 cm-1).
Data for O2-(3,5-dimethoxybenzyl)-1-(N,N-diethylamino)diazen-
1-ium-1,2-diolate (6): 90% yield as a yellow oil; 1H NMR (300 MHz,
CDCl3) δ 1.05 (t, 6H, J ) 7 Hz), 3.05 (q, 4H, J ) 7 Hz), 3.80 (s, 6H),
5.27 (s, 2H), 6.41 (t, 1H), 6.53 (d, 2H).
Preparation of O2-(3-(Dimethylamino)benzyl)- and O2-(3,5-Bis-
(dimethylamino)benzyl)-1-(N,N-diethylamino)diazen-1-ium-1,2-di-
olate (7, 8). The methyl esters of the commercially available 3-amino-
and 3,5-diaminobenzoic acids were prepared according to literature
procedures.36 Each amino group was then methylated to the corre-
sponding trimethylammonium salt using a 5-fold excess of methyl
iodide.37 Reduction of the ammonium salts with a large excess of lithium
aluminum hydride38 gave the desired 3-(dimethylamino)- and 3,5-bis-
(dimethylamino)benzyl alcohols, respectively, in 80% yield. The
bromide precursors were prepared by reacting the corresponding benzyl
alcohols with bromide ion.34 These bromides were too reactive to be
isolated and were coupled immediately with diazeniumdiolate 15 to
afford 7 and 8 in yields of ca. 20%.
Preparation of Authentic Samples for Product Analysis. The
required benzaldehyde41 and oxime42 derivatives were synthesized (from
the corresponding benzyl alcohols) according to literature procedures.
Acknowledgment. J.P.T. gratefully acknowledges the Na-
tional Institutes of Health (Grant R01 GM58109) for generous
support of this research. A Camille Dreyfus Teacher-Scholar
Award and an Alfred P. Sloan Research Fellowship also support
this research.
Data for O2-(3-(dimethylamino)benzyl)-1-(N,N-diethylamino)-
Supporting Information Available: Quantification of the
amount of NO formed following thermal dissociation of varying
concentrations of diazeniumdiolate 15 (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
1
diazen-1-ium-1,2-diolate (7): 30% yield as a yellow oil; H NMR
(36) (a) Methyl 3-aminobenzoate: Elhadi, F. E., Ollis, W. D., Stoddart, J. F. J.
Chem. Soc., Perkin Trans. 1 1982, 8, 1727-1730. (b) Methyl 3,5-
diaminobenzoate: Kraft, A.; Osterod, F. J. Chem. Soc., Perkin Trans. 1
1998, 6, 1019-1025.
(37) Drewes, S. E.; Magojo, H. E. M.; Sutton, D. A. J. Chem. Soc., Perkin
Trans. 1 1975, 13, 1283-1284.
(38) (a) 3-(Dimethylamino)benzyl alcohol: Sakurai, H.; Izuoka, A.; Sugawara,
T. J. Am. Chem. Soc. 2000, 122, 9723-9734. (b) 3,5-Bis(dimethylamino)-
benzyl alcohol: Connolly, C. J. C.; Hamby, J. M.; Schroeder, M. C.;
Barvian, M.; Lu G. H.; Paneck, R. L.; Amar, A.; Shen, C.; Kraker, A. J.;
Fry, D. W.; Klohs, W. D.; Doherty, A. M. Biol. Org. Med. Chem. Lett.
1997, 7, 2415-2420.
JA026900S
(39) Stewart, G. M.; Fox, M. A. J. Am. Chem. Soc. 1996, 118, 4354-4360.
(40) Duffley, R. P.; Handrick, G. R.; Uliss, D. B.; Lambert, G.; Dalzell, H. C.;
Razdan, R. K. Synthesis 1980, 9, 733-736.
(41) Sakurai, H.; Izuoka, A.; Sugawara, T. J. Am. Chem. Soc. 2000, 122, 9723-
9734.
(42) Gale, D. J.; Wilshire, J. F. K. Aust. J. Chem. 1975, 28, 2447-2458.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9811