Studies of structure-activity relationship of nitroxide free radicals and their precursors as modifiers against oxidative damage

Article abstract of DOI:10.1021/jm9802160

The protective effects of stable nitroxides, as well as their hydroxylamine and amine precursors, have been tested in Chinese hamster V79 cells subjected to H₂O₂ exposure at fixed concentration or exposure to ionizing radiation. Cytotoxicity was evaluated by monitoring the viability of the cells assessed by the clonogenic assay. The compounds tested at fixed concentration varied in terms of ring size, oxidation state, and ring substituents. Electrochemical studies were carried out to measure the redox midpoint potentials. The studies show-that in the case of protection against H₂O₂ exposure, the protection was determined-by the ring size, oxidation state, and redox midpoint potentials. In general the protection factors followed the order nitroxides > hydroxylamines > amines. Both the six- membered ring nitroxides and substituted five-membered ring nitroxides were efficient protectors. For six-membered ring nitroxides, the compounds exhibiting the lowest midpoint potentials exhibited maximal protection. In the case of X-radiation, nitroxides were the most protective though some hydroxylamines were also efficient. The amines were in some cases found to sensitize the toxicity of aerobic radiation exposure. The protection observed by the nitroxides was not dependent on the ring size. However, the ring substituents had significant influence on the protection. Compounds containing a basic side chain were found to provide enhanced protection. The results in this study suggest that these compounds are novel antioxidants which can provide cytoprotection in mammalian cells against diverse types of oxidative insult and identify structural determinants optimal for protection against individual types of damage.

Full text of DOI:10.1021/jm9802160

J . Med. Chem. 1998, 41, 3477-3492
3477
Stu d ies of Str u ctu r e-Activity Rela tion sh ip of Nitr oxid e F r ee Ra d ica ls a n d
Th eir P r ecu r sor s a s Mod ifier s Aga in st Oxid a tive Da m a ge
Murali C. Krishna,† William DeGraff,† Olga H. Hankovszky,‡ Cec´ılia P. Sa´r,‡ Tama´s Ka´lai,‡ J o´zsef J ek_o,§
Angelo Russo,† J ames B. Mitchell,† and Ka´lma´n Hideg*,‡
Radiation Biology Branch, National Cancer Institute, Bethesda, Maryland 20892, Institute of Organic and Medicinal
Chemistry, University of Pe´cs, H-7643 Pe´cs, P.O. Box 99, Hungary, and ICN Alkaloida Company Ltd., H-4440 Tiszavasva´ri,
P.O. Box 1, Hungary
Received April 8, 1998
The protective effects of stable nitroxides, as well as their hydroxylamine and amine precursors,
have been tested in Chinese hamster V79 cells subjected to H2O2 exposure at fixed concentration
or exposure to ionizing radiation. Cytotoxicity was evaluated by monitoring the viability of
the cells assessed by the clonogenic assay. The compounds tested at fixed concentration varied
in terms of ring size, oxidation state, and ring substituents. Electrochemical studies were
carried out to measure the redox midpoint potentials. The studies show that in the case of
protection against H2O2 exposure, the protection was determined by the ring size, oxidation
state, and redox midpoint potentials. In general the protection factors followed the order
nitroxides > hydroxylamines > amines. Both the six-membered ring nitroxides and substituted
five-membered ring nitroxides were efficient protectors. For six-membered ring nitroxides,
the compounds exhibiting the lowest midpoint potentials exhibited maximal protection. In
the case of X-radiation, nitroxides were the most protective though some hydroxylamines were
also efficient. The amines were in some cases found to sensitize the toxicity of aerobic radiation
exposure. The protection observed by the nitroxides was not dependent on the ring size.
However, the ring substituents had significant influence on the protection. Compounds
containing a basic side chain were found to provide enhanced protection. The results in this
study suggest that these compounds are novel antioxidants which can provide cytoprotection
in mammalian cells against diverse types of oxidative insult and identify structural determi-
nants optimal for protection against individual types of damage.
In tr od u ction
stress”. In most cases the defense provided by the
enzymatic and nonenzymatic antioxidants is adequate.
However, in acute situations, such as radiation-induced
normal tissue toxicity, anthracycline-induced cardiomy-
opathy, and iron overload, supplementation with exog-
enous antioxidants may be necessary to minimize tissue
damage. However, the targets and sites of free radical
damage vary depending on the kind of stress. For
example, organic peroxides exert cytotoxicity by damag-
ing cell membranes, whereas ionizing radiation induces
cytotoxicity by causing DNA double-strand breaks.
Therefore, the sites of damage need to be identified, and
appropriate antioxidant strategies need to be devised
to effectively limit tissue injury depending on the type
of insult. This has prompted a search for efficient
antioxidants which can augment the normal antioxidant
defense and provide defense against a stress imposed
by a wide range of modalities in chronic as well as acute
pathological conditions. Desired features in an effective
antioxidant would include (a) ability to localize in
subcellular compartments, (b) ability to react and
scavenge with a wide range of reactive species, and (c)
ability to be regenerated to the active form efficiently.
Reactive oxygen species (ROS) are byproducts of
normal metabolic processes in aerobic environments.1
These species can exert cytotoxicity by damaging critical
cellular components necessary for viability.2 Examples
of the ROS are superoxide, hydrogen peroxide, and
hydroxyl radicals produced by sources such as biore-
ductively activated xenobiotics and ultraviolet and
ionizing radiation. Low-molecular-weight complexes of
transition metals such as iron and copper can also
catalyze the formation of some of these oxidizing species.
In addition to exerting toxicity by oxidation, ROS have
been also implicated in the activation of factors such as
NF-κB via mechanisms involving redox reactions. Detox-
ification of these oxidants is provided by enzymatic and
nonenzymatic mechanisms. Enzymatic antioxidants
include superoxide dismutase (SOD), catalase (Cat), and
glutathione peroxidase (GPX) which catalytically detox-
ify oxidants, while nonenzymatic antioxidants are pri-
marily reducing agents such as vitamin C and vitamin
E which can scavenge oxidants by H atom donation in
a stoichiometric manner.3 Imbalances in the detoxifi-
cation of ROS relative to their production lead to what
is a widely accepted phenomenon called “oxidative
Stable nitroxide free radicals, which have been used
as biophysical probes, have been identified as novel
antioxidants. By undergoing one-electron-transfer reac-
tions, nitroxides are readily reduced to the hydroxy-
lamines4,5 or oxidized to the oxoammonium cation.6,7
Consequently, all three forms of the oxidation states
* Address correspondence to Ka´lma´n Hideg at University of Pe´cs.
Phone/Fax: (36) (72) 325-731. E-mail: KHIDEG@main.pote.hu.
† National Cancer Institute.
‡ University of Pe´cs.
§ ICN Alkaloida Co. Ltd.
S0022-2623(98)00216-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 08/12/1998

3478 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Krishna et al.
might exist in a biological system. The nitroxide/
oxoammonium pair behaves as an efficient and electro-
chemically reversible redox pair and is involved in the
catalytic decomposition of superoxide6,8 as well as
inducing catalase-like activity in heme proteins.9 Though
not directly oxidizing, piperidin-1-oxyl nitroxides are
often used as cooxidants in organic chemistry.10 The
hydroxylamine, on the other hand, can function as a
typical reducing agent like vitamins C and E to scavenge
oxidants,11 while the nitroxide participates in radical-
radical recombination reactions.12 The metabolism of
nitroxides in biological systems has been investigated
in detail and found to be reduced to the corresponding
hydroxylamines through enzyme-related mechanisms.4
The oxidation of the hydroxylamines can also occur
efficiently. Consequently, a distribution of nitroxide/
hydroxylamine is achieved regardless of which form is
administered in vivo. Sterically hindered five- and six-
membered cyclic amines without H atoms at R-positions
have the ability of undergoing oxidation to paramagnetic
N-oxyl compounds.13 The influence of the ring structure
on the nitroxide/hydroxylamine distribution in cells and
tissues has been studied. Nitroxides belonging to the
six-membered piperidine ring structure have been shown
to be rapidly converted to the hydroxylamine, unlike the
five-membered pyrrolidine or pyrroline nitroxides.4
scavenging of short-lived radicals generated in close
proximity to DNA.
To study the structural requirements of nitroxide for
general antioxidant effects as well as radioprotection,
a systematic screening of these agents was performed
using an in vitro assay. The effect of ring size, substit-
uents, and oxidation state of the test compounds at a
fixed concentration, in providing protection to mam-
malian cells exposed to oxidative damage induced by
exposure to H2O2 or ionizing radiation, was evaluated
using the clonogenic viability of Chinese hamster lung
fibroblasts. The results indicate that the requirements
of an agent to be effective in providing protection against
H2O2-induced toxicity are predominantly influenced by
the ring size, redox potential, and oxidation state. In
the case of protection against ionizing radiation, the ring
substituents and the oxidation state were the main
determinants of the radiation modification.
Ch em istr y
Generally speaking the structure of nitroxides can be
modified by three methods: first, five- or six-membered
ring can be constructed as it is required; second, the
nitroxide function can be reduced partially, totally
removed, or restored; third, the functional group on the
ring can be altered as well. Nitroxides (14c, 16c, 22c,
23c, 25c, 36c, 37c, 40c, 55c) were reduced to secondary
amines with Fe/AcOH24 without alteration of other
functional groups attached to the nitroxide ring. In a
reverse process sterically hindered amine 55a could be
oxidized to nitroxide with H2O2 in the presence of Na2-
WO4.13 The N-hydroxylamines (1b, 2b, 5b, 6b, 9b,
11b-17b, 19b, 22b, 23b, 25b-27b, 29b, 36b-38b,
40b, 42b, 48b, 52b, 53b, 55b) could be obtained by
refluxing nitroxides with EtOH saturated with HCl
gas.13 The N-hydroxylamines in basic or buffered media
were oxidized to nitroxides with PbO2 or MnO2 ac-
companied by bubbling oxygen through the solution.
The synthesis of all pyrroline or piperidine nitroxides
utilized 4-oxo-2,2,6,6-tetramethylpiperidine13 as starting
material. Favorskii rearrangement of 3,5-dibromo-4-
oxo-2,2,6,6-tetramethylpiperidine (59) with sodium meth-
oxide afforded the pyrroline methyl ester (42a ), which
could be reduced to allylic alcohol (14a ) by SMEAH. This
compound was readily oxidized to R,â-unsaturated al-
dehyde26 36a which was methylated at its amino
function to tertiary amine 36d (Scheme 1).
Secondary amines (18c-20c, 25c, 28c) were obtained
by sodium borohydride reduction of nonisolated Schiff’s
bases synthesized from aldehydes26,27 (36c, 37c). The
secondary amine could be alkylated with methyl iodide
in the presence of NaH to yield tertiary amines (26c,
29c). Tertiary amine 10c in which the nitrogen atom
is attached to a secondary carbon atom could be
achieved by reductive alkylation with sodium cyanoboro-
hydride28,29 starting from TEMPON (39c) and dibutyl-
amine. Assessing the fact that paramagnetic ethers
exhibited excellent NMR relaxitivity,30 several of them,
such as 15c, were synthesized by alkylating the corre-
sponding alcohol 14c (Scheme 2).
It was found that compared to nitrones, which trap
short-lived radicals to form stable nitroxide adducts,
several nitroxide free radicals possess better antioxidant
properties and decrease the oxidative damage caused
•-
by reactive oxygen-centered free radicals (HO•, O2
,
HO2 ).8 Studies conducted in this laboratory have
identified protective effects of nitroxides in several
cellular and in vivo studies.14 For example, nitroxides
and hydroxylamines have been shown to protect mam-
malian cells exposed to superoxide, hydrogen peroxide
(H2O2), organic hydroperoxides, and redox cycling an-
ticancer agents.15-18 A variety of sterically hindered
pyrrolinecarboxamides were prepared and tested for
antiarrhythmic effects.19 It was recently demonstrated
that one of these compounds, N-[(2,2,5,5-tetramethyl-
3-pyrroline-3-carboxamido)propyl]phthalimide which is
currently under clinical trial, was oxidized in vivo to
stable nontoxic nitroxide free radical as detected by in
vivo EPR spectroscopy in living animals.20 Mechanisms
explaining the protective effects include (a) SOD-mimic
activity, (b) oxidizing redox active transition metals, (c)
reducing hypervalent heme states, (d) radical-radical
recombination, and (e) chain-breaking antioxidants.
However, in the case of exposure to ionizing radiation,
nitroxides were effective in inhibiting the aerobic cyto-
toxicity, whereas the hydroxylamines, which are reduc-
ing agents, were ineffective similar to other reducing
agents such as vitamin C and vitamin E.21 The thiol-
based radioprotectors, where the sulfhydryl agent pro-
vides radioprotection either by radical scavenging or by
chemical repair, inhibit radiation-induced DNA le-
sions.22 The most effective thiol-based radioprotectors
require the presence of basic functions (amino, amidino,
or guanidino).23,24 The enhanced protection afforded by
thiols with basic functional groups has been attributed
to the better affinity of the positively charged substit-
uents on the thiol to the DNA thus allowing the
•
The most convenient method for obtaining paramag-
netic tertiary amines (17c, 21c, 22c) was the alkylating
of secondary amines with allylic bromide 60.31 This
paramagnetic alkylating agent was used to get am-

SAR of Nitroxide Free Radicals and Precursors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18 3479
Sch em e 1a
Sch em e 3a
a
Reagents: (A) Fe powder, AcOH, 3 h, 40-50 °C; (B) 30% H2O2,
MeOH, 3 days; (C) HCl/EtOH, 30 min; (D) K2CO3, CHCl3, then
MnO2, 30 min; (E) MeONa, MeOH, 1 h; (F) SMEAH, toluene, 40
°C, 1 h; (G) MnO2, CHCl3, 30 min; (H) MeI, acetone, reflux, 4 h.
a
Reagents: (M) pyrrolidine, CHCl3, reflux, 3 h; (N) n-tributyl-
amine, acetone, reflux, 6 days; (O) triphenylphosphine, acetone,
reflux, 3 h; (P) o-dibromoxilol, K2CO3, CHCl3, reflux, 3 h; (Q)
S-methylisothiouronium sulfate, EtOH, reflux, 3 h; (R) Et3N,
MeSO2Cl, CH2Cl2, rt, 3 h; (S) NH3, cyclohexane.
Sch em e 2a
and ammonia. The investigation of these compounds
is well-supported so that parent compounds, i.e., guani-
dine and urea, often appear in bioactive molecules and
drugs. Besides the amines which proved the most
efficient agents, we synthesized their sulfonylated 13c
derivative. Compound 13c was obtained with standard
procedure, e.g., reacting amine 9c with methanesulfonyl
chloride in the presence of triethylamine (Scheme 3).
Paramagnetic amides (45c, 46c, 48c, 49c) were
achieved by acylating the corresponding amines with
paramagnetic mixed anhydride ester 6334 or 1-acylimi-
dazole 64. To avoid catalytic hydrogenation of double
bond in pyrrolines, pyrrolidine derivative 66 was ob-
tained by reaction of 1-oxyl-3-bromo-4-oxo-2,2,6,6-tet-
ramethylpiperidine35 (65) with 1,3-diaminopropane.
Fusing compound 66 with phthalimide without solvent
at 120 °C yielded compound 52c while ammonia gas was
released.36 This method had to be used in synthesis of
55c from 9c, since in this case the alkylation did not
give the desired product, while alkylation of phthalimide
with allylic bromide 60 furnished 53c in good yield
(Scheme 4).
a
Reagents: (I) (i) 1-aminoadamantane, cat. p-toluenesulfonic
acid, toluene, reflux, 12 h, (ii) NaBH4, EtOH, 0 °C then reflux for
h; (J ) NaH, THF, reflux, 1 h then MeI, reflux 6 h; (K)
dibutylamine hydrochloride, NaCNBH3, MeOH, reflux, 48 h; (L)
NaH, THF, rt, 1 h then EtBr, DMF, reflux, 2 h.
1
monium (30d , 31d ) and phosphonium (32e) salts from
tertiary amines and triphenylphosphines.32 The isoin-
dolyl derivative 24c was obtained most conveniently by
alkylating allylic amine 6133 with R,R′-dibromo-o-xylene.
Attractive challenging target was to achieve spin-
labeled guanidine 33c and urea 43c from amine 61 and
S-methylisothiouronium salt and from isocyanate 6233
Resu lts a n d Discu ssion
The compounds evaluated in this study as potential
modifiers of cytotoxicity of both ionizing radiation and
H2O2 are listed in Tables 1-10. Figure 1 shows the
general structure of the compounds tested. For each
ring type, three possible oxidation states may exist:

3480 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Krishna et al.
Sch em e 4a
tion of 100 µM, present during the 1-h H2O2 exposure.
This concentration was selected for several reasons.
First, based on our experience with Tempol (5c), a final
concentration of 100 µM provides partial protection
against H2O2 cytotoxicity, whereas higher concentra-
tions (1000 µM) completely protect.15 A final concentra-
tion of 100 µM of each compound would, therefore, afford
comparison to our lead compound and allow identifica-
tion of agents with greater protective properties than
Tempol (5c). Second, because numerous compounds
were evaluated, only small quantities were synthesized
making the use of higher concentrations prohibitive. The
protection afforded by the compounds against H2O2-
induced cytotoxicity is shown in Tables 1-10 and Figure
2 (Supporting Information). The results shown in
Figure 2 indicate that the test compounds exhibited a
broad range in modifying the cytotoxic effects of H2O2.
Most agents tested were effective in providing protective
effects against exposure. In comparison to Tempol (5c)
which gave a PF of 2.5, 12 compounds yielded greater
PFs. The most efficient protector identified by the
screen was Tempo (2c) which provided a PF of 3.5. With
respect to structure and the protective effects against
H2O2 exposure, the following observations can be made.
The efficacy of stable nitroxides as well as the
corresponding hydroxylamines in reversing the free
radical-mediated damage induced by oxidative stress
prompted this study to identify structural requirements
of this class of compounds for optimal antioxidant
effects. The model chosen to screen for the efficacy of
the antioxidant effects of the test compounds was by
exposing Chinese hamster V79 cells to hydrogen per-
oxide for a fixed duration of time and assessing the
clonogenic viability in the absence and presence of a
fixed concentration of the compound to be tested.
Hydrogen peroxide is produced cellularly as a result of
incomplete reduction of oxygen to water or also produced
by phagocytic cells as well as by redox cycling xenobiotic
drugs. The damage associated with exposure to hydro-
gen peroxide is understood to be mediated by free
radical generation presumably by the generation of •OH
radical or an equally strong oxidant catalyzed by redox
active transition-metal complexes (eq 1).
a
Reagents: (T) 3-aminopropanol, CHCl3, reflux, 30 min; or 1,3-
diaminopropane, THF, 0 °C f rt, 1 h; (U) 1,3-diaminopropane,
H2O/THF, rt, 2 h; (V) phthalimide, 120 °C, 1 h; (W) phthalimide
K2CO3, KOH, 18-K-6, dioxane, reflux, 4 h; (Z) phthalimide, 120
°C, 50 min.
H2O2 + Fe2+-chelate f
•OH + -OH + Fe3+-chelate (1)
•OH + LH f L• + H2O (initiation)
L• + O2 f LOO• (propagation)
(2)
(3)
LOO• + LH f LOOH + L• (propagation) (4)
F igu r e 1. General chemical structures of the nitroxides
studied.
LOOH f LO• + •OH (propagation)
(5)
(6)
namely, the fully reduced secondary amine (type a ), the
hydroxylamine (type b), and the nitroxide (type c). In
some cases the redox midpoint potential of the nitroxide/
oxoammonium cation redox pair was estimated.
LOO• + LOO• f products (termination)
Earlier studies have shown that both free radical
scavengers as well as metal chelators such as desferal
can significantly inhibit both the cytotoxicity as well as
the oxidation of cellular components. Thus cellular
incubation of H2O2 represents a valid modality of
oxidant stress and has been widely used to screen for
agents to reverse this damage. The damage is associ-
Hyd r ogen P er oxid e Tr ea tm en t. The mean sur-
viving fraction of V79 cells following a 1-h exposure to
H2O2 alone (500 µM) was 0.14 ( 0.016. Possible
modulation of H2O2-mediated cytotoxicity by the test
compounds was evaluated using a fixed final concentra-

SAR of Nitroxide Free Radicals and Precursors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18 3481
Ta ble 1. Alcohols, Amines, and Amides
yield,
%
H2O2
prot
radiat
prot
compd
n
R
method
C
mp, °C
170
formulaa
mol wt
179.69
ref
43
1b
0
H
78
C8H17NO‚HCl
2.3
(0.52)
0.9
0.70
(0.10)
4.60
1c
2b
2c
3c
4c
5a
5b
5c
6b
6c
7c
8c
9b
9c
10c
11b
0
1
1
0
0
1
1
1
1
1
0
0
1
1
1
1
H
C8H16NO
142.22
193.72
156.25
158.22
172.25
157.26
209.72
172.25
237.77
200.30
157.24
171.26
245.19
171.26
283.48
436.89
50
43
44
13
26
44
43
13
43
30
44
44
43
44, 13
43
43
(0.06)
1.5
(0.20)
3.5
(0.28)
1.4
(0.14)
1.1
(0.40)
0.68
(0.10)
4.20
(0.00)
2.90
(1.40)
4.7
(0.20)
0.90
(0.08)
1.05
(0.04)
5.10
(0.08)
NT
H
C
67
180
dec
C9H19NO‚HCl
C9H18NO
H
OH
C8H16NO2
CH2OH
OH
C9H18NO2
(0.00)
0.4
C9H19NO
(0.00)
OH
C
C
65
63
>240
C9H19NO2‚HCl
C9H18NO2
NTb
OH
2.5
(0.23)
1.6
(0.30)
1.9
(0.50)
1.1
(0.00)
0.7
(0.00)
1.0
(0.00)
2.8
(0.14)
2.9
OCH2CH3
OCH2CH3
NH2
120
C11H23NO2‚HCl
C11H23NO2
C8H17N2O
NT
12.9
(0.02)
13.7
(0.00)
1.10
(0.04)
18.5
CH2NH2
NH2
C9H19N2O
C
74
>230
C9H20N2O‚2HCl
C9H19N2O
NH2
(0.31)
NT
N(n-Bu)2
K
C
35
61
oil
C17H35N2O
C18H37N3O2.3HCl
(0.64)
1.9
>230
1.40
(0.30)
1.8
(0.00)
NT
(0.13)
0.85
(0.20)
0.77
12b
13b
1
1
NHCOCH3
NHSO2CH3
C
C
73
78
128-130
182-184
C11H22N2O2‚HCl
C10H22N2O3S‚HCl
250.77
286.82
43
43
(0.05)
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
RRNO• + H+ + HO2• f RRNO+ + H2O2
RRNO+ + O2-• f RRNO• + O2
(7)
(8)
ated with DNA single- and double-strand breaks (in the
presence of specific chelators) as well as damage to other
cellular organelles.
In this study nitroxides of four different ring classes
were evaluated such as the six-membered piperidine
class (I), the six-membered 1,2,5,6-tetrahydropyridine
class (II), the saturated five-membered pyrrolidine class
(III), the unsaturated five-membered pyrroline class
(IV), and nitrones (V) (Figure 1). Electrochemical
studies on these agents show that while the hydroxy-
lamines undergo an irreversible oxidation around 200
mV vs NHE, at higher potentials, a reversible redox
couple consistent with the nitroxide and the correspond-
ing one-electron oxidation product, the oxoammonium
cation, exists around in the range 700-1100 mV for the
compounds tested. The nitroxide/oxoammonium redox
couple has been shown in earlier studies to mediate
enzyme mimetic reactions catalytically, such as SOD-
mimicking activity.6
The catalytic efficiencies of superoxide dismutation
were found to depend on the ease of oxidation of the
nitroxide to the corresponding oxoammonium cation,
which represents the rate-limiting step. The catalytic
rate constants were found to be inversely correlated to
the redox midpoint potential of the nitroxide/oxoammo-
nium couple.8 The nitroxide/oxoammonium couple has
also been postulated to induce catalase-like activity in
heme proteins.9 Either of these roles can confer anti-
oxidant activity in the nitroxides. In addition, nitrox-
ides have been shown to prevent the generation of
oxidants via Fenton reactions (eq 1) by maintaining the
metals in the oxidized state.15

3482 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Ta ble 2. Alcohols and Amines
Krishna et al.

SAR of Nitroxide Free Radicals and Precursors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18 3483
Ta ble 2 (Continued)

3484 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Ta ble 2 (Continued)
Krishna et al.
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
RRNO• + H+ + Fe2+-chelate f
RRNOH + Fe3+-chelate (9)
bered nitroxides. This is a remarkable fact because of
the lower toxicity of five-membered ring maintaining
the similar H2O2 protecting activity.38,39 Saturated five-
membered alcohol 4c exhibited a smaller protecting
activity than its unsaturated derivative 14c. Among
saturated (50c, 51c) and unsaturated (41c, 44c) acid
derivatives, there is no significant difference in H2O2
protecting activity. In the range of well-available
unsaturated acid derivatives, we observed that ester 42c
has better protecting activity than the corresponding
acid 41c. Among all the compounds which exhibit the
best protecting activity are amines 9c, 10c, 21c, 25c,
and 26c; however, quaternary salts 30d , 31d , 33c, 34c,
and 35c show small protecting activity. The amine-
substituted five-membered and six-membered nitroxides
(9c, 10c, 25c 26c) show good protecting activity. The
order of amines was not found to have significant
influence on protecting activity as was observed in the
case of 25c and 26c.
The hydroxylamines, on the other hand, were found
to behave like classic reducing agents such as ascorbate
and tocopherol by being oxidized to the corresponding
nitroxide by the oxidant in the detoxifying reaction.
RRNOH + •OH f RRNO• + H2O
(10)
However, unlike ascorbate, tocopherol, and other phe-
nolic antioxidants, the product (nitroxide) is also capable
of further detoxification modalities such radical-radical
recombination reactions and interrupting chain propa-
gation of lipid peroxidation reactions.37
RRNO• + L• f RRNOL
(11)
For a given ring size, the effect of the midpoint redox
potential of the nitroxide/oxoammonium catalytic couple
has also been examined for its influence on the protec-
tive effects. On the basis of the results obtained in this
study, some observations can be made.
Among the six-membered piperidines, where the
nitroxide and the hydroxylamines were compared at
similar concentrations, nitroxides were found to be
better protectors. However, the protection was inversely
proportional to the midpoint potentials of the nitroxide/
oxoammonium couple for the piperidines (Figure 3A).
The nitroxides which were easily oxidizable provided
better protection. However, in the case of five-mem-
bered nitroxides such as the pyrrolidines and the
pyrrolines such a correlation was absent (Figure 3B).
However, in this case, several hydroxylamines exhibited
better protection than the corresponding nitroxides
suggesting that a H atom donation might be the
predominant operating factor in these cases, similar to
antioxidants such as ascorbate, thiols, etc.
St r u ct u r e-Act ivit y R ela t ion sh ip s. The com-
pounds of type a (amines) were in general ineffective
in providing protection or in some cases actually en-
hanced the cytotoxicity of H2O2 exposure. The nitrox-
ides and hydroxylamines were in general protective.
The well-known six-membered nitroxides (2c, 5c, 9c,
39c) were investigated earlier and showed better pro-
tecting activity then the corresponding five-membered
nitroxides (1c, 3c, 7c, 38c). Redox behavior alone might
not be able to account for this behavior since no
significant differences in their midpoint potentials were
found. However, on the basis of several earlier reports
which found six-membered nitroxides more reactive
with radicals and are more likely to be converted to the
corresponding hydroxylamine than the corresponding
five-membered counterparts,4 the enhanced ability of
the pideridine class of nitroxides can be attributed to
their reactivities. This has been explained in terms of
the access to the nitroxide moiety in the ring to
reactants. In the case of six-membered ring nitroxides,
the “boat” or “chair” conformations of the piperidine ring
provide easy access to the nitroxide moiety to participate
in electron-transfer reactions involved in antioxidant
effects.
Ion izin g Ra d ia tion Tr ea tm en t. The mean surviv-
ing fraction of V79 cells following a 12-Gy dose of
radiation alone was 0.02 ( 0.002. Possible modulation
of radiation-mediated cytotoxicity by the test compounds
was evaluated using a fixed final concentration of 10
mM, present during the 10 min prior to radiation
exposure. This concentration was selected for several
reasons. From prior results obtained with Tempol, a
final concentration of 10 mM provides partial protection
against radiation-induced cytotoxicity. A final concen-
tration of 10 mM of each compound would therefore
afford comparison to our lead compound and allow
identification of agents with greater protective proper-
ties than Tempol. As stated above, because numerous
compounds were evaluated, only small quantities were
synthesized making the use of higher concentrations
prohibitive. Additionally, our earlier studies showed
The protecting activity is also influenced by ring
saturation and functional groups; for instance, the
protecting activity of substituted pyrroline nitroxides
(21c, 25c, 26c) approaches the efficiency of six-mem-

SAR of Nitroxide Free Radicals and Precursors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18 3485
Ta ble 3. Aldehydes
yield,
%
H2O2
prot
radiat
prot
compd
n
method
A, G
mp, °C
oil
formulaa
C9H15NO
mol wt
153.22
ref
43
36a
0
62, 89
0.9
(0.06)
2.7
(0.29)
1.3
0.07
(0.02)
0.13
(0.30)
2.70
(0.23)
0.60
(0.10)
0.40
(0.00)
1.50
(0.15)
6.10
36b
36c
36d
37a
37b
37c
0
0
0
1
1
1
C
D
H
A
C
69
91
85
45
77
153-158
77-79
oil
C9H15NO2‚HCl
205.68
168.22
167.25
167.25
219.71
182.24
43
C9H14NO2
26, 43
43
(0.00)
C
C
C
C
10H17NO
NTb
oil
10H17NO
43
NT
NT
194-197
10H17NO2‚HCl
10H16NO2
43
27
0.8
(0.00)
(0.25)
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
Ta ble 4. Ketones
yield,
%
H2O2
prot
radiat
prot
compd
n
method
C
mp, °C
formulaa
mol wt
193.67
ref
43
38b
0
68
151-153
C8H15NO2‚HCl
1.1
(0.00)
0.9
(0.00)
0.5
1.0
(0.20)
10.2
(0.45)
1.10
(0.10)
0.68
38c
39a
39b
39c
0
1
1
1
C8H14NO2
C9H17NO
156.20
155.24
207.70
170.23
13
47
(0.00)
C9H17NO2‚HCl
C9H16NO2
35
NTb
(0.15)
7.10
13, 47
1.1
(0.00)
(0.85)
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
Ta ble 5. Ketones
H2O2
prot
radiat
prot
compd
method
A
yield, %
57
mp, °C
oil
formulaa
10H17NO
mol wt
167.25
ref
43
40a
C
C
C
NTb
NT
NT
0.63
(0.13)
0.15
(0.02)
4.80
40b
40c
C
80
181-183
10H17NO2‚HCl
219.71
182.24
43
10H16NO2
48, 49
(0.20)
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
that nitroxides, but not the hydroxylamines, were
radioprotective under aerobic conditions.21 The protec-
tion afforded by the compounds against radiation-
induced cytotoxicity is shown in Tables 1-10 and Figure
3. The studies show that the protective effects of the
test compounds varied from protection factors of 1 to

3486 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Ta ble 6. Acids, Esters, and Amides
Krishna et al.
H2O2
prot
radiat
prot
compd
R
method
yield, %
mp, °C
formulaa
C9H14NO3
mol wt
184.24
ref
13
41c
CO2H
1.0
(0.00)
1.3
(0.35)
2.7
(0.46)
1.5
(0.29)
1.2
(0.00)
1.0
(0.00)
1.2
(0.04)
1.1
NTb
NT
NT
42a
42b
42c
43c
44c
45c
46c
47c
48a
48b
48c
49c
CO2Me
CO2Me
CO2Me
E
C
82
64
oil
C10H17NO2
183.25
235.71
198.24
198.24
183.23
227.28
241.31
241.26
225.33
314.26
240.33
358.41
43
43
13
43
13
43
43
33
19
43
43
43
143-145
C10H17NO3‚HCl
C10H16NO3
0.94
(0.08)
NT
NHCONH2
S
70
189-190
C9H16N3O2
CONH2
C9H15N2O2
10.7
(3.80)
NT
CONH(CH2)2OH
CONH(CH2)3OH
CONHCH2CO2H
CONH(CH2)3NH2
CONH(CH2)3NH2
CONH(CH2)3NH2
CONH(CH2)3NMe2
T1
T1
62
84
99-100
95-96
C11H19N2O3
C12H21N2O3
C11H17N2O4
C12H23N3O
NT
(0.14)
0.9
(0.20)
NT
0.98
(0.02)
0.52
(0.07)
NT
C
66
hygrosc
104-108
94-96
C12H23N3O2‚2HCl
C12H22N3O2
C14H26N3O2‚C2H2O4
1.0
(0.12)
1.1
(0.17)
NT
T1, T2
T1
73, 55
59
NT
15.0
(4.20)
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
Ta ble 7. Acids, Amides, and Imides
H2O2
prot
radiat
prot
compd
R
method
yield, %
mp, °C
formulaa
C9H16NO3
mol wt
186.23
ref
13
50c
CO2H
1.0
NTb
(0.00)
1.7
(0.00)
NT
51c
52b
CONH2
C9H17N2O2
185.25
409.91
13
43
11.2
(1.80)
1.26
C
68
dec
C20H27N3O4‚HCl
(0.00)
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
30 (Figure 4 in Supporting Information). In comparison
to Tempol (5c) which gave a PF of 2.2, 19 compounds
yielded greater PFs. In some cases the agents provided
sensitization of radiation effects with sensitization
factors ranging between 1 and 0.1. A correlation
between the protective effects and the structure and
oxidation state was made to evaluate the determinant
factors governing the radiation-modifying effects.
Among the compounds studied, the nitroxides were
found to be the most effective in providing the protection
followed by the hydroxylamines under identical concen-
trations. Though the nitroxides and hydroxylamines
were both effective in reversing the toxicity mediated
by H2O2, the results obtained from radiation studies
differ presumably as a result of kinetic reasons as well
as differences in the sites of damage.
The possible chemical reactions involved in the pro-
tective reactions for nitroxides are as follows. If TH is
a target molecule which is a critical biomolecule such
as DNA, abstraction of the H atom by the radiation-
induced species (X•) causes the initial lesion.
TH + X• f T• + XH (radical formation) (12)

SAR of Nitroxide Free Radicals and Precursors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18 3487
Ta ble 8. Imides
H2O2
prot
radiat
prot
compd
A
method
C
yield, %
78
mp, °C
formulaa
mol wt
336.82
ref
43
53b
CH2
118-120
C17H20N2O3‚HCl
3.3
(0.52)
1.1
(0.23)
2.2
(0.23)
0.49
(0.03)
1.90
(0.23)
1.30
54a
54b
(CH2)3NHCO
(CH2)3NHCO
C20H25N3O3‚HCl
C20H25N3O4‚HCl
391.90
407.90
19
19
(0.17)
a
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula.
Ta ble 9. Imides
yield,
%
H2O2
prot
radiat
prot
compd
method
A
mp, °C
formulaa
mol wt
286.37
ref
43
55a
54
128-129
C
C
17H22N2O2
0.8
(0.00)
1.5
1.70
(0.29)
55b
C
72
>235
17H22N2O3‚HCl
338.83
43
NTb
(0.14)
a
b
Analyses for C, H, N, S, Br, and Cl are within (0.4% of the expected value for the formula. NT, not tested.
Ta ble 10. Nitrones
H2O2
prot
radiat
prot
compd
R1
R2
R3
formulaa
C6H11NO
mol wt
113.15
ref
44
56
57
58
H
CH3
CH3
0.9
(0.17)
1.1
(0.09)
0.8
0.93
0.91
1.10
CH3
CH3
H
CH3
CH3
C6H11NO
C7H13NO
113.15
127.18
51
52
CH3
(0.12)
In the presence of oxygen or other hypoxic cell radiation
sensitizers such as nitric oxide, the damage on the
molecule is fixed by radical-radical recombination
reactions.
If X• represents the species generated by ionizing
radiation which mediate biological damage, then protec-
tion by radical scavenging can proceed by:
H
+ + X• + RRNO• f
XH + RRNO+ (electron transfer) (15)
T• + O2 f TOO•
(13)
X• + RRNO• f
These intermediates lead to further breakdown products
leading to single- and double-strand breaks in DNA.
However, thiol-based radioprotectors have been shown
to be effective H atom donors and compete with O2 and
restore the damage.
RRNOX (radical-radical recombination) (16)
Protection by chemical repair is also feasible with
nitroxides:
H
+ + T• + RRNO• f TH + RRNO+
(17)
T• + RSH f TH + RS•
(14)
Thus, nitroxides can afford radioprotection by radical
scavenging as well as by chemical repair.
The hydroxylamines can mediate both electron-
transfer and chemical repair reactions and provide
The thiyl radical, being a weak oxidant, is presumed to
be minimally involved in mediating further biologic
damage.

3488 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Krishna et al.
the DNA to similar extent. In a recent study, Milligan
et al.40 show that a diradical intermediate with lifetime
of 1 µs is involved in the formation of dsbs. Hence
agents which can scavenge such intermediates should
effectively inhibit the radiation-induced cytotoxicity.
While both hydroxylamine and nitroxides are effective
radical scavengers, the mechanisms and the kinetics are
expected to be different.
Str u ctu r e-Activity Rela tion sh ip s. The unsub-
stituted base compounds (1c, 2c) afforded modest ra-
dioprotective activity similar to that found for Tempol
(5c). Effective radioprotection from nitroxides was
observed when the substituent contained an amino
group in the side chain. The protection however, was
not found to dependent on the order of the amine;
primary amines (7c-9c), secondary amines (19c), or
tertiary amines (16c, 17c, 22c, 23c, 49c) were all found
to have significant radioprotecting activity. The posi-
tively charged isothiouronium salt (34c) also exhibited
remarkable radioprotective activity.
Only c type nitroxide compounds or in some cases b
type hydroxylamines showed X-ray protecting activity.
The latter compounds, which are N-hydroxy derivatives,
could be active because of the possible propensity to
undergo spontaneous oxidation back to nitroxide. In
addition it was found that upon reducing the nitroxide
free radical function to amine, the protecting activity
disappeared. This effect was independent of the type
of side chain or other functional group.
In addition to the amino substituents, five- and six-
membered nitroxide compounds (c type) containing a
formyl function (36c, 37c) were found to have protecting
activity, but in their amine oxidation state (a type), they
exhibited significant radiosensitization effects (36a ,
37a ).
F igu r e 3. Protection factors of hydrogen peroxide-induced
cytotoxicity as a function of redox potential of selected com-
pounds: (A) six-membered ring nitroxides and (B) five-
membered ring nitroxides.
protective effects.
Oxidation state of the ring appears to be a critical
factor in influencing the protective effects of the com-
pounds. The fully reduced amines (compounds type a )
predominantly exhibited sensitizing effects, whereas the
nitroxides and hydroxylamines exhibited protective
effects. Among the nitroxides tested, a comparison of
the protective effects with the corresponding redox
midpoint potentials shows that there is no significant
correlation between the observed protection and the
redox potentials. This indicates that even for an amine-
based ring oxidation state to be an efficient sensitizer,
DNA seems to be the likely target, and therefore,
accumulation at DNA to sufficient concentrations con-
fers effective radiosensitizing properties.
X• + RRNOH f
HX + RRNO• (electron-transfer) (18)
T• + RRNOH f
TH + RRNO• (chemical repair) (19)
The fully reduced amines were found to provide radi-
osensitization in some cases. These effects are consis-
tent with the reactions listed below:
X• + RRNH f XH + RRN•
X• + TH f T• + XH
(20)
(21)
(22)
Con clu sion s
RRN• + TH f T• + RRNH (damage)
The results from the present study show that the
nitroxide-based antioxidants represent a novel class of
recycling antioxidants which have both catalytic as well
as stoichiometric scavenging effects. A general screen
of their effectiveness against two types of damage such
as H2O2 exposure and ionizing radiation shows that
while they are effective in both cases, the structural
requirements are distinctly different. For protection
against H2O2 exposure, the kinetic effects predominate,
and both six-membered ring and substituted five-
membered ring nitroxides were found to be effective. In
the case of ionizing radiation, accumulation of the agent
at the site of damage was critical. In this case, nitrox-
T• + RRN• f RRN-T (damage fixation) (23)
The ring size was not found to be of significant effect
in determining the radiation response. DNA binding
characteristics of the nitroxide-based radioprotectors
have been evaluated by nonequilibrium dialysis of DNA
with individual nitroxides. These studies show that the
positively charged nitroxides had higher degrees of
association with DNA as compared to the negatively
charged nitroxides. However, since the binding to DNA
should be significantly similar for a given ring substitu-
ent, nitroxide and hydroxylamine should accumulate at

SAR of Nitroxide Free Radicals and Precursors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18 3489
and dried (0.53 g, 70%) as off-white crystals: mp 215-216 °C;
TSP [M + H]+ 304. Anal. (C19H32N2O‚2HCl) C, H, N, Cl.
Compounds 1b, 2b, 5b, 6b, 9b, 11b-17b, 19b, 22b, 23b,
26b, 27b, 29b, 36b-38b, 40b, 42b, 48b, 52b, 53b, and 55b
were prepared following a procedure similar to method C.
Meth od D. 1-Oxyl-3-for m yl-2,2,5,5-tetr a m eth yl-3-p yr -
r olin e (36c). N-Hydroxylamine hydrochloride salt 36b (2.06
g, 10 mmol) was dissolved in saturated K2CO3 solution (15 mL)
and extracted with CHCl3. To the organic phase were added
drying agent (MgSO4) and catalytic amount of MnO2 (50 mg),
and this mixture was aerated for 30 min. Then the reaction
mixture was filtered and evaporated. The crystalline product
36c was filtered off and evaporated with hexane (1.53 g, 91%)
as yellow crystals: mp 77-79 °C; EI (m/z, %) 168 (M+, 35),
138 (41), 123 (100), 95 (88). Anal. (C9 H14NO2) C, H, N.
Meth od E. Meth yl (2,2,5,5-Tetr a m eth yl-3-p yr r olin -3-
yl)ca r boxyla te (42a ). 2,2,6,6-Tetramethyl-3,5-dibromo-4-
piperidone hydrobromide13 (59) (39.39 g, 0.1 mol) was added
during about 1 h to a stirred methanol solution of sodium
methoxide (30%) (57 mL, 0.3 mol) at room temperature; then
the formed NaBr was filtered off, the filtrate evaporated to
dryness, and the residual pale oil distilled under reduced
pressure (1.3 mmHg, 56-58 °C) to yield the pure product 42a
(15.0 g, 82%) as an oil: EI (m/z, %) 183 (M+, 1), 168 (100), 136
(49), 108 (64). Anal. (C10H17NO2) C, H, N.
Meth od F . 3-(Hyd r oxym eth yl)-2,2,5,5-tetr a m eth yl-3-
p yr r olin e (14a ). To a solution of ester 42a (1.83 g, 10 mmol)
in toluene (10 mL) was added SMEAH (70% solution in
toluene, 6.4 mL, 23 mmol) dropwise in argon atmosphere. The
reaction mixture was kept at 40 °C for 1 h and then added to
a NaOH solution (20%, 20 mL). THF (10 mL) was also added,
and the organic phase was separated, dried (MgSO4), and
evaporated. The crystalline product 14a was filtered and
washed with hexane (1.21 g, 78%) as pale-yellow crystals: mp
66-69 °C; EI (m/z, %) 155 (M+, 1), 140 (40), 122 (18), 110 (100).
Anal. (C9H17NO) C, H, N.
ides with basic side groups were most effective. With
respect to toxicity, none of the agents evaluated in the
present study exhibited cytotoxicity. The toxicity of only
a few selected nitroxides has been evaluated in mice
where sufficient concentrations of the nitroxides for
antioxidant and radioprotective purposes were achieved
without untoward toxicity.41,42 The results from this
screen will hopefully provide optimal candidates for
more intensive evaluation as agents with respect to
toxicity and for chemical modification of damaging
agents such as H2O2 and ionizing radiation.
Exp er im en ta l Section
The syntheses of the compounds are described by way of
typical examples: the methods are denoted in the same way
as in Schemes 1-4. The structural and molecular formulas
and physical properties of the compounds are summarized in
Tables 1-10.
Melting points (uncorrected) were determined on a Boetius
micromelting point apparatus. Elemental analyses (C, H, N,
S) were obtained on a Carlo-Erba EA 1110 apparatus or (Hal)
determined titrimetrically by Scho¨niger’s method. Mass spec-
tra were recorded on a VG TRIO-2 instrument in the EI mode
(70 eV, direct inlet) or with thermospray technique. Samples
were analyzed in the bypass mode; 10 µL of the sample
solution in CH3OH was introduced via the thermospray
interface. The mobile phase was CH3OH/H2O, 1:1, solution
containing 0.1 M NH4OAc. The capillary tip temperature was
230 °C, the electrode voltage was 180 V, and the source
temperature was 210 °C. The ESR spectra were obtained from
10-5 M solution, using a BRUKER 300-E spectrometer. All
the monoradicals exhibited three equidistant lines with aN
14.7-15.5 G. Flash column chromatography was performed
with silica gel 60 (230-400 mesh) (Merck).
)
Met h od A. 3-[N-(1-Ad a m a n t yl)a m in om et h yl]-2,2,5,5-
tetr a m eth yl-3-p yr r olin e (25a ). The nitroxide 25c (3.04 g,
10 mmol) was dissolved in AcOH (10 mL); then Fe powder (2
g, 36 mmol) was added. The reaction mixture was stirred at
40-50 °C for 3 h, diluted with water (10 mL), and filtered from
unreacted Fe. The filtrate was basified by adding solid K2CO3,
then extracted with CHCl3 (3 × 20 mL), dried (MgSO4), and
filtered again, and the solvent was evaporated. The residue
was flash-chromatographed on silica gel (Merck 60) with
CHCl3-Et2O and evaporated to dryness. The product was
crystallized from Et2O-hexane, to give the base 25a (1.78 g,
62%) as off-white crystals: mp 116-118 °C; EI (m/z, %) 288
(M+, 2), 273 (65), 135 (100), 122 (81). Anal. (C19H32N2) C, H,
N.
Preparation of hydrochloride salt: The base (0.86 g, 3 mmol)
was acidified with HCl in EtOH to pH 3 and then diluted with
Et2O to start crystallization (0.92 g, 85%). The analytical
sample of the salt of 25a /HCl was recrystallized from EtOH-
Et2O (mp 226 °C dec). Anal. (C19H34N2Cl2) C, H, N, Cl.
Compounds 14a , 16a , 22a , 23a , 36a , 37a , 40a , and 55a
were prepared following a procedure similar to method A.
Met h od B. 1-Oxyl-4-(N-p h t h a lim id -1-yl)-2,2,6,6-t et -
r a m eth ylp ip er id in e (55c). To a solution of amine 55a (2.86
g, 10 mmol) in methanol (10 mL) was added 30% H2O2 solution
(5 mL, 50 mmol), and the solution was allowed to stand for 3
days. Then the reaction mixture was evaporated to one-half
volume and extracted with CHCl3. The organic phase was
dried (MgSO4) and evaporated. The product was crystallized
from CHCl3-hexane to give nitroxide 55c (2.17 g, 72%) as pale-
yellow crystals: mp 134-136 °C; EI (m/z, %) 301 (M+, 2), 287
(20), 147 (85), 76 (100). Anal. (C17H21N2O3) C, H, N.
Meth od C. 1-Hyd r oxy-3-[N-(1-a d a m a n tyl)a m in om eth -
yl]-2,2,5,5-tetr a m eth yl-3-p yr r olin e (25b). The paramag-
netic compound 25c (0.61 g, 2 mmol) was dissolved in EtOH
saturated with HCl gas (10 mL), then refluxed until the color
of nitroxide disappeared (about 30 min), and diluted with Et2O
to start crystallization. The crystalline hydrochloride 25b that
separated on refrigeration was filtered off, washed with Et2O,
Met h od G. 3-F or m yl-2,2,5,5-t et r a m et h yl-3-p yr r olin e
(36a ). A stirred solution of 14a (1.55 g, 10 mmol) in CHCl3
(30 mL) was refluxed with active MnO2 (5 g, 56 mmol) until
oxidation of alcohol 14a was completed and then filtered, the
filtrate evaporated to dryness, and the residual solidified when
cooled to give the aldehyde 36a (1.36 g, 89%) as a pale-yellow
oil: EI (m/z, %) 153 (M+, 1), 138 (48), 110 (100), 95 (44). Anal.
(C9H15NO) C, H, N.
Meth od H. 3-F or m yl-1,2,2,5,5-p en ta m eth yl-3-p yr r o-
lin e (36d ). A solution of aldehyde 36a 25 (0.77 g, 5 mmol) and
MeI (1.42 g, 10 mmol) in acetone (10 mL) was refluxed for 4 h
and then evaporated to dryness. The residual solid was
dissolved in water, basified with solid K2CO3, extracted with
CHCl3 (3 × 20 mL), dried (MgSO4), and evaporated to dryness.
The oily residue was flash-chromatographed on silica gel
(Merck 60) to get the pure compound 36d (0.71 g, 85%): EI
(m/z, %) 167 (M+, 11), 152 (41), 124 (100), 109 (42). Anal.
(C10H17NO) C, H, N.
Meth od I. 1-Oxyl-3-[N-(1-a d a m a n tyl)a m in om eth yl]-
2,2,5,5-tetr a m eth yl-3-p yr r olin e (25c). A solution of alde-
hyde 36c26 (1.68 g, 10 mmol) and 1-aminoadamantane (1.66
g, 11 mmol) in toluene was refluxed in the presence of catalytic
amount of p-toluenesulfonic acid, in a flask equipped with a
Dean-Stark water separator. After the separation of the
calculated amount of water (about 12 h), the mixture was
concentrated under reduced pressure (water pump). The
remaining Schiff base was dissolved in EtOH (20 mL) and
added slowly to a stirred suspension of NaBH4 (0.57 g, 15
mmol) in EtOH (30 mL) at 0 °C; then the mixture refluxed for
1 h. It was diluted with water (50 mL) and evaporated in a
vacuum to about one-half volume, and the oily aqueous residue
was extracted with CHCl3 (3 × 50 mL). The extract was dried
(MgSO4), filtered, and concentrated. The crude base 25c was
recrystallized from a mixture of Et2O and hexanes (2.18 g,
72%) as yellow crystals: mp 109-111 °C; EI (m/z, %) 303 (M+,
10), 273 (34), 164 (48), 135 (100). Anal. (C19H31N2O) C, H, N.
The base was converted into its tosylate salt. To the base
in acetone was added an equivalent amount of p-toluene-

3490 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Krishna et al.
sulfonic acid in acetone (10 mL), and the mixture was cooled.
The precipitated salt 25c/Ts was filtered off, washed with
Et2O, and dried to give 25c/Ts (3.76 g, 79%) as yellow
crystals: mp 250-251 °C. Anal. (C19H31N2O‚C7H8O3S) C, H,
N, S.
Compound 30d was prepared following a procedure similar
to method N.
Meth od O. 1-Oxyl-3-(tr iph en ylph osph on iu m ylm eth yl)-
2,2,5,5-tetr a m eth yl-3-p yr r olin e Br om id e (32e). A solution
of 1-oxyl-3-(bromomethyl)-2,2,5,5-tetramethyl-3-pyrroline31 (60)
(2.33 g, 10 mmol) and triphenylphosphine (2.62 g, 10 mmol)
was heated in acetone (10 mL) for 3 h and then diluted with
Et2O to start crystallization. The solid residue was recrystal-
lized from CHCl3-Et2O to give the pure compound 32e (3.51
g, 71%) as pale-brown crystals: mp 224-226 °C; TSP 416 (M
+ H)+. Anal. (C27H30BrNOP) C, H, N, Br.
Meth od P . 1-Oxyl-N-(2-isoin d olin ylm eth yl)-2,2,5,5-tet-
r a m eth yl-3-p yr r olin e (24c). A solution of 1-oxyl-3-(ami-
nomethyl)-2,2,5,5-tetramethyl-3-pyrroline33 (61) (1.69 g, 10
mmol), and o-dibromoxilol (2.64 g, 10 mmol), and K2CO3 (2.76
g, 20 mmol) in CHCl3 (50 mL) was stirred and refluxed for 3
h and then evaporated, and the residue was diluted with water
(20 mL) and extracted with Et2O (3 × 20 mL). The extract
was dried (MgSO4), filtered, and concentrated. The residue
was purified by chromatography (CHCl3-Et2O) to give the
base 24c (1.85 g, 68%) as pale-yellow crystals: mp 89-90 °C;
TSP 272 (M + H)+. Anal. (C17H23N2O) C, H, N.
Meth od Q. 1-Oxyl-3-(gu a n id in om eth yl)-2,2,5,5-tetr a m -
eth yl-3-p yr r olin e Hyd r ogen Su lfa te Sa lt (33c). A suspen-
sion of 1-oxyl-3-(aminomethyl)-2,2,5,5-tetramethyl-3-pyrro-
line33 (61) (340 mg, 2 mmol) and S-methylisothiouronium
sulfate (278 mg, 1 mmol) in EtOH (10 mL) was refluxed for 3
h, then the hot reaction mixture was filtered, and the filtrate
was diluted with Et2O to precipitate off-white crystalline
product 33c (287 mg, 55%) as yellow crystals: mp >180 °C
dec; TSP 212 (M + H)+. Anal. (C10H19N4O‚1/2H2SO4) C, H, N,
S.
Meth od R. 1-Oxyl-4-(m eth ylsu lfon yla m in o)-2,2,6,6-tet-
r a m eth ylp ip er id in e (13c). To a stirred solution of amino
compound 9c28 (342 mg, 2 mmol) and triethylamine (TEA) (303
mg, 3 mmol) in dry CH2Cl2 (20 mL) was added methanesulfo-
nyl chloride (252 mg, 2.2 mmol) at 0 °C, and the reaction
mixture was stirred at room temperature for 3 h, then washed
with saturated aqueous NaCl solution (10 mL), dried (MgSO4),
and evaporated to dryness. The solid residue was recrystal-
lized from CHCl3-hexane to give the pure orange compound
13c (409 mg, 82%) as yellow crystals: mp 141-143 °C; EI (m/
z, %) 249 (M+, 3), 235 (23), 124 (61), 84 (100). Anal.
(C10H21N2O3S) C, H, N, S.
Meth od S. 1-Oxyl-3-u r eid o-2,2,5,5-tetr a m eth yl-3-p yr -
r olin e (43c). To a cyclohexane solution of the freshly pre-
pared isocyanate 6233 (0.91 g, 5 mmol) was introduced ammo-
nia gas. The precipitated solid urea 43c was filtered off and
recrystallized from CHCl3-Et2O to give a pure compound (0.69
g, 70%) as yellow crystals: mp 189-190 °C; EI (m/z, %) 198
(M+, 7), 168 (41), 110 (58), 41 (100). Anal. (C9H16N3O2) C, H,
N.
Meth od T1. 1-Oxyl-N-(3-h yd r oxyp r op yl)-3-ca r boxa -
m id o-2,2,5,5-tetr a m eth yl-3-p yr r olin e (46c). To a solution
of 1-oxyl-3-[(ethoxycarbonyl)oxycarbonyl]-2,2,5,5-tetramethyl-
3-pyrroline34 (63) (2.56 g, 10 mmol) in dry CHCl3 (20 mL) was
added 3-aminopropanole (0.75 g, 10 mmol), and the mixture
refluxed for 30 min. Then evaporation to dryness and recrys-
tallization from Et2O-hexane gave 46c (2.03 g, 84%) as yellow
crystals: mp 95-96 °C; EI (m/z, %) 241 (M+, 2), 211 (32), 136
(32), 110 (100). Anal. (C12H21N2O3) C, H, N.
Compounds 18c-20c, and 28c were prepared following a
procedure similar to method I.
Meth od J . 1-Oxyl-3-[N-m eth yl-N-(1-a d a m a n tyl)a m i-
n om eth yl]-2,2,5,5-tetr a m eth yl-3-p yr r olin e (26c). To a
stirred solution of 25c (608 mg, 2 mmol) in dry tetrahydrofuran
(THF) (20 mL) in argon atmosphere was added NaH (60 mg,
2.5 mmol), and the mixture refluxed for 1 h. After MeI (284
mg, 2.0 mmol) was added to the reaction mixture and refluxed
for 6 h, the cooled reaction mixture was washed with saturated
NaCl solution (10 mL), and the organic phase was dried
(MgSO4), filtered, and evaporated. The residue was chromato-
graphed (hexanes-Et2O) to separate the nonmethylated start-
ing compound. The pure compound 26c was recrystallized
from CHCl3-hexanes (394 mg, 62%) as yellow crystals: mp
110-112 °C; EI (m/z, %) 317 (M+, 2), 287 (5), 165 (55), 135
(100). Anal. (C20H33N2O) C, H, N.
Compound 29c was prepared following a procedure similar
to method J .
Meth od K. 1-Oxyl-4-(N,N-n -d ibu tyla m in o)-2,2,6,6-tet-
r a m eth ylp ip er id in e (10c). To solution of ketone 39c (TEM-
PONE)44 (1.70 g, 10 mmol) and di-n-butylamine hydrochloride
(6.63 g, 40 mmol) in dry methanol (100 mL) was added
NaCNBH3 (3.77 g, 60 mmol), and the reaction mixture was
refluxed for 48 h and finally evaporated to dryness. The
residue was taken up in water (30 mL), acidified by adding
diluted H2SO4 (5%), and extracted with CHCl3 (3 × 20 mL) to
remove the unreacted ketone. The aqueous phase was basified
by adding solid K2CO3, extracted again with CHCl3 (3 × 20
mL), then dried, evaporated, and purified by flash chroma-
tography to give the product 10c (0.99 g, 35%) as a deep-red
thick oil: TSP 284 (M + H)+. Anal. (C17H35N2O) C, H, N.
Meth od L. 1-Oxyl-3-(eth oxym eth yl)-2,2,5,5-tetr am eth yl-
3-p yr r olin e (15c). To a suspension of NaH (0.24 g, 10 mmol)
in dry THF (15 mL) was added alcohol 14c26 (1.70 g, 10 mmol)
dropwise in dry THF (10 mL), and the mixture was stirred at
room temperature for 1 h. After that EtBr (1.09 g, 10 mmol)
in dry DMF (10 mL) was added dropwise, and the reaction
mixture was refluxed for 2 h. Then the cooled reaction mixture
was extracted with brine (20 mL), and the organic phase was
dried (MgSO4), filtered, and evaporated. The residue was
flash-chromatographed (hexanes-Et2O) to give the pure prod-
uct 15c (1.13 g, 57%) as a yellow oil: EI (m/z, %) 198 (M+, 10),
138 (9), 122 (26), 107 (100). Anal. (C11H20NO2) C, H, N.
Meth od M. 1-Oxyl-3-(N-p yr r olid in ylm eth yl)-2,2,5,5-
tetr a m eth yl-3-p yr r olin e (22c). A solution of 1-oxyl-3-(bro-
momethyl)-2,2,5,5-tetramethyl-3-pyrroline31 (60) (2.33 g, 10
mmol) and pyrrolidine (2.13 g, 30 mmol) in CHCl3 (30 mL)
was refluxed for 3 h and then evaporated, and the residue was
diluted with water (20 mL) and extracted with Et2O (3 × 20
mL). The extract was dried (MgSO4), filtered, and concen-
trated. The residue was purified by chromatography (hex-
anes-Et2O) to give the base 22c (1.74 g, 78%) as a red oil.
Anal. (C13H23N2O) C, H, N.
The pure base 22c was converted to its tosylate salt (3.08
g, quantitatively) in the same way as described in method I:
yellow crystals; mp 129-132 °C; TSP 224 [M + H]+. Anal.
(C13H23N2O‚C7H8O3S) C, H, N, S.
Compounds 17c and 21c were prepared following a proce-
dure similar to method M.
Met h od N. 1-Oxyl-3-(t r ib u t yla m m on iu m ylm et h yl)-
2,2,5,5-tetr a m eth yl-3-p yr r olin e Br om id e (31d ). A solution
of 1-oxyl-3-(bromomethyl)-2,2,5,5-tetramethyl-3-pyrroline31 (60)
(2.33 g, 10 mmol) and the excess of n-tributylamine (3.71 g,
20 mmol) was heated in acetone (10 mL) for 6 days and then
diluted with Et2O to start crystallization. The solid residue
was recrystallized from acetone-Et2O to give the pure com-
pound 31d (2.64 g, 63%) as yellow crystals: mp 180-181 °C;
EI (m/z, %) 338 (M+, 1), 281 (2), 202 (8), 142 (100). Anal.
(C21H42BrN2O) C, H, N, Br.
Compounds 45c, 48c, and 49c were prepared following a
procedure similar to method T1.
Meth od T2. 1-Oxyl-N-(3-a m in op r op yl)-3-ca r boxa m id o-
2,2,5,5-tetr a m eth yl-3-p yr r olin e (48c). To a stirred solution
of 1,3-diaminopropane (2.22 g, 30 mmol) in dry THF (50 mL)
was added 1-oxyl-3-(N-imidazolylcarbonyl)-2,2,5,5-tetramethyl-
3-pyrroline26 (64) (2.56 g, 10 mmol) at 0 °C. Then the reaction
mixture was warmed to room temperature, 1 h later evapo-
rated, and flash-chromatographed with CHCl3-Et2O. The
first band was the biradical (0.43 g; mp 177-178 °C), and the
second yellow band was the monoradical 48c (1.32 g, 55%) as
yellow crystals: mp 104-108 °C; EI (m/z, %) 240 (M+, 8), 226
(18), 136 (65), 41 (100). Anal. (C12H22N3O2) C, H, N.

SAR of Nitroxide Free Radicals and Precursors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18 3491
Meth od U. 1-Oxyl-3-[N-(3-a m in op r op yl)-3-ca r boxa m -
id o]-2,2,5,5-tetr a m eth ylp yr r olid in e (66). To a stirred solu-
tion of 1,3-diaminopropane (2.22 g, 30 mmol) in water-THF
(2:1, 20 mL) was added 1-oxyl-3-bromo-4-oxo-2,2,6,6-tetram-
ethylpiperidine35 (65) (2.49 g, 10 mmol) in portions. The
solution was extracted after stirring for 2 h with CHCl3. The
organic phase was dried, evaporated, and purified by flash
chromatography to give the product 66 (1.50 g, 62%) as yellow
crystals: mp 102-105 °C; EI (m/z, %) 242 (M+, 18), 212 (11),
156 (54), 41 (100). Anal. (C12H24N3O2) C, H, N.
acetic acid (3:1) and stained with crystal violet. Colonies
containing >50 cells were scored.
The results for H2O2 treatment were formulated to express
a protection factor (PF, derived by dividing the surviving
fraction of nitroxide plus H2O2 treatment by the surviving
fraction of H2O2 treatment alone). A PF > 1.0 means that the
agent provided protection against H2O2 cytotoxicity. PF values
were also derived for radiation studies (PF, derived by dividing
the surviving fraction at 12 Gy of nitroxide treatment by the
surviving fraction of 12 Gy of treatment alone). PFs > 1.0
mean that the agent provided radiation protection, and values
< 1.0 mean that radiation sensitization occurred. The PFs
from individual experiments were pooled to find the mean PF
( standard deviation. Each nitroxide was then compared to
the untreated irradiated control (PF ) 1.0) using a two-tailed
paired Students t-test to determine statistical significance.45
Meth od V. N-[(1-Oxyl-2,2,5,5-tetr a m eth ylp yr r olid in e-
3-ca r boxa m id o)p r op yl]p h th a lim id e (52c). A mixture of
equivalent amounts of 1-oxyl-3-[N-(3-aminopropyl)-3-carboxa-
mido]-2,2,5,5-tetramethylpyrrolidine19 (66) (2.42 g, 10 mmol)
and phthalimide (1.47 g, 10 mmol) was heated at 120 °C
without any solvent until the evolution of ammonia had ceased
(about 60 min). After cooling the melt was dissolved in CHCl3
(20 mL) and flash-chromatographed on silica gel (eluted with
CHCl3-Et2O). The pure product was crystallized from CHCl3-
hexane to give 52c (2.15 g, 58%) as yellow crystals: mp 118-
120 °C; EI (m/z, %) 372 (M+, 15), 358 (9), 342 (89), 41 (100).
Anal. (C20H26N3O4) C, H, N.
Ack n ow led gm en t. This work was supported in part
by grants from the Hungarian National Research Foun-
dation (T 021277 and T 017842).
Su p p or tin g In for m a tion Ava ila ble: Graphical presenta-
tion of protection factors against H2O2- and radiation-induced
cytotoxicity (Figures 2 and 4) (2 pages). Ordering information
is given on any current masthead page.
Meth od W. 1-Oxyl-3-(N-p h th a lim id -1-ylm eth yl)-2,2,5,5-
tetr a m eth yl-3-p yr r olin e (53c). A vigorously stirred solution
of phthalimide (1.47 g, 10 mmol), 1-oxyl-3-(bromomethyl)-
2,2,5,5-tetramethyl-3-pyrroline31 (60) (2.33 g, 10 mmol), pow-
dered K2CO3 (2 g), powdered KOH (100 mg), and 18-crown-6
(300 mg) in dioxane (30 mL) was refluxed for 4 h and filtered,
and the filtrate was evaporated to dryness. Residue was flash-
chromatographed with CHCl3-Et2O and then recrystallized
from CHCl3-hexane to give 53c (2.18 g, 73%) as yellow
crystals: mp 127-129 °C; EI (m/z, %) 299 (M+, 13), 285 (27),
269 (100), 213 (42). Anal. (C17H19N2O3) C, H, N.
Refer en ces
(1) Chance, B.; Sies, H.; Boveris, A. Hydroperoxide metabolism in
mammalian organs. Physiol. Rev. 1979, 59, 527-605.
(2) Imlay, J . A.; Chin, S. M.; Linn, S. Toxic DNA damage by
hydrogen peroxide through the Fenton reaction in vivo and in
vitro. Science 1988, 240, 640-642.
(3) Halliwell, B.; Gutteridge, J . M. C. Free Radicals in Biology and
Medicine; Clarendon Press: 1989; pp 210-211.
(4) Swartz, H. M. Principles of the metabolism of nitroxides and
their implications for spin trapping. Free Radical Res. Commun.
1990, 9, 399-405.
(5) Paleos, C. M.; Pais, D. Ready Reduction of Some Nitroxide
Radicals with Ascorbic Acid. J . Chem. Soc., Chem. Commun.
1977, 428.
(6) Krishna, M. C.; Grahame, D. A.; Samuni, A.; Mitchell, J . B.;
Russo, A. Oxoammonium cation intermediate in the nitroxide-
catalyzed dismutation of superoxide. Proc. Natl. Acad. Sci.
U.S.A. 1992, 89, 5537-5541.
(7) Rozantsev, E. G.; Kagan, E. S. H.; Sholle, V. D. Triacetonamine
in the Chemistry of Nitroxyl Radicals. In Bioactive Spin Labels;
Zhdanov, R. I., Ed.; Springer-Verlag: Berlin, 1992; pp 83-118.
(8) Krishna, M. C.; Russo, A.; Mitchell, J . B.; Goldstein, S.; Dafni,
H.; Samuni, A. Do nitroxides antioxidants act as scavengers of
superoxide or as SOD mimics? J . Biol. Chem. 1996, 271, 26026-
26031.
Meth od Z. 1-Oxyl-4-(N-p h th a lim id -1-yl)-2,2,6,6-tetr a -
m eth ylp ip er id in e (55c). A mixture of equivalent amounts
of 1-oxyl-4-amino-2,2,6,6-tetramethylpiperidine44 (9c) (1.71 g,
10 mmol) and phthalimide (1.47 g, 10 mmol) was heated at
120 °C without any solvent until the evolution of ammonia
had ceased (about 50 min). After cooling the melt was
dissolved in CHCl3 (20 mL) and flash-chromatographed on
silica gel (eluted with CHCl3-Et2O). The pure product was
crystallized from CHCl3-hexane to give 55c (2.01 g, 67%) as
yellow crystals: mp 134-136 °C; EI (m/z, %) 301 (M+, 2), 287
(20), 147 (85), 76 (100). Anal. (C17H21N2O3) C, H, N.
Electr och em istr y. Cyclic voltammetry was performed on
the nitroxides in phosphate-buffered saline using a glassy
carbon working electrode and a platinum auxiliary electrode.
All other conditions are similar to that reported.6
(9) Krishna, M. C.; Samuni, A.; Taira, J .; Goldstein, S.; Mitchell, J .
B.; Russo, A. Stimulation by nitroxides of catalase-like activity
of hemeproteins. J . Biol. Chem. 1996, 271, 26018-26025.
(10) Nooy, A. E. J .; Besemer, A. C.; van Bekkum, H. On the Use of
Stable Organic Nitroxyl Radicals for the Oxidation of Primary
and Secondary Alcohols. Synthesis 1996, 1153-1174.
(11) Zhdanov, R. I.; Komarov, P. G. Sterically hindered hydroxy-
lamines as bioactive spin labels. Free Radical Res. Commun.
1990, 9, 367-377.
(12) Chateauneuf, J .; Lusztyk, J .; Ingold, K. U. Absolute rate
constants for the reactions of some carbon-centered radicals with
2,2,6,6-tetramethylpiperidine-N-oxyl. J . Org. Chem. 1988, 53,
1629-1632.
(13) Rozantsev, E. G. Free Nitroxyl Radicals; Plenum Press: New
York, 1970.
(14) Mitchell, J . B.; Krishna, M. C.; Samuni, A.; Russo, A.; Hahn, S.
M. Nitroxides as Protectors Against Oxidative Stress. Reactive
Oxygen Species in Biological Systems: An Interdisciplinary
Approach; 1997, in press.
(15) Mitchell, J . B.; Samuni, A.; Krishna, M. C.; DeGraff, W. G.; Ahn,
M. S.; Samuni, U.; Russo, A. Biologically active metal-indepen-
dent superoxide dismutase mimics. Biochemistry 1990, 29,
2802-2807.
(16) Samuni, A.; Mitchell, J . B.; DeGraff, W.; Krishna, C. M.; Samuni,
U.; Russo, A. Nitroxide SOD-mimics: Modes of action. Free
Radical Res. Commun. 1991, 12-13, 187-194.
(17) Krishna, C. M.; DeGraff, W.; Tamura, S.; Gonzalez, F.; Samuni,
A.; Russo, A.; Mitchell, J . B. Mechanisms of hypoxic and aerobic
cytotoxicity of Mitomycin C in Chinese hamster V79 cells. Cancer
Res. 1991, 51, 6622-6628.
Cell Cu ltu r e. Chinese hamster V79 cells were grown in
F12 medium supplemented with 10% fetal calf serum, penicil-
lin, and streptomycin. Survival was assessed in all studies
by the clonogenic assay. The plating efficiency ranged between
80% and 90%. Stock cultures of exponentially growing cells
were trypsinized, rinsed, and plated (5 × 105 cells/dish) into a
number of 100-mm Petri dishes and incubated for 16 h at 37
°C prior to experimental protocols. Nitroxides were added to
exponentially growing cells in complete F12 medium (final
concentration, 100 µM) at room temperature immediately prior
to treatment with H2O2 (final concentration, 500 µM) for 1 h.
For radiation studies, nitroxides were added to exponentially
growing cells in complete F12 medium (final concentration,
10 mM) at room temperature 10 min prior to a single radiation
dose of 12 Gy. Plates were irradiated at room temperature
with a cobalt-60 irradiator at a dose rate of 5.0 Gy/min. Full
electron equilibrium was ensured for all irradiations. The time
required for irradiation (at room temperature) was approxi-
mately 10 min. Immediately after treatment, cells were
rinsed, trypsinized, counted, and plated for macroscopic colony
formation. Using these conditions, none of the nitroxides
exerted cytotoxicity alone. Each nitroxide/H2O2 or nitroxide/
radiation determination was plated in triplicate, and experi-
ments were repeated a minimum of two times. Plates were
incubated for 7 days; colonies were then fixed with methanol/

3492 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 18
Krishna et al.
(18) Sosnovsky, G.; Bell, P. In the search for new anticancer drugs.
29. A study on the correlation of lipophilicities, ionization
constants and anticancer activities of aminoxyl labeled TEPA
congeners. Life Sci. 1998, 62, 639-648.
(19) Hankovszky, H. O.; Hideg, K.; Bo´di, I.; Frank, L. New Antiar-
rhythmic Agents. 2,2,5,5-Tetramethyl-3-pyrroline-3-carboxam-
ides and 2,2,5,5-tetramethylpyrrolidine-3-carboxamides. J . Med.
Chem. 1986, 29, 1138-1152.
(20) Twomey, P.; Taira, J .; DeGraff, W.; Mitchell, J . B.; Russo, A.;
Krishna, M. C.; Hankovszky, H. O.; Frank, L.; Hideg, K. In Vivo
Nitroxide Free Radical Production from the new Anti-arrhythmic
Drug as Evidenced by EPR Spectroscopy. Free Radical Biol. Med.
1997, 22, 909-916.
(21) Mitchell, J . B.; DeGraff, W.; Kaufman, D.; Krishna, M. C.;
Samuni, A.; Finkelstein, E.; Ahn, M. S.; Hahn, S. M.; Gamson,
J .; Russo, A. Inhibition of oxygen-dependent radiation-induced
damage by the nitroxide superoxide dismutase mimic, Tempol.
Arch. Biochem. Biophys. 1991, 289, 62-70.
(22) Bacq, Z. M. Chemical Protection Against Ionizing Radiation;
Thomas: Springfield, 1965.
(23) Aguilera, J . A.; Newton, G. L.; Fahey, R. C.; Ward, J . F. Thiol
uptake by Chinese hamster V79 cells and aerobic radioprotection
as a function of net charge on the thiol. Radiat. Res. 1992, 130,
194-204.
(24) Bump, E. A.; Foye, W. O. Radiosensitizers and Radioprotective
Agents. In Burger’s Medicinal Chemistry and Drug Discovery,
5th ed.; Wolff, M. E., Ed.; Wiley: New York, 1997; Vol. 4, pp
3-69.
(25) Sa´r, P. C.; Ka´lai, T.; Ba´ra´cz, M. N.; J erkovich, Gy.; Hideg, K.
Selective Reduction of Nitrones and Nitroxides to Functionalized
Secondary Amines. Synth. Commun. 1995, 25, 2929-2940.
(26) Hideg, K.; Hankovszky, H. O.; Lex, L.; Kulcsa´r, Gy. Nitroxyls;
VI. Synthesis and Reactions of 3-Hydroxymethyl-2,2,5,5-tetram-
ethyl-2,5-dihydropyrrole-1-oxyl and 3-Formyl Derivatives. Syn-
thesis 1980, 911-914.
(27) Csek_o, J .; Hankovszky, H. O.; Hideg, K. Synthesis of Novel,
Highly Reactive 1-Oxyl- 2,2,6,6-tetramethyl-1,2,5,6-tetrahydro-
pyridine Derivatives. Can. J . Chem. 1985, 63, 940- 943.
(28) Rauckman, E. W.; Rosen, G. M.; Hord, W. W. Use of Sodium
Cyanoborohydride in the Synthesis of Biradical Nitroxides. Org.
Prep. Proc. Int. 1977, 9, 53-56.
(29) Rosen, M. G. Use of Sodium Cyanoborohydride in the Prepara-
tion of Biologically Active Nitroxides. J . Med. Chem. 1974, 17,
358-360.
(35) Sosnovsky, G.; Cai, Z. A. Study of the Favorskii Rearrangement
with 3-Bromo-4-oxo-2,2,6,6-teramethylpiperidin-1-oxyl. J . Org.
Chem. 1995, 60, 3414-3418.
(36) Hideg, K.; Hankovszky, H. O.; Frank, L.; Bo´di, I.; Csa´k, J .
2-[(2,2,5,5-Tetramethyl-3-pyrrolin-3-carbonyl)]-amino Deriva-
tives. U.S. Patent 4 731 376, 1988.
(37) Kieber, D. J .; Blough, N. V. Fluorescence detection of carbon-
centered radicals in aqueous solution. Free Radical Res. Com-
mun. 1990, 10, 109-117.
(38) Dikalov, S.; Skatchkov, M.; Fink, B.; Bassenge, E. Quantification
of Superoxide Radicals and Peroxynitrite in Vascular Cells Using
Oxidation of Sterically Hindered Hydroxylamines and Electron
Spin Resonance. Nitric Oxide 1997, 1, 423-431.
(39) Afzal, V.; Brasch, R. C.; Niteczki, D. E.; Wolff, S. Nitroxyl spin
labels contrast enhancers for magnetic resonance imaging.
Studies of acute toxicity and mutagenesis. Invest. Radiol. 1984,
19, 549-552.
(40) Milligan, J . R.; Ng, J . Y.; Wu, C. C.; Aguilera, J . A.; Fahey, R.
C.; Ward, J . F. DNA repair by thiols in air shows two radicals
make a double strand break. Radiat. Res. 1995, 143, 273-280.
(41) Hahn, S. M.; Tochner, Z.; Krishna, C. M.; Glass, J .; Wilson, L.;
Samuni, A.; Sprague, M.; Venzon, D.; Glatstein, E.; Mitchell, J .
B. Tempol, a stable free raidcal, is a novel murine radiation
protector. Cancer Res. 1992, 52, 1750-1753.
(42) Hahn, S. M.; Wilson, L.; Krishna, M. C.; Liebmann, J .; DeGraff,
W.; Gamson, J .; Samuni, A.; Venzon, D.; Mitchell, J . B. Identi-
fication of nitroxide radioprotectors. Radiat. Res. 1992, 132, 87-
93.
(43) This article.
(44) Aldrich Catalogue Handbook of Fine Chemicals; Aldrich: Mil-
waukee, WI, 1996-1997.
(45) Snedecor, G. W.; Cochran, W. G. Statistical Methods; Iowa State
University: Ames, 1980; pp 83-106.
(46) Hideg, K.; Sa´r, C. P.; Hankovszky, O. H.; J erkovich, Gy. Allylic
nitroxyl spin label reagents. Synthesis 1991, 616-620.
(47) Reanal Catalogue of Fine Chemicals, no. 20813-0-21.
(48) Hankovszky, H. O.; Hideg, K.; Lex, L.; Kulcsa´r, Gy.; Hala´sz, H.
A. Methods for preparation of heterobifunctional nitroxides: R,â-
unsaturated ketones, â-ketoesters, cyano-nitro-derivatives. Can.
J . Chem. 1982, 60, 1432-1438.
(49) Keana, J . F. W.; Hideg, K.; Birrell, G. B.; Hankovszky, H. O.;
Ferguson, G.; Parvez, M. New mono- and difunctionalized
2,2,5,5-tetramethylpyrrolidine- and ∆3-pyrroline-1-oxyl nitroxide
spin label reagents. Can. J . Chem. 1982, 60, 1439-1447.
(50) Keana, J . F. W.; Lee, T. D.; Bernard, E. M. Side-Chain
Substituted 2,2,5,5-Tetramethylpyrrolidine-N-oxyl (Proxyl) Ni-
troxides. A New Series of Lipid Spin Labels Showing Improved
Properties for the Study of Biological Membranes. J . Am. Chem.
Soc. 1976 98, 3052-3053.
(30) Vallet, P.; Haverbeke, Y.; Bonnet, A. P.; Subra, G.; Chapat, J .-
P.; Muller, R. N. Relaxivity Enhancement of Low Molecular
Weight Nitroxide Stable Free Radicals: Importance of Structure
and Medium. Magn. Reson. Med. 1994, 32, 11-15.
(31) Hankovszky, H. O.; Hideg, K.; Lex, L. Nitroxyls; VII. Synthesis
and Reactions of Highly Reactive 1-Oxyl-2,2,5,5-tetramethyl-2,5-
dihydropyrrole-3-ylmethyl Sulfonates. Synthesis 1980, 914-916.
(32) Belkin, S.; Mehlhorn, R. J .; Hideg, K.; Hankovszky, H. O.;
Packer, L. Reduction and Destruction Rates of Nitroxide Spin
Probes. Arch. Biochem. Biophys. 1987, 256, 232-243.
(33) Hankovszky, H. O.; Hideg, K.; Lex, L. Nitroxyls; VIII. Synthesis
of Nitroxylphosphinimines; A Convenient Route to Amine,
Isothiocyanate, Aminocarbonylaziridine, and Carbodiimide Ni-
troxyls. Synthesis 1981, 147-149.
(51) Keana, J . F. W. New aspects of nitroxide chemistry. In Spin
Labeling II, Theory and Applications; Berliner, L. J ., Ed.;
Academic Press: New York, 1979; pp 115-172.
(52) Delpierre, G. R.; Lamchen, M. Nitrones. Part I. Cycloaddition
of Unsymmetrical Olefins to the 1-Pyrroline 1-Oxides. J . Chem.
Soc. 1963, 4693-4701.
(34) Hankovszky, H. O.; Hideg, K.; Tigyi, J . Nitroxides, II. 1-Oxyl-
2,2,5,5-tetramethylpyrroline-3-carboxylic acid derivatives. Acta
Chim. Acad. Sci. Hung. 1978, 98, 339-348.
J M9802160