Unsymmetrical azo initiators increase efficiency of radical generation in aqueous dispersions, liposomal membranes, and lipoproteins

Article abstract of DOI:10.1021/ja9934605

Lipid peroxidation studies often employ the use of azo initiators to produce a slow, steady source of free radicals, but the lack of initiators capable of efficiently generating radicals in lipid aggregates such as micelles and membranes has created persistent problems in these investigations. We report here the synthesis and study of unsymmetrically substituted (hydrophilic/hydrophobic) azo initiators C-8, C-12, and C-16 that increase the efficiency of radical generation in lipophilic regions of aqueous emulsions such as micelles and liposomes. Radical generation from these initiators was monitored in micelles, liposomes, and lipoproteins by the use of two radical scavengers, one that scavengers lipophilic peroxyl radicals and one that scavenges hydrophilic peroxyls. The lipophilic radical scavenger used was the well-known antioxidant α-tocopherol and the hydrophilic radical scavenger used was uric acid. Two peroxyl radicals are trapped by each of these scavengers, tocopherol presumably being biased toward reacting with lipid soluble radicals, uric acid presumably reacting preferentially with water-soluble radicals. In Triton X-100 micelles the unsymmetrical initiators C-8 and C-16 display an increase in both α-TOH (α- tocopherol) trapping and in overall radical generation efficiency compared to the symmetrical initiators C-0 (hydrophilic) and MeOAMVN (lipophilic). The unsymmetrical azo initiators performance in liposomes was excellent (increased cage escape with lipid compartment access). In low-density lipoprotein oxidations, the initiators C-8, C-12, and C-16 also provided advantages over C-0 and MeOAMVN. The hydrophilic/hydrophobic character of the two radicals generated from the unsymmetrical initiators is an important factor for separating the geminate radical pair. These initiators, when compared to the widely used symmetrical azo initiators, provide an advantage of free radical production, lipophilic access, and constant radical generation in the investigation of lipid peroxidation in various media.

Full text of DOI:10.1021/ja9934605

4032
J. Am. Chem. Soc. 2000, 122, 4032-4038
Unsymmetrical Azo Initiators Increase Efficiency of Radical
Generation in Aqueous Dispersions, Liposomal Membranes, and
Lipoproteins
Sean M. Culbertson and Ned A. Porter*
Contribution from the Department of Chemistry, Duke UniVersity, Durham, North Carolina 27708, and
Department of Chemistry, Vanderbilt UniVersity, NashVille, Tennessee 37235
ReceiVed September 24, 1999. ReVised Manuscript ReceiVed February 28, 2000
Abstract: Lipid peroxidation studies often employ the use of azo initiators to produce a slow, steady source
of free radicals, but the lack of initiators capable of efficiently generating radicals in lipid aggregates such as
micelles and membranes has created persistent problems in these investigations. We report here the synthesis
and study of unsymmetrically substituted (hydrophilic/hydrophobic) azo initiators C-8, C-12, and C-16 that
increase the efficiency of radical generation in lipophilic regions of aqueous emulsions such as micelles and
liposomes. Radical generation from these initiators was monitored in micelles, liposomes, and lipoproteins by
the use of two radical scavengers, one that scavengers lipophilic peroxyl radicals and one that scavenges
hydrophilic peroxyls. The lipophilic radical scavenger used was the well-known antioxidant R-tocopherol and
the hydrophilic radical scavenger used was uric acid. Two peroxyl radicals are trapped by each of these
scavengers, tocopherol presumably being biased toward reacting with lipid soluble radicals, uric acid presumably
reacting preferentially with water-soluble radicals. In Triton X-100 micelles the unsymmetrical initiators C-8
and C-16 display an increase in both R-TOH (R-tocopherol) trapping and in overall radical generation efficiency
compared to the symmetrical initiators C-0 (hydrophilic) and MeOAMVN (lipophilic). The unsymmetrical
azo initiators performance in liposomes was excellent (increased cage escape with lipid compartment access).
In low-density lipoprotein oxidations, the initiators C-8, C-12, and C-16 also provided advantages over C-0
and MeOAMVN. The hydrophilic/hydrophobic character of the two radicals generated from the unsymmetrical
initiators is an important factor for separating the geminate radical pair. These initiators, when compared to
the widely used symmetrical azo initiators, provide an advantage of free radical production, lipophilic access,
and constant radical generation in the investigation of lipid peroxidation in various media.
Lipid peroxidation has been increasingly implicated in a
viscosity of the lipid environment in lipid bilayers and the
macromolecular “cage” nature of a lipoprotein significantly
prevents the cage escape of the lipophilic radicals formed in
the decomposition of AMVN.3 Studies of the peroxidation of
low-density lipoproteins (LDL), for example, include the use
of high concentrations of AMVN to produce initiation at a
reasonable rate. Such high initiator concentrations may com-
promise the integrity of the lipoprotein.
number of pathological events including atherosclerosis, liver
disease, tumorigenesis, and various neurological disorders, such
as Alzheimer’s disease.1 Initiation and antioxidant defenses are
of key importance to the mechanism(s) of lipid peroxidation,
and R-tocopherol (R-TOH) has attracted attention as a peroxi-
dation chain breaker since it is a naturally abundant and potent
lipophilic antioxidant.2
Lipid peroxidation studies often employ the use of azo
initiators to produce a slow, steady source of free radicals, but
the lack of initiators capable of efficiently generating radicals
in lipid regions has created persistent problems in these
investigations.3 The azo compounds AMVN (2,2′-azobis(2,4-
dimethylvaleronitrile)) and AAPH (2,2′-azobis(2-amidinopro-
pane) dihydrochloride), frequently used to investigate these
oxidations, present difficulties.4,5 For example, the increased
While the hydrophilic radicals derived from AAPH exhibit
efficient cage escape in water, the transfer of these radicals into
the lipid region of a molecular aggregate is dependent upon
some transport mechanism(s). In fact, it has been demonstrated
that peroxyl radicals derived from AAPH in the aqueous phase
do not initiate low-density lipoprotein (LDL) oxidation in the
absence of R-TOH.6 Indeed, there is an increasing body of
evidence suggesting that subtle changes in the initiator source,
rate of decomposition, or location of radical formation can
influence mechanistic pathways of peroxidation.4
* To whom correspondence should be addressed. E-mail: n.porter@
vanderbilt.edu.
(1) (a) Nicholson, A. C.; Hajjar, D. P. Am. Scientist 1995, 83, 460-
467. (b) Feher, J.; Lengyel, G.; Blazovics, A. Scand. J. Gastro. 1998, 288S,
38-46. (c) Welsch, C. W. Free Radical Biol. Med. 1995, 18, 757-773.
(d) Knight, J. A. Ann. Clin. Lab. Sci. 1997, 27, 11-25.
(2) Burton, G. W.; Joyce, A.; Ingold, K. U. Arch. Biochem. Biphys. 1983,
221, 281-290.
(3) Kigoshi, M.; Sato, K.; Niki, E. Bull. Chem. Soc. Jpn. 1993, 66, 2954-
2959.
(4) Bowry, V. W.; Ingold, K. U. Acc. Chem. Res. 1999, 32, 27-34.
(5) Noguchi, N.; Yamashita, H.; Gotoh, N.; Yamamoto, R.; Niki, E. Free
Radical Biol. Med. 1998, 24, 259-268.
(6) Neuzil, J.; Thomas, S. R.; Stocker, R. Free Radical Biol. Med. 1997,
22, 57-71.
10.1021/ja9934605 CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/14/2000

Efficiency of Radical Generation
J. Am. Chem. Soc., Vol. 122, No. 17, 2000 4033
Scheme 1
Scheme 2
b
Given the limitations of existing initiators, there is a need
for new compounds that efficiently generate radicals in lipid
aggregate structures such as lipoproteins, liposomes, and aque-
ous dispersions (or micelles).5 We report here the synthesis and
the study of the new unsymmetrical azo compounds, C-8, C-12,
and C-16 (Scheme 1), that decompose at convenient tempera-
tures with an increased lipophilic radical generation efficiency.
the pure unsymmetrical initiator was generated, after chroma-
tography, by BOC removal with trifluoroacetic acid.
•
These new initiators generate one amphiphilic radical, RL , and
•
one hydrophilic radical, RH , in the geminate radical pair. The
Initiator Decomposition. The unsymmetrical initiators were
designed for use in lipid emulsions or molecular aggregates and
determining the rates for decomposition of the compounds in a
variety of media was therefore of critical importance. Anionic
SDS (sodium dodecyl sulfate), neutral Triton X-100, and
cationic TTAB (tetradecyltrimethylammonium bromide) mi-
celles were chosen for study since micelles from these surfac-
tants are in common usage. Micelles generated from these
surfactants are appropriate to test micellar charge effects on the
azo initiators since SDS and TTAB micelles bear negative and
positive charges respectively while Triton micelles are un-
charged. Liposomal bilayer membranes were also chosen for
study and multilamellar vesicle (MLV) aggregates were prepared
from plant phosphatidyl choline (PC) as previously described.
results of this study suggest that the amphiphilic radical
associates with the lipid aggregate while the hydrophilic radical
escapes to the aqueous environment. The use of the “hydro-
phobic effect” to separate the geminate radical pair apparently
overcomes the difficulties of cage escape and lipid compartment
access that plagues the use of the symmetrical initiators AMVN
and AAPH.
Results and Discussion
Initiator Synthesis. Initiator structures based on the imida-
zoline functional group were chosen since the parent compound,
C-0 (2,2′-azobis(N,N′-dimethyleneisobutyramidine) dihydro-
chloride), was previously shown to have a substantially shorter
half-life for decomposition than that of the acyclic counterpart,
AAPH.7 Thus, C-0 has a rate constant for decomposition in
water at 50 °C five times that of AAPH.
The synthesis of C-8, C-12, and C-16 was based on the
displacement of ammonia from amidines by 1,2-diamines.8 The
general approach is outlined in Scheme 2. The 1,2-diamine 4
was prepared by LAH reduction of the amide and 4 was then
reacted with the unsymmetrical azo compound 5. The compound
5 was prepared by exchange of ethylenediamine on to AAPH.
The bis hydrochloride salt of 5 crystallizes from ethanol-ether
solvent mixtures.
Purification of the unsymmetrical initiators, C-8, C-12, and
C-16, proved to be difficult. Recrystallization was successful
for purification of C-16 while the other two initiators proved
to be contaminated with azo compound 5 as an impurity under
all recrystallization conditions attempted. Chromatography of
the compounds was also unsuccessful. A purification method
for use with each of the initiators was developed that involved
their conversion to the bis-BOC protected compounds 6. The
protected imidazoline chromatographs well on silica gel and
Kinetics for initiator decomposition were determined by
measurement of the disappearance of the azo absorption in the
UV at 366 nm.7 Experiments with C-0 in methanol solvent gave
good first-order kinetics over 3 half-lives. Rate constants for
C-0 decomposition, kd, were determined at 65 °C, 2.3 × 10-4
s-1, τ1/2 ) 0.8 h; 58 °C, 1.1 × 10-4 s-1, τ1/2 ) 1.8 h; and 50
°C, 3.9 × 10-5 s-1, τ1/2 ) 5.0 h. Extrapolation of these rate
constants to 37 °C gives a value of kd ) 7.1 × 10-6 s-1, τ1/2
)
27 h at this temperature. Rate constants determined for C-8,
C-12, and C-16 were similar to those measured for C-0 in
methanol. Rates of decomposition of C-0 could be determined
in phosphate saline buffer at pH 7.4 as well as in this buffer
containing SDS, TTAB, and Triton X-100 micelles. The
unsymmetrical initiators C-8, C-12, and C-16 are themselves
amphiphiles and it was impossible to monitor the loss of the
azo chromophore for these initiators in micellar media. Rate
constants could not be determined by UV for any of the initiators
in PC multilamellar liposomes since these emulsions are milky
white and not conducive to UV analysis. Rate constants for
decomposition of C-0 in pH 7.4 buffer, kd, were determined at
80 °C, 7.3 × 10-4 s-1, τ1/2 ) 0.3 h; 65 °C, 1.4 × 10-4 s-1, τ1/2
) 1.4 h; 50 °C, 2.4 × 10-5 s-1, τ1/2 ) 8 h; and 37 °C, 4.7 ×
10-6 s-1, τ1/2 ) 41 h. Rate constants for decomposition of C-0
in pH 7.4 buffer micelles at 37 °C were as follows: 0.5 M SDS,
(7) Fujie, H.; Shiraki, K.; Miyagawa, T.; Mnamii, N. Pure Appl. Chem.
1992, A29, 741-751.
(8) Dougherty, T. J. J. Am. Chem. Soc. 1961, 83, 4849-4853.

4034 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Culbertson and Porter
Scheme 3
radicals that have escaped the cage and that are trapped by uric
acid while the lipophilic term, eL, reports the efficiency of escape
of radicals that leads to scavenging by the lipophilic scavenger
R-tocopherol.
3.3 × 10-6 s-1, τ1/2 ) 58 h; 0.1 M Triton X-100, 4.5 × 10-6
s-1, τ1/2 ) 43 h; and 0.1 M TTAB, 4.5 × 10-6 s-1, τ1/2 ) 43
h. Surfactant concentrations were chosen such that the experi-
ment was carried out at least 2 orders of magnitude above the
critical micelle concentration and such that the initiator con-
centration was approximately 1% of the surfactant concentration.
Free Radical Generation and Scavenging. Radical genera-
tion was monitored by the use of two antioxidants, one
lipophilic, R-tocopherol (R-TOH) and one hydrophilic, uric acid
(UA).9 Both R-TOH and UA are efficient radical chain breakers
that scavenge two free radicals per antioxidant molecule.9,10 The
widely used aqueous antioxidant ascorbic acid is known to
reduce R-tocopheroxyl radicals (R-TO•) and regenerate R-TOH
across the interface of micelles, liposomes, and LDL.11,12 Uric
acid does not exhibit this regenerating capability with R-TOH
in any system so far investigated.9 The experiments carried out
monitored the rate of disappearance of R-tocopherol and uric
acid during the decomposition of a given initiator. Since we
monitor radical consumption by the use of two radical scaven-
gers, R-TOH and UA, the sum of rates of consumption for the
two scavengers should give information about the efficiency
of radical generation from a given initiator under given
conditions of decomposition.
If the efficiency of radical generation from an initiator were
100%, then the sum of the rates of scavenger consumption
would be equal to the rate of initiator decomposition since the
initiator generates two radicals and both antioxidants (R-
tocopherol and uric acid) consume two radicals. Furthermore,
the partitioning of the radical formed from the initiator between
R-TOH and UA presumably gives information about the locus
of the scavenged radical in the emulsion. Lipophilic radicals
would presumably be scavenged by R-TOH while hydrophilic
radicals would be trapped by UA. The chemistry described in
these studies is outlined in Scheme 3. Values for observed
initiator free radical flux, as measured by consumption of
scavengers, are given by 2eT‚kd[initiator], where eT is the total
efficiency of radical escape from the initial radical pair formed
in the decomposition (100‚escape fraction), and kd[initiator] is
the rate of geminate radical pair formation. Two radicals are
generated per initiator decomposition accounting for that factor
in eq 1; an efficiency (eT) of 100% means that all radicals
formed from an initiator escape from the solvent cage and react
with a radical scavenger. Equations 2 and 3 separate the total
efficiency of radical generation into lipophilic and hydrophilic
components (eL and eH). The hydrophobic term, eH, represents
To determine the values of eT, eL and eH, the rate of
consumption of both R-tocopherol and uric acid was determined.
Dividing these rates by the kd[initiator] appropriate for the
medium of study gives eL and eH; the sum of these two
efficiencies is eT.9,13 Rates of consumption of R-tocopherol and
uric acid were determined by monitoring the disappearance of
these compounds by HPLC. In a typical experiment in micelles
or liposomes, initiator (∼2 mM), surfactant (0.1-0.5 M),
R-tocopherol (∼0.1 mM), uric acid (∼0.1 mM), and methyl
linoleate (∼5 mM)14 were vortexed in pH 7.4 phosphate buffer
and kept at 37 °C for several hours (<0.1 half-lives) during
which time aliquots were taken and analyzed for R-tocopherol
and uric acid. Rates of tocopherol and urate consumption were
determined and eT, eL, and eH were calculated by the use of eq
3.
Scavenging Efficiencies. Table 1 and Figures 1-3 sum-
marize the results obtained from the azo initiators tested in
TTAB, SDS, and Triton-X100 micelles. The following general
trends are noted: (a) In TTAB micelles, a general trend for
nearly an equal split between lipophilic and hydrophilic
scavenging, eL and eH, was observed for all of the intiators and
•
eT was high. (b) In SDS micelles R-TOH was the primary
antioxidant consumed for C-0 and the unsymmetrical initiators,
C-8 and C-16 (eL > eH). (c) In Triton X-100, radical generation
from C-0 was significantly less than that observed for the
unsymmetrical initiators. The eL term for the unsymmetrical
initiators was very high. (d) The symmetrical initiator MeOAM-
VN (2,2′-azobis(4-methoxy-2,4-dimethylvaleronitrile)) has a low
efficiency of radical generation in Triton X-100 micelles (eT )
9%), consistent with earlier reports that symmetrical initiators
such as AMVN and MeOAMVN have low efficiencies of
radical generation in surfactant emulsions.5,11 Some of the eT’s
determined were greater than 100%, a result that is obviously
in error. These errors are likely due to the assumptions made in
the calculations, particularly with regard to the extrapolation
of kd values from C-0 to the unsymmetric compounds.15
The anionic, neutral, and cationic micelles each interact
differently with the cationic azo initiators tested. Of the micelles
investigated, TTAB shows the highest consumption of uric acid,
which may be due to the surfactant’s cationic character.16 The
charge on the micelle has an effect of gathering scavenger to
(13) Barclay, L. R. C.; Locke, S. J.; MacNeil, J. M.; VanKessel, J.;
Burton, G. W.; Ingold, K. U. J. Am. Chem. Soc. 1984, 106, 2479-2481.
(14) In TTAB micelles the presence of methyl linoleate with R-TOH
and no initiator caused significant antioxidant consumption. Thus, methyl
linoleate was excluded from these experiments.
(9) (a) Niki, E.; Saito, M.; Yoshikawa, Y.; Yamamoto, Y.; Kamiya, Y.
Bull. Chem. Soc. Jpn. 1986, 59, 471-477. (b) Niki, E.; Saito, T.; Kawakami,
A.; Kamiya, Y. J. Biol. Chem. 1984, 259, 4177-4182.
(10) Simic, M. G.; Javanovic, S. V. J. Am. Chem. Soc. 1989, 111, 5778-
5782.
(11) Bowry, V. W.; Stocker, R. J. Am. Chem. Soc. 1993, 115, 6029-
6044.
(12) Candenas, E. In Free Radicals, OxidatiVe Stress, and Antioxidants:
Pathological and Physiological Significance; Ozben, T., Ed.; Plenum
Press: New York, 1998; pp 237-251.
(15) Preliminary experiments have been carried out measuring the rates
of initiator decomposition in molecular aggregates such as micelles and
liposomes by nitrogen evolution. These preliminary studies suggest that
the assumption of a common rate of decomposition for C-0, C-8, C-12,
and C-16 is reasonable, To be submitted for publication: S. M. Culbertson,
N. A. Porter, L. R. C. Barclay, and M. Vinqvist.
(16) If the micelle captures urate because of charge effects, urate will
compete effectively for lipophilic peroxyl radicals. C-0 decomposes in the
aqueous phase where UA can scavenge a large fraction of the free radicals
before partitioning into the lipid phase.

Efficiency of Radical Generation
J. Am. Chem. Soc., Vol. 122, No. 17, 2000 4035
Table 1. Radical Generation Efficiency of Azo Initiators in Micelles
initiator
C-0
surfactant
d[RToc]/dt [initiator]
d[urate]/dt [initiator]
kd
eL
eH
eT
SDS
3.4 × 10-6
1.5 × 10-6
1.5 × 10-6
1.5 × 10-6
4.0 × 10-6
2.1 × 10-6
1.7 × 10-6
4.1 × 10-6
8.2 × 10-7
2.2 × 10-6
2.5 × 10-7
1.2 × 10-6
1.8 × 10-6
4.5 × 10-7
1.3 × 10-6
1.8 × 10-6
2.0 × 10-8
9.0 × 10-7
1.4 × 10-6
6.4 × 10-7
3.3 × 10-6
4.5 × 10-6
4.5 × 10-6
3.3 × 10-6
4.5 × 10-6
4.5 × 10-6
3.3 × 10-6
4.5 × 10-6
4.5 × 10-6
3.2 × 10-5
102
34
33
44
88
47
52
90
18
7
7
27
40
14
29
40
1
20
32
2
109
61
73
58
117
87
53
110
50
X-100
TTABS
SDS
X-100
TTABS
SDS
X-100
TTABS
X-100
C-8
C-16
MeOAMVN
9
Figure 1. Efficiencies for scavenging by R-tocopherol (eL), uric acid
(eH), and total (eT) in TTAB. The e’s are as defined in the text.
Efficiencies based on R-TOH and uric acid (UA) consumption rates at
37 °C in air. The kd values for C-0 were measured in 10 mM PBS (pH
7.4) for liposomes and LDL, and in each respective micelle. The kd
values for the unsymmetrical initiators were estimated by using C-0
values. Micelles composed of TTAB (0.1 M) containing initiator 2 mM,
R-TOH e 140 µM, and UA e 150 µM.
Figure 3. Efficiencies for scavenging by R-tocopherol (eL), uric acid
(eH), and total (eT) in Triton-X100. See caption for Figure 1. Micelles
composed of Triton X-100 (0.1 M) and methyl linoleate (5.0 mM).
from C-0. What is nevertheless surprising is the fact that nearly
all of the radicals generated by the water-soluble intitiator are
trapped by the hydrophobic inhibitor, R-TOH. Indeed, R-TOH
is the dominant radical scavenger for C-8 and C-16 as well.
We suggest that the anionic surface of SDS micelles acts to
trap cationic azo initiator or their derived peroxyl radicals by
salt exchange, significantly favoring scavenging by R-TOH near
the micelle surface. In short, the cationic initiators or the radicals
formed from each initiator associate with negatively charged
SDS micelles as does lipophilic R-TOH. The radicals and the
scavenger are associated with the same aggregate leading to a
bias for consumption of R-TOH. Association of cations with
negatively charged micelles is the basis of surfactant catalysis;
the phenomenon is well-known.17 We speculate that the dif-
ferences in overall efficiency between C-0 and the unsym-
metrical intitiators (C-8 and C-16) in SDS is due to exchange
of the radicals out of the initial geminate radical pair micelle
aggregate for C-0. This exchange process would be expected
to be slower for the more lipophilic radical derived from C-8
or C-16. An alternate explanation for the high efficiency
observed from C-0 is the possibility that its decompostion occurs
outside of the aggregate while C-8 and C-16 decompose while
associated with the micelle. Radicals derived from C-0 will
likely associate with different micelles while radicals derived
from C-8 and C-16 are confined, at least initially, to the same
micelle. This leads to a low eT for the unsymmetrical initiators.
In Triton X-100 micelles, no ion association with the micelle
is expected. The hydrophilic/hydrophobic character of the two
radicals generated from the unsymmetrical initiators should be
the dominant factor leading to differences between C-0 and the
Figure 2. Efficiencies for scavenging by R-tocopherol (eL), uric acid
(eH), and total (eT) in SDS. See caption for Figure 1. Micelles composed
of SDS (0.5 M) and methyl linoleate (5.0 mM).
the micellar surface. Urate is an anion at pH 7.4,10 and it
presumably associates with the cationic micelle surface where
it can scavenge a higher percentage of radicals generated near
the aqueous interface. The total efficiency of scavenging of
radicals in SDS micelles (eT) is ∼100% for the water-soluble
symmetric initiator C-0 and approximately 50% for the unsym-
metric initiators. The acyclic analogue of C-0, AAPH, has a
high efficiency of radical generation, consistent with the results
(17) See, for example: Fendler, J. H.; Fendler, E. J. Catalysis in Micellar
and Macromolecular Systems; Academic Press: New York, 1975.

4036 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Culbertson and Porter
Figure 5. Efficiencies for scavenging by R-tocopherol (eL), uric acid
(eH), and total (eT) in low-density lipoproteins (LDL). See caption for
Figure 1. Oxidation of LDL (0.75 mg protein/mL) containing endog-
enous R-TOH (12 µM), added UA (60 µM), and 0.5 mM initiator,
except 0.25 mM for MeOAMVN.
Figure 4. Efficiencies for scavenging by R-tocopherol (eL), uric acid
(eH), and total (eT) in phosphatidyl choline liposomes. See caption for
Figure 1. Liposomes prepared from plant PC (10.3 mM) containing
initiator (2 mM), R-TOH e140 µM, and UA (150 µM).
the rate constant for decomposition of all of the initiators in
LDL and liposomes is assumed to be comparable to rate
constants determined in pH 7.4 buffer and this assumption may
lead to errors in the calculated e’s.15 By the same token, the
rate constant for decomposition of MeOAMVN in LDL was
assumed comparable to rates of decompostion in acetonitrile
to calculate e’s.5 The calculated efficiencies may reflect the fact
that this assumption is not valid. Experiments are in progress
to determine initiator kd values in lipoproteins and other
molecular aggregates which would avoid the necessity of any
rate constant assumptions.15
The experiments presented here demonstrate that unsym-
metrical azo initiators increase free radical generation efficiency
in the lipid phase of neutral micelles, PC liposomes, and LDL.
These initiators, when compared to the widely used symmetrical
azo initiators, provide an advantage of free radical production,
lipophilic access, and consistent radical generation in the
investigation of lipid peroxidation in various media. They may
provide valuable tools in the investigation of lipid peroxidation,
antioxidant protection, or even emulsion polymerization. Fur-
thermore, the results show that charge on a radical and molecular
aggregate host can have a significant effect on the radical
chemistry. Negatively charged micelles such as SDS gather
cationic radicals derived from the initiators and this has impact
on the ultimate fate of the radicals.
amphiphilic initiators, C-8 and C-16. The results suggest that
the hydrophobic effect is an important factor leading to
separation of the geminate radical pair for these initiators. We
also note the observed high efficiencies of radical generation
for C-8 and C-16 (eT ∼ 100%) as compared to MeOAMVN
(9%, see Table 1) in Triton X-100 micelles. MeOAMVN is a
symmetric lipophilic initiator and the geminate radical pair
separates with very low efficiency in the micellar media.
The greatest effect on lipophilic radical generation observed
was in phosphatidyl choline liposomes. The data for studies in
liposomes are presented in Figure 4. Overall efficiency for C-0
was good, but the fraction of radicals scavenged in the lipophilic
phase was low. MeOAMVN efficiency in phosphatidyl choline
liposomes was very low overall with a negligible consumption
of R-TOH. In contrast, the unsymmetrical azo initiators
performance in liposomes was excellent. The high overall
efficiency observed suggests that most radicals derived from
these initiators escaped the initial cage. The high R-TOH
consumption also supports the conclusion that most radicals had
access to the lipid phase of these multilamellar vesicles.18
Unsymmetrical azo initiators also provided increased lipo-
philic radical generation in LDL, Figure 5. The low lipophilic
scavenging (eL) of C-0 suggests that the aqueous peroxyl
radicals it generates do not have efficient access to the lipid
core of LDL.6,11 Each of the unsymmetrical azo initiators tested
displayed significantly greater lipophilic scavenging eL than did
C-0. The addition of the alkyl chain to one radical in the pair
apparently allows greater cage escape and access to LDL lipid
regions. It was surprising that the symmetrical MeOAMVN
initiator showed a higher overall efficiency in LDL than was
observed in either micelles or liposomes.19 We point out that
Experimental Section
r-Aminodecanoic Acid (1a).20 Diethyl acetamidomalonate (24.3 g,
0.11 mol) and 1-bromooctane (29 g, 0.15 mol) were heated under reflux
in a solution of sodium (2.5 g, 0.11 mol) in EtOH (85 mL) for 24 h.
Upon cooling, the mixture was poured onto ice/water (160 mL) and
the precipitate filtered and washed with water. The crude solid was
placed into a 500 mL round-bottom flask, concentrated HCl (180 mL)
and DMF (20 mL) were added, and then the mixure was heated at
reflux for 24 h. The mixture was allowed to cool, poured into a solution
of EtOH/water (3:1), and neutralized with concentrated aqueous NH3.
The precipitate was filtered and washed with EtOH/water; crude yield
of 1a 14.6 g (71%). Crude product 1b was carried directly onto the
next step. Crude yield of 1c 29.5 g (80%).
(18) The initiators C-8, C-12, and C-16 primarily decompose in the lipid
aggregate where both free radicals generated can be quickly scavenged by
R-TOH in the immediate vicinity.
(19) It has been reported that MeOAMVN peroxyl radical formation in
LDL as measured by the single, lipophilic antioxidant DPPD gave an e ×
kd value of 4.14 × 10-6 (or % e ) 13%).5 Our experience with the
antioxidant DPPD implicated it did not function the same as R-TOH in
micelles and liposomes tested. In fact, results suggest that there is an
interconnection (possibly a regeneration of uryl radical) between UA and
DPPD across the aqueous/lipid interface (data not shown).
Methyl r-Aminodecanoate Hydrochloride (2a).20 The crude amino
acid 1a (25.9 g, 0.14 mol) was heated at reflux in a mixture of thionyl
chloride (11.2 mL, 0.15 mol) and MeOH (60 mL) for 24 h. Ninety

Efficiency of Radical Generation
J. Am. Chem. Soc., Vol. 122, No. 17, 2000 4037
percent of the solvent was removed and the precipitated product filtered
and recrystallized from MeOH. 2a: Yield 29.0 g (88%). 1H NMR (300
MHz, CD3OD) δ 4.05 (t, 1 H); 3.85 (s, 3 H); 1.90 (m, 2 H); 1.35 (m,
first from CHCl3 and then from MeOH. The free base was then
dissolved in MeOH saturated with HCl gas, stirred for 1 h, evaporated,
and then recrystallized from MeOH/Et2O to give a white powder. Mp
198 °C dec. 1H NMR (300 MHz, CD3OD) δ 4.05 (s, 8 H); 1.55 (s, 12
H). MS (FAB) calcd for [C12H23N6]+ 251.20, found 251.22.
1
12 H); 0.90 (t, 3 H). 2b: Crude yield 25.4 g (overall 79%). H NMR
(300 MHz, CD3OD) δ 4.05 (t, 1 H); 3.85 (s, 3 H); 1.90 (m, 2 H); 1.35
1
(br m, 20 H); 0.90 (t, 3 H). 2c: Yield 37.2 g (82%). Mp 103 °C. H
2,2′-Azo(2-amidinopropane)[2-(2-imidazolin-2-yl)propane] Di-
hydrochloride (5). The unsymmetrical initiator 5 was synthesized
similiarly to C-0 with a few modifications. In a 500 mL three-necked
flask were dissolved AAPH (12.0 g, 44.3 mmol) and sodium methoxide
(1.44 g, 26.7 mmol) in 450 mL of anhydrous MeOH at room
temperature. After 30 min, ethylenediamine (1.48 mL, 22.2 mmol) was
added and argon bubbled vigorously through the solution. NH3 in the
exhaust was trapped in water containing methyl red indicator. Titration
of the NH3 showed the reaction to be complete after 52 h. The solvent
was evaporated and the resulting solid was washed with 125 mL of
EtOH at 40 °C. The filtrate was treated with HCl gas and then
evaporated after 30 min. The solid was washed with30 mL of EtOH at
room temperature and then rinsed with 30 mL of Et2O. The remaining
residue was dissolved in 125 mL of EtOH at 40 °C, filtered, and
recrystallized with the addition of Et2O. 5: Yield 2.43 g (37%). Mp
143 °C dec. 1H NMR (300 MHz, d6DMSO) δ 10.49 (s, 2 H); 9.40 (s,
2 H); 9.10 (s, 2H); 3.90 (s, 4 H); 1.50 (d, 12 H).
NMR (300 MHz, CD3OD) δ 4.05 (t, 1 H); 3.85 (s, 3 H); 1.90 (m, 2
H); 1.35 (br m, 28 H); 0.90 (t, 3 H). MS (FAB) calcd for [C19H40-
NO2]+ 314.3, found 314.3.
r-Aminodecanamide (3a).21 The amino methylester hydrochloride
2a (29.0 g, 0.12 mol) was dissolved in 250 mL of methanol and
saturated with NH3 at 0 °C. The solution was resaturated with NH3 at
0 °C after 5 h and once more after 2 d. After 4 d the solvent was
removed. The resulting amide, amide hydrochloride, and NH4Cl were
treated with 10% NaOH and extracted with CHCl3. Recrystallization
from EtOH yielded 15.1 g (66%) of solid R-amino amide 3a. 1H NMR
(300 MHz, CD3OD) δ 3.85 (t, 1 H); 1.85 (m, 2 H); 1.30 (br m, 12 H);
0.90 (t, 3 H). 3b: Yield 16.5 g (79%). Mp 95 °C. 1H NMR (300 MHz,
CD3OD) δ 3.84 (t, 1 H); 1.85 (m, 2 H); 1.30 (br m, 20 H); 0.90 (t, 3
H). 3c: Yield 10.4 g (45%). Mp 98 °C.22 1H NMR (300 MHz, CDCl3)
δ 7.10 (s, 2 H); 5.50 (s, 2 H); 3.35 (m, 1 H); 1.85 (m, 2 H); 1.25 (m,
28 H); 0.90 (m, 3 H).
1,2-Diaminodecane (4a).22,23 The R-amino amide 3a (12.0 g, 64.5
mmol) was added in small portions to a suspension of LiAlH4 (5.39 g,
142 mmol) in 140 mL of anhydrous THF. The reaction mixture was
stirred and heated under reflux for 24 h. It was then cooled to 5 °C
and cautiously quenched by adding, successively and dropwise, water
(5.4 mL), 10% NaOH (8.1 mL), and finally water (13.5 mL). The
precipitated, granular alumina was filtered, washed with hot THF, and
then extracted three times with boiling THF. The combined filtrates
were taken to dryness in vacuo at 50 °C. Crude yield: 4a 80%, 4b
92%, 4c 92%. A portion of the crude 1,2-diaminodecane (5.8 g, 34
mmol) was dissolved in MeOH (125 mL) and treated with (BOC)2O
(18.2 g, 83 mmol) and triethylamine (11.65 mL, 83 mmol). After the
mixture was stirred for 2 h, the solvent was evaporated to give a crude
oil. Gradient FC (5%, 10%, 20%, 40%, 80%, 100% EtOAc in hexanes
with 0.2% TEA) gave the BOC-protected diaminodecane as a white
solid, 9.2 g (73%). The solid was dissolved in MeOH and treated with
excess HCl gas. Recrystallization from MeOH/Et2O provided a purified
dihydrochloride that was converted to the free base by treatment with
aqueous NaOH and extraction with EtOAc. 4a: Yield 4.42 g (60%)
of a colorless oil. 1H NMR (300 MHz, CDCl3) δ 2.74 (dd, 1 H); 2.65
(m, 1 H); 2.44 (dd, 1 H); 1.28 (br m, 14 H); 0.89 (t, 3 H). 4b: Yield
colorless, semisolid (41%). 1H NMR (300 MHz, CDCl3) δ 2.75 (dd, 1
H); 2.65 (m, 1 H); 2.45 (dd, 1 H); 1.30 (br m, 22 H); 0.90 (t, 3 H). 4c:
Yield white powder (56%). Mp 59 °C. 1H NMR (300 MHz, CDCl3) δ
2.74 (dd, 1 H); 2.65 (m, 1 H); 2.45 (dd, 1 H); 1.28 (br m, 30 H); 0.89
(t, 3 H).
2,2′-Azobis[2-(2-imidazolin-2-yl)propane] Dihydrochloride (C-
0).24 In a 500 mL three-necked round-bottom flask was dissolved a
sample of AAPH (10 g, 0.037 mol) in 200 mL of anhydrous MeOH
which was treated with ethylenediamine (12.5 mL, 0.185 mol) at room
temperature. One neck of the flask was connected to an argon line that
had a hollow glass tube long enough to bubble Ar directly into the
solvent. The second neck served as the exhaust over the surface of the
solvent. The exhaust line was submersed into an erlynmeyer flask of
water containing a small amount of the indicator methyl red. The
remaining neck of the flask was stoppered after the addition of the
reactants. The reaction was followed to 87% completion after 36 h by
titration of the trapped NH3 in water.26 Ninety percent of the solvent
was evaporated and the precipitated product filtered and recrystallized
2,2′-Azo[2-(2-imidiazolin-2-yl)propane]{2-[2-(4-n-octyl)imidazo-
lin-2-yl]propane} Dihydrochloride(C-8). In a three-necked flask was
dissolved azo initiator 5 (6.5 g, 22 mmol) and sodium methoxide (0.65
g, 12 mmol) in 200 mL of anhydrous MeOH. After the mixture was
stirred for 30 min, 1,2-diaminiodecane (1.88 g, 11 mmol) was added
as argon bubbled through the solution and exhausted into a water trap.
The reaction was followed for 48 h (84% completion) by titration of
trapped NH3. The argon bubbling was stopped and sodium methoxide
(1.31 g, 24.2 mmol) was added. After 1 h the solvent was evaporated
and the solid washed with CHCl3. The filtrate was saved and evaporated
to dryness in vacuo. The crude solid and anhydrous sodium carbonate
(25 g, 0.30 mol) were suspended in 80 mL of EtOH.25 With vigorous
stirring, (BOC)2O (33.3 g, 0.15 mol) was added and reacted for 3 d.
After the first 3 d additional (BOC)2O (9.5 g, 0.04 mol) was added
and reacted another 24 h. Evaporation of the solvent and extraction of
the solid with EtOAc yielded a crude BOC protected initiator. Gradient
FC (5%, 10%, 20%, 30% EtOAc in hexanes) provided purified BOC
1
protected initiator 6, yield 2.3 g (37%). H NMR (300 MHz, CDCl3)
δ 3.94 (m, 1 H); 3.82 (m, 1 H); 3.79 (s, 4 H); 3.40 (dd, 1 H); 1.58 (m,
12 H); 1.42 (s, 18 H); 1.30 (br m, 12 H); 0.89 (t, 3 H). HRMS (FAB)
calcd for [C30H55O4N6]+ 563.4285, found 563.4294. Deprotection with
trifluroacetic acid gave a TFA salt that was dissolved in water, treated
with 50% NaOH until strongly basic, and extracted with CHCl3. The
resulting free base was then treated with HCl in MeOH to provide the
desired dihydrochloride. Recrystallization from MeOH/Et2O yielded
pure C-8 (0.63 g, 13% overall) as a white powder. Mp 156 °C dec. 1H
NMR (400 MHz, CD3OD) δ 4.41 (m, 1 H); 4.13 (t, 1 H); 4.05 (s, 4
H); 3.71 (dd, 1 H); 1.75 (br m, 2 H); 1.55 (s, 12 H); 1.35 (br m, 12 H);
0.90 (t, 3 H). HRMS (FAB) calcd for [C20H39N6]+ 363.3236, found
363.3252. Anal. calcd for C20H40N6Cl2: C, 55.14; H, 9.26; N, 19.31.
Found: C, 55.34; H, 9.20; N, 19.24.
2,2′-Azo[2-(2-imidiazolin-2-yl)propane]{2-[2-(4-n-dodecyl)imida-
zolin-2-yl]propane} Dihydrochloride (C-12). The unsymmetrical azo
initiator C-12 was synthesized and purified the same way as C-8.
C-12: Yield 0.62 g (19% overall) of a purified white powder. Mp 160
1
°C dec. H NMR (400 MHz, CD3OD) δ 4.40 (m, 1 H); 4.13 (t, 1 H);
4.05 (s, 4 H); 3.71 (dd, 1 H); 1.75 (br m, 2 H); 1.55 (s, 12 H); 1.35 (br
m, 20 H); 0.90 (t, 3 H). HRMS (FAB) calcd for [C24H47N6]+ 419.3862,
found 419.3857. Anal. calcd for C24H48N6Cl2: C, 58.62; H, 9.85; N,
17.10. Found: C, 58.75; H, 9.77; N, 17.04.
2,2′-Azo[2-(2-imidiazolin-2-yl)propane]{2-[2-(4-n-hexadecyl)imi-
dazolin-2-yl]propane} Dihydrochloride (C-16). The unsymmetrical
azo initiator C-16 was synthesized the same way as C-8, but was
purified by fractional recrystallization rather than protection and
chromatography. Breifly, the crude reaction product was first treated
with HCl gas in MeOH, evaporated, then washed with a small amount
of EtOH at room temperature. Four successive recrystallizations from
MeOH/Et2O yielded C-16 (1.7 g, 44%) as a pure white powder. Mp
(20) Gibbons, W. A.; Huges, R. A.; Charalambous, M.; Christodoulou,
M.; Szeto, A.; Aulabaugh, A. E.; Mascagni, P.; Toth, I. Liebigs Ann. Chem.
1990, 1175-1183.
(21) Brunner, H.; Schmidt, M.; Unger, G. Eur. J. Med. Chem. Chim.
Ther. 1985, 20, 509-512.
(22) Kempter, V. G.; Moser, G. J. Prakt. Chem. 1966, 34, 104-111.
(23) Amundsen, A. R.; Whelan, J.; Bosnich, B. Inorg. Chem. 1979, 18,
206-208.
(24) (a) Hammond, G. S.; Neuman, R. C., Jr. J. Am. Chem. Soc. 1963,
85, 1501-1508. (b) Neuman, R. C., Jr.; Hammond, G. S.; Dougherty, T. J.
J. Am. Chem. Soc. 1962, 84, 1506-1507.
(26) Reynaud, P.; Brion, J.; Menard, G. Bull. Soc. Chim. Fr. 1978, 9-10,
II-449-II-456.
(25) Sen, A. B.; Shanker, K. J. Prakt. Chem. 1965, 29, 304-311.

4038 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Culbertson and Porter
150 °C dec. 1H NMR (400 MHz, CD3OD) δ 4.40 (m, 1 H); 4.13 (t, 1
H); 4.05 (s, 4 H); 3.71 (dd, 1 H); 1.75 (br m, 2 H); 1.56 (s, 12 H); 1.35
(br m, 28 H); 0.90 (t, 3 H). 13C NMR (75.4 MHz, CD3OD) δ 174.4,
173.3, 71.6, 59.3, 51.0, 46.0, 45.9, 36.0, 35.9, 35.9, 35.9, 33.0, 30.7,
30.7, 30.6, 30.6, 30.5, 30.5, 30.4, 28.3, 25.6, 25.5, 23.7, 23.6, 23.6,
19.5, 19.4, 14.4 ppm. HRMS (FAB) calcd for [C28H55N6]+ 475.4490,
found 475.4510. Anal. calcd for C228H56N6Cl2: C, 61.38; H, 10.31; N,
15.35. Found: C, 61.66; H, 10.36; N, 15.06.
δ-TOH (12.0 µM) in the mobile phase) intermittently between sample
injections of 15-75 µL. Columns were flushed with MeOH/H2O (50:
50) after each use.
Urate Analysis.29 Uric acid standards were prepared from a stock
solution (1.25 mM UA in PBS pH 7.4). The stock solution was diluted
with a mobile phase to produce a working standard of 2.25 µM UA.
The mobile phase consisted of filtered and degassed sodium acetate
buffer (40 mM, pH 4.75), Na2EDTA (0.54 mM), dodecyltriethylam-
monium phosphate (1.5 mM, Regis Chemical Co., Morton Grove, IL),
and 7.5% MeOH. Electrochemical detector settings were +0.60 to
+0.65 V with a range setting of 50 nA and the flow rate was 1.5 mL/
min. The standard curves were generated by intermittently injecting
5-25 µL of working standard between sample injections. Samples of
UA were diluted with 1.0-2.5 mL of mobile phase and 15 µL was
injected. Columns were flushed with MeOH/H2O (50:50) after each
use.
Free Radical Generation Efficiency in Liposomes. The generation
of free radicals in MLV liposomal membranes was carried out at 37
°C in air using plant PC (10.3 mM). PC and lipid soluble components
(R-TOH and initiator, except for C-0) were dissolved in MeOH and
placed in a 5 mL flask. The MeOH was evaporated in vacuo to obtain
a thin film. PBS buffer (4.4 mL) and UA (0.6 mL, 1.25 mM in PBS)
were added and the PC film was slowly peeled off by shaking and
then vortexing to obtain white, milky liposome suspensions.30 Aliquots
for R-TOH and UA were removed at various intervals, and treated as
described above.
Free Radical Generation Efficiency in Lipoproteins. (a) Lipo-
protein Isolation.31 Whole blood from fasting, normolipidemic healthy
subjects was collected in a 450 mL ACD blood collection bag (Baxter)
containing the following: 1.61 g of dextrose; 1.66 g of sodium citrate
dihydrate; 188 mg of anhydrous citric acid; and 140 mg of monobasic
sodium phosphate monohydrate. The bag containing blood was
centrifuged at 4200 rpm for 10 min at 22 °C and the plasma was
collected. The low-density lipoprotein (LDL) was isolated from plasma
over 15 h by density gradient sequential ultracentrifugation at 14 °C
using a Beckman Optima LE-80K centrifuge and a Ti 70 rotor. Each
spin was preformed at 504 000 g for 5.5 h. Lipoproteins were dialyzed
extensively against 10 mM PBS, sterilized by passage through a Millex-
HA 0.45 µM filter, and stored at 4 °C under argon. Protein concentra-
tions of the LDL preparations were determined by the method of
Lowry.32 LDL isolation was confirmed with the use of SDS PAGE
separation of associated apoproteins and Beckman LIPO Gel electro-
proesis of intact lipoproteins.33
(b) Oxidation of LDL. Lipoprotein concentrations were adjusted
to 0.75 mg of protein/mL with PBS containing 60 µM UA and allowed
to equilibrate to 37 °C for 5 min in a 10 mL flask. To the stirred solution
was added initiator (0.065 M) in MeOH (except C-0 was in PBS) to
give a final C-0, C-8, C-12, or C-16 concentration of 0.5 mM, or a
MeOAMVN concentration of 0.25 mM. Following addition of initiator
(time zero) aliquots for R-TOH and UA were removed at various
intervals and treated as described above.
Determination of Rate Constants for Azo Decomposition. The
kd values were measured by following the loss of azo chromophore at
366 nm.7,24 The kd values for the initiators C-8, C-12, and C-16 were
measured in MeOH containing 0.2% concentrated HCl. Values for kd
in methanol were the following: C-0 (7.1 × 10-6), C-8 (8.3 × 10-6),
C-12 (8.0 × 10-6), and C-16 (7.4 × 10-6). MeOH experiments were
conducted at three temperatures, 50, 58, and 65 °C, over the course of
several half-lives, then extrapolated to calculated kd at 37 °C. kd values
for C-0 were also measured in PBS (phosphate buffered saline) buffer
pH 7.4 (4.7 × 10-6), SDS (3.3 × 10-6), Triton X-100 (4.5 × 10-6),
and TTAB micelles (4.5 × 10-6). SDS micelles were 0.5 M in PBS
buffer pH 7.4; Triton X-100 and TTAB micelles were both 0.1 M
diluted in PBS buffer pH 7.4. Aqueous experiments for C-0 were
conducted at 37 °C and followed the decomposition for the first 2 h.
Inititial rates were used because interfering absorbances began to grow
in and result in nonlinear rate analysis for extended decomposition
studies. The UV initial rate method was not successful for kd analysis
of unsymmetrical azo initiators, C-8, C-12, and C-16, in aqueous
dispersions. The kd values for compounds C-8, C-12, and C-16 were
only measured in methanol. However, considering the unsymmetrical
azo initiators are structurally similar to C-0 and also show similiar
decomposition rates in MeOH, the kd values obtained in buffer and
micelles were assumed to be comparable.
Free Radical Generation Efficiency In Micelles. The generation
of free radicals in aqueous dispersions was carried out at 37 °C in air
using either SDS (0.5 M), Triton X-100 (0.1 M), or TTAB (0.1 M). In
a typical experiment R-TOH (150 µL, ∼4.5 mM in MeOH) and methyl
linoleate (8.3 µL (neat); no methyl linoleate was added for TTAB
micelles)14 were mixed in a 5 mL flask. Next, C-8 (150 µL, 0.065 M
in MeOH) was added, swirled, and followed by addition of surfactant
(4 mL;either 0.625 or 0.125 M in PBS pH 7.4 depending on desired
final concentration). (Note that in the case of water-soluble C-0, the
initiator was added in PBS pH 7.4 after the addition of surfactant, and
pure MeOH (150 µL) was added in place of the initiator solution in
methanol.) Finally, uric acid (600 µL, 1.25 mM in PBS pH 7.4) and
PBS (100 µL) were added bringing the total volume to 5 mL. The
dispersions were thoroughly mixed by vortexing 2 min.
Oxidation and Extraction of Antioxidants. The flask containing
all components of the experiment was incubated at 37 °C for 5 min
before the initial timepoint was removed. Aliquots were taken for
R-TOH analysis (200 µL) and UA analysis (20 µL, or 50 µL) and placed
directly on ice. Typically, 4 to 6 timepoints were taken during the first
3 h of oxidation. To the 200 µL R-TOH aliquot was added BHT (30
µL, 3 mM in MeOH) and a known amount of the internal standard,
δ-TOH (100 µM in MeOH). Extraction of all the aliquots was
performed with ice-cold MeOH (1 mL) and ice-cold hexanes (5 mL)
in sequence, vortexed vigorously after the addition of each solvent (ca.
15 s) and then centrifuged at 1700 rpm for approximately 1 min with
the use of an Adams analytical centrifuge.27 The hexane phase was
removed by pipet and concentrated under argon, and then stored at
-78 °C until analysis. The UA aliquots were simply stored at -78 °C
until analysis by HPLC.
r-TOH Analysis.28Samples of R-TOH were re-suspended in 0.5-
1.5 mL of HPLC solvent and analyzed by HPLC with electrochemical
detection. The mobile phase was composed of methanol/reagent alcohol
(60:40) and lithium perchlorate (20 mM). Electrochemical detector
settings were +0.60 to +0.65 V with a range setting of 0.5 µA. The
flow rate was 1.0 or 1.5 mL/min, which provided good separation of
δ-TOH and R-TOH within 7 min. External standard curves were created
by injecting 5-25 µL of working standard (R-TOH (9.0 µM) and
Acknowledgment. Support of this research from an NSF
Award (Grant No. 9996188) and NIH (HL17921) is gratefully
acknowledged.
Supporting Information Available: Experimental details
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA9934605
(29) Motchnik, P. A.; Frei, B.; Ames, B. N. Methods Enzymol. 1994,
234, 269-279.
(30) Komuro, E. K.; Takahashi, M.; Morita, T.; Tsuchiya, J.; Yoshitaka,
A.; Yamamoto, Y.; Niki, E. J. Phys. Org. Chem. 1990, 3, 309-315.
(31) Schumaker, V. N.; Puppione, D. L. Methods Enzymol. 1986, 128,
155-181.
(32) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. J. Biol.
Chem. 1951, 193, 265-275.
(33) (a) Sattler, W.; Mohr, D.; Stocker, R. Methods Enzymol. 1994, 233,
469-489. (b) Woodburn, K.; Kessel, D.; Sykes, E. Technical Information
Bulletin T-1734, Beckman Instruments, Inc., 1993.
(27) Kenar, J. A.; Havrilla, C. M.; Porter, N. A.; Guyton, J. R.; Brown,
S. A.; Klemp, K. F.; Selinger, E. Chem. Res. Toxicol. 1996, 9, 737-744.
(28) Lang, J. K.; Gohil, K.; Packer, L. Anal. Biochem. 1986, 157, 106-
116.