Antibody-catalyzed oxidative degradation of nicotine using riboflavin

Article abstract of DOI:10.1016/j.bmc.2004.07.006

Tobacco abuse remains a major cause of death worldwide despite ample evidence linking nicotine to various disease states. Consequently, immunopharmacotherapeutic approaches for the treatment of nicotine abuse have received increasing attention. Although a number of nicotine-binding antibodies have been disclosed, no antibody catalysts exist which efficiently degrade nicotine into pharmacologically inactive substances. Herein, we report the first catalytic antibodies which can oxidatively degrade nicotine. These biocatalysts use the micronutrient riboflavin and visible light as a source of singlet oxygen for the production of reactive oxygen species. Along with various known nicotine metabolites, antibody-catalyzed nicotine oxidations produce two novel nicotine oxidation products that were also detected in control ozonation reactions of nicotine. The reaction is efficient, with multiple turnovers of catalyst observed and total consumption of nicotine attained. These results demonstrate the potential of harnessing riboflavin as an endogenous sensitizer for antibody-catalyzed oxidations and demonstrate a new approach for the development of an active vaccine for the treatment of nicotine addiction using in vivo catalytically active antibodies.

Full text of DOI:10.1016/j.bmc.2004.07.006

Bioorganic & Medicinal Chemistry 12 (2004) 4981–4987
Antibody-catalyzed oxidative degradation of nicotine
using riboflavin
Tobin J. Dickerson, Noboru Yamamoto and Kim D. Janda*
Department of Chemistry and The Skaggs Institute for Chemical Biology, The Scripps Research Institute 10550 North
Torrey Pines Road La Jolla, CA 92037, USA
Accepted 21 June 2004
Available online 27July 2004
Abstract—Tobacco abuse remains a major cause of death worldwide despite ample evidence linking nicotine to various disease
states. Consequently, immunopharmacotherapeutic approaches for the treatment of nicotine abuse have received increasing atten-
tion. Although a number of nicotine-binding antibodies have been disclosed, no antibody catalysts exist which efficiently degrade
nicotine into pharmacologically inactive substances. Herein, we report the first catalytic antibodies which can oxidatively degrade
nicotine. These biocatalysts use the micronutrient riboflavin and visible light as a source of singlet oxygen for the production of
reactive oxygen species. Along with various known nicotine metabolites, antibody-catalyzed nicotine oxidations produce two novel
nicotine oxidation products that were also detected in control ozonation reactions of nicotine. The reaction is efficient, with multiple
turnovers of catalyst observed and total consumption of nicotine attained. These results demonstrate the potential of harnessing
riboflavin as an endogenous sensitizer for antibody-catalyzed oxidations and demonstrate a new approach for the development
of an active vaccine for the treatment of nicotine addiction using in vivo catalytically active antibodies.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
Efficient de novo catalysts for the degradation of drugs
of abuse have been highly sought after, especially in the
context of catalytic antibodies capable of cocaine
hydrolysis.5 However, no antibody catalysts have been
described in the context of nicotine addiction, although
numerous nicotine-binding antibodies are known.5 Our
interest in nicotine and the oxidative biochemical path-
ways utilized in its processing are grounded upon an
immunological viewpoint; we have focused efforts on
methods to procure tailored antibodies that not only
recognize nicotine, but catalyze its degradation. The lat-
ter of these two goals has proven significantly more
demanding as to obtain catalytic antibodies with the
energetic capabilities to degrade nicotine via an oxida-
tive mechanism, we were faced with a hapten design
challenge wherein neither a transition-state analogue,6
nor reactive immunization7 approach would be
applicable.
Nicotine is the most widely used addictive drug in the
world. The drug itself has been intimately linked with
cigarette smoking, the leading preventable cause of
death in the United States,1,2 hence, extensive studies
have been performed into both the psychoactive proper-
ties and metabolism of nicotine.3 The metabolic fate of
nicotine in mammals comes mainly from P-450 oxida-
tion and six main metabolites of nicotine are known
(Scheme 1). The exact mechanisms of formation of all
metabolites are not well understood; however, nicotine
is primarily converted to the pharmacologically inactive
and long-lived metabolite, cotinine or one of its corre-
sponding secondary metabolites, such as 3-hydroxycoti-
nine.4 In certain bacteria, tobacco alkaloids can serve as
a carbon and nitrogen source through catabolism by
various biochemical processes. Yet, there are no known
biological catalysts that specifically and efficiently de-
grade nicotine under physiologically relevant
conditions.
Clues for the design of antibody catalysts for the oxida-
tive degradation of nicotine come from both natural
metabolic pathways and work from Lerner and co-
workers whom have provided the first direct evidence
that all antibodies, regardless of specificity or source,
have the innate ability to produce reactive oxygen spe-
cies.8–12 The importance of these studies to our antibody
*
Corresponding author. Tel.: +1-858-784-2516; fax: +1-858-784-
2595; e-mail: kdjanda@scripps.edu
0968-0896/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.07.006

4982
T. J. Dickerson et al. / Bioorg. Med. Chem. 12 (2004) 4981–4987
O
N
N
Cotinine
N
N
O
N
N
Nicotine-N-oxide
N-Methylnicotinium
Ion
N
N
Nicotine
N
N
N
N
Nicotine-∆4',5'-enamine
HO
O
–
HO
CO2
N
H
OH
N
Nicotine
Glucuronide
Nornicotine
Scheme 1. The six major metabolites of nicotine in mammals. The major metabolites, cotinine, and nicotine-N-oxide, are clearly derived from
oxidation processes, but all are synthesize via formal oxidations.
catalyst efforts include the observation that for the gen-
eration of these oxidants to occur, a source of singlet
oxygen (1O2) is necessary. Given a suitable source of
1O2, an antibody oxidative pathway for nicotine degra-
dation could parallel many of the oxidative products
found in natural biochemical systems, albeit from a
completely different source.
tine addiction.5 Recently, antibodies that specifically rec-
ognize nicotine have emerged as a promising method for
attenuating the psychoactive effects of nicotine.5,14 Be-
cause of its addictive properties we viewed nicotine as
an attractive target for the development of novel anti-
body-based treatments, however, as there are no cata-
lysts either enzyme or antibody based that can
specifically and efficiently degrade this molecule, it also
represented a formidable energetic challenge for anti-
body catalysis.
To this end, an efficient endogenous photosensitizing
chromophore was required. Many sensitizers, such as
porphyrins, could have been investigated, however, we
chose riboflavin, or vitamin B2. Our reasoning for exam-
ining riboflavin stems from reports, which describe the
interaction of riboflavin with immunoglobulins in ser-
um, however, the functional significance of this interac-
tion has, to date, not been understood.13 Based upon the
inherent ability of riboflavin to sensitize molecular oxy-
gen, coupled with its proximity to circulating antibodies,
and the antibody oxidative cycle detailed by Lerner and
co-workers, we have sought to investigate whether the
immune system could contain a set of catalytic antibod-
In an effort to identify systems where antibody-catalyzed
oxidation of nicotine could be feasible, we examined the
ability of the micronutrient riboflavin to assist in
this energetically demanding oxidation. Antibodies are
known to interact with riboflavin in vivo, however, the
significance of antibody–riboflavin union is still cryptic
since its discovery 20 years ago.13 In light of the recent
discovery that all antibodies can produce reactive oxy-
gen species when given a source of singlet oxygen,8–12
we hypothesized that a subset of antibodies could use
riboflavin as an innate source of singlet oxygen, thereby
unifying foreign particle recognition and destruction
into one molecular complex. In this context, the oxida-
tion of a bound protein or organic molecule could occur
catalytically through the ternary interaction between
antibody–antigen and riboflavin.
1
ies that utilize riboflavin as a source of O2. Thus, the
fundamental nature of an antibody–riboflavin chemical
interaction could provide a functional oxidation system
in vivo for the catalytic destruction of nicotine and pro-
vide a powerful tool in the treatment of nicotine addic-
tion by passive, or possibly active, immunization.
To characterize potential catalytic antibodies for oxida-
tive nicotine degradation, we first examined several con-
trol oxidation reactions for both product identity and
relative distribution. Because antibodies have been
shown to produce H2O2,8,9 we first treated nicotine with
this oxidant. As previously reported, this reaction yields
nicotine-N-oxide as the major product.15 Our second
2. Results and discussion
In light of the countless deaths attributable to tobacco
abuse each year and the low success rates of current nico-
tine cessation therapies, extensive efforts have been
devoted to the production of novel treatments for nico-

T. J. Dickerson et al. / Bioorg. Med. Chem. 12 (2004) 4981–4987
4983
oxidant examined was ozone, again, based on the work
H
of Lerner and co-workers.10–12 Treatment of nicotine
with ozone provided seven major products as identified
by cation-exchange HPLC. Five of these products, nor-
nicotine, anabasine, b-nicotyrine, cotinine, and nicotine-
N-oxide have been reported previously to be formed
with ozone,16 while two new products, 1 and 2, have
not (Scheme 2). We have isolated these compound and
surprisingly, neither corresponded to common nicotine
metabolites. Purification and analytical characterization
revealed the structures of these compounds to be 3-(1-
hydroxy-2-pyrrolidinyl)pyridine 1 and polyhydroxylated
cotinine derivative 2 (Fig. 2). Lastly, the reaction of nic-
otine with potassium superoxide was studied, however
presumably due to its instability in water, only hydrogen
peroxide was generated, yielding nicotine-N-oxide.
N
N
Nicotine
H
N
H
N
OH
N
O
O
3
Figure 1. Structures of nicotine and structurally congruent hapten 3
used to elicit tight binding nicotine antibodies.
To elicit antibody catalysts for nicotine degradation, a
traditional transition-state analogue approach was
abandoned for a hapten–substrate tact. Our reasoning
here was similar to previous reports from our group
on photochemical antibody-catalyzed reactions where
an external source of energy was present.17,18 We ini-
tially screened a panel of monoclonal antibodies previ-
ously elicited against a hapten 3, which is highly
congruent to nicotine (Fig. 1).19 Antibodies in this panel
bound nicotine in the range of nanomolar to micro-
molar affinity. However, no catalysts were observed
after either irradiation with UV light, or irradiation with
white light in the presence of riboflavin. Interestingly,
some of these antibodies actually protected nicotine
from oxidation relative to control reactions containing
nicotine and riboflavin alone.
hapten 4 (hapten code: TD1) was synthesized from nor-
nicotine and glutaric anhydride to elicit nicotine-binding
antibodies (Scheme 2). Our design principle was thus
grounded upon two aspects: (1) Substrate recognition
by antibodies to 4 would be sufficient, but not tight as
seen previously with hapten 3; thus, 4 represents the glo-
bal scaffold of nicotine less the charge on the pyrrolidine
nitrogen. (2) The oxidation of nicotine dramatically
changes the charge distribution within the molecule
(e.g., formation of N-oxides), and as such, product inhi-
bition using antibodies raised against 4 was expected to
be minimal. Hapten 4 was conjugated to keyhole limpet
hemocyanin (KLH) and bovine serum albumin (BSA)
using standard protocols and mice were immunized with
the 4-KLH conjugate. The titer against 4-KLH was
measured to be ꢀ25,000, indicating that the immuno-
genicity of hapten 4 was comparable to titers observed
in our active cocaine vaccine studies20 and significantly
greater than our previous nicotine hapten 3.19 The
resulting panel of 26 monoclonal antibodies was puri-
fied21 and screened for nicotine oxidation in the presence
of riboflavin. The range of binding of these antibodies to
In light of these findings, we envisioned that antibodies
that bound nicotine weakly would provide a reasonable
local concentration of nicotine in the vicinity of the anti-
body without protecting the nicotine molecule from oxi-
dation by generated reactive oxygen species. To this end,
O
N
N
N
O
H
N
N
N
O3
N
N
N
H
N
N
N
OH
O
N
N
OH
O
N
H
O
N
1
2
O
O
O
N
N
O
H
DIEA
72%
O
OH
N
N
4
Scheme 2. Chemical ozonation of nicotine and hapten 4 used to elicit nicotine-binding antibodies.

4984
T. J. Dickerson et al. / Bioorg. Med. Chem. 12 (2004) 4981–4987
nicotine was much weaker as expected from our hapten
design and was found to be between 0.2 and 50mM as
determined by competition ELISA. Using riboflavin
(60lM) and a source of white light, selected monoclonal
antibodies from the panel raised against 4-KLH were
able to catalyze the oxidation of nicotine to products
strikingly similar to the ozone positive control reaction.
Of the 26 antibodies screened, the most proficient
were TD1-10E8 (Kd,app =6.3mM) and TD1-36H10
(Kd,app =6.3mM), which were selected for further study.
The products that were observed in both the uncata-
lyzed synthetic ozone control reaction as well as with
TD1-10E8 and TD1-36H10 oxidations were identical
with regard to the seven major products. In essence, syn-
thetically produced ozone provided an accurate Ôblue-
printÕ for the antibody-catalyzed oxidations. Of the
seven products observed with antibodies TD1-10E8
and TD1-36H10, the greatest overall rate acceleration
was seen with compounds 1 and 2. Both of these com-
pounds were formed at approximately 10-fold rate
enhancements over the uncatalyzed reactions of nicotine
and riboflavin alone (Fig. 2). Control reactions in which
visible light or oxygen were omitted showed no produc-
tion of compounds 1, 2, or degradation of nicotine. The
presence of these two new products is interesting since a
powerful oxidant must be invoked in the formation of 1
as it has only previously been observed in metal-cata-
lyzed oxidation systems.22 Furthermore, compound 2
also implicates potent oxidants, potentially hydroxyl
radical or a more stable version of this oxidant,23 capa-
ble of heavily oxidizing bound antigens.
A more detailed analysis of the formation of 1 and 2 was
performed to glean additional information into the spe-
cific nature of this reaction. The kinetic profile of the
reaction revealed that approximately 50% of the total
nicotine was converted to products 1 and 2 (Fig. 2). In
the presence of TD1-10E8, the initial rates of formation
of 1 and 2 were 0.54 and 0.67lM/min, respectively, and
the relative formation of 1:2 was found to be 1.05:1 (Fig.
2). In contrast, TD1-36H10 formed 1 and 2 at initial
rates of 0.67and 0.49 lM/min, respectively, and pro-
duced these compounds in a 1:1.75 ratio (Fig. 2). In
comparison, the uncatalyzed control reaction in which
nicotine is treated with ozone produces these two prod-
ucts in a ratio of 1.62:1. These differences are remarka-
ble, and presumably, these two antibodies have
unique combining site architectures that allows for
altered product distributions and a disfavored reaction
ratio.24
Figure 2. Time course plots of the formation of 1 (A) and 2 (B). Filled
circles: TD1-10E8, riboflavin, nicotine, and visible light; Filled squares:
TD1-36H10, riboflavin, nicotine, and visible light; Filled diamonds:
riboflavin, nicotine, and visible light. Data was fit using Kaleidagraph
3.0.5 (Abelbeck software) using nonlinear regression analysis.
such as the use of superoxide dismutase to destroy any
superoxide, also did not diminish catalytic activity. In
the presence of SOD (150U/mL), no effect was observed
on the product ratio (1:1.03, 1:2). In total, this data indi-
cates that although the identity of the antibody-pro-
duced reactive oxygen species is unknown, this oxidant
is likely protected from bulk solvent, yet accessible to
the antibody combining site.
To inhibit the reaction, hapten 4 could not be used as its
structural congruency is too close to the substrate and
would presumably undergo oxidation, thus we explored
the use of various reactive oxygen species scavengers
including methioinine and thiosulfate.25 Neither com-
pound eliminated the catalytic activity of TD1-10E8; in-
deed, methionine had no effect on the product ratio
(1.02:1, 1:2), while a drastically different ratio of com-
pounds was observed in antibody-catalyzed reactions
in the presence of thiosulfate (2.02:1, 1:2). Furthermore,
other methods for quenching reactive oxygen species,
It is critical to note that independent of the identity of the
reactive oxygen species, after only 6h of white light irra-
diation in the presence of either antibody (20lM), no
detectable nicotine remains in the reaction mixtures
(starting nicotine concentration=200lM), implying
multiple turnover events by the antibody. As a further
control, nicotine was also irradiated in the presence of
riboflavin and BSA, a protein known to bind small mole-
cules and accelerate some reactions.26 Product ratios
in this reaction greatly differed from our antibody-

T. J. Dickerson et al. / Bioorg. Med. Chem. 12 (2004) 4981–4987
4985
catalyzed reactions (1:4, 1:2). Furthermore, in contrast to
our antibody-catalyzed reactions, following a short burst
of reactive oxidant production (<1 turnover), a drastic
decrease in rate due to photooxidation of the protein
was observed.9
iodine, dinitrophenylhydrazine, ceric ammonium mo-
lybdate, ninhydrin, or potassium permanganate as
appropriate. H NMR spectra were recorded on either
1
a Varian INOVA-400 or Bruker DRX-600 spectrometer
at 400 and 600MHz, respectively. 13C NMR spectra
were recorded on a Varian INOVA-400 spectrometer
at 100MHz. Matrix-assisted laser desorption/ionization
(MALDI) FTMS experiments were performed on an
IonSpec FTMS mass spectrometer. Electrospray ioniza-
tion (ESI) mass spectrometry experiments were
performed on an API 100 Perkin Elmer SCIEX single
quadrupole mass spectrometer. Analytical high pressure
liquid chromatography (HPLC) was performed on a
Hitachi D-7000 instrument equipped with a Pheno-
menex Partisil 10 SCX cation-exchange analytical
column, a Hitachi L-7100 pump, Hitachi L-7200
autosampler, a Hitachi L-7455 diode array detector,
and mobile phases comprised of 0.3M sodium acetate/
methanol (95:5, v/v, pH4.5), at a flow rate of
1mL/min. The L-7100 pump was controlled using
Hitachi-HSM software on a Dell Optiplex GX110 PC
computer.
A subset of immunoglobulins has been previously
reported to bind riboflavin,13 and as such, we tested
the most efficient antibody catalysts for binding by equi-
librium dialysis to determine if riboflavin binding was
essential for efficient transfer of singlet oxygen to the
antibody. We observed no specific riboflavin binding
(Kd P 1mM), thus it is clear that binding is not a pre-
requisite for antibody-catalyzed oxidation processes. In-
deed, these results demonstrate that antibodies,
independent of intimate riboflavin-binding ability, can
utilize riboflavin as an innate sensitizer and source of
singlet oxygen for the production of powerful oxidants.
3. Conclusion
In summary we have demonstrated that antibody-cata-
lyzed oxidation of nicotine is feasible, and may be an
unrecognized method for the destruction of nicotine in
biological systems. The concentrations of riboflavin
(ꢀ10nM)27 and specific antibody (ꢀ1lM)28 in vivo
are reasonable for riboflavin-assisted antibody oxida-
tions to have biological relevance. We emphasize that
the oxidation and product distribution seen for nicotine
degradation would not have previously been envisioned
to be in the repertoire of antibody catalysis based on tra-
ditional hapten design using transition-state or reactive
immunization theory.
4.2. Synthesis of TD1 Hapten 4
Glutaric anhydride (77mg, 0.68mmol), N,N-dimethyl-
aminopyridine (8mg, 0.07mmol), and diisopropylethyl-
amine (235lL, 1.35mmol) were added to a solution of
nornicotine (100mg, 0.68mmol) in CH2Cl2 (6mL), at
room temperature and the reaction mixture allowed to
stir for 18h. The crude mixture was concentrated, and
then purified by preparative TLC using CHCl3/MeOH
(3:2) as the eluent. The desired band was triturated with
CHCl3/MeOH (3:2) and concentrated to obtain the TD1
hapten 4 as a clear oil (152mg, 86%). Hapten conjuga-
tion to BSA and KLH and monoclonal antibody pro-
duction was performed. Approximate loadings of the
TD1-BSA conjugate were determined by MALDI-
TOF mass spectrometry and revealed that approxi-
mately 19 copies of the hapten were present per BSA
molecule.
Our active vaccine approach does not require the pres-
ence of tight-binding antibodies, and in fact, these re-
sults suggest that antibodies with modest affinity may
serve as more efficient catalysts for nicotine destruction.
Furthermore, a weaker binding constant may also allow
for the binding of various pharmacologically and chem-
ically active nicotine metabolites, such cotinine29 and
nornicotine,30,31 and their subsequent oxidation and
clearance. Hapten 4 is highly immunogenic, as is evi-
denced by the high circulating titer in mice, and could
lead to the production of a novel active vaccine in which
multiple catalytic antibodies could be elicited in vivo
against a given drug of abuse. We envision that this
strategy is applicable to the destruction of other drugs
of abuse containing bonds that can be oxidized (T.J.D
and K.D.J., unpublished results).
4.3. Nicotine competition ELISA
An ELISA plate (CoStar; 96 well) was coated uncov-
ered overnight at 37°C with 25lL of a nicotine-BSA
conjugate previously reported to resemble free nico-
tine22 diluted to 5lg/mL. To this, methanol (50lL/well)
was added and allowed to sit 5min at room tempera-
ture. The methanol was then removed from the wells
and the plates allowed to dry. Wells were then blocked
with 50lL of Blotto (5% skim milk powder in PBS) for
5min at 37°C. Typically, 25lL of a solution containing
a 1:100 dilution of mAb in Blotto along with varied se-
rial dilutions of nicotine (50lM–50mM) was then
added and incubated 1h at 37°C. After washing,
25lL of a 1:1000 dilution of a goat-antimouse/horse-
radish peroxidase conjugate in Blotto was added and
incubated for 30min. The plate was developed with
the colorimetric reagent ABTS for 30min and the
absorbance measured on an ELISA plate reader at
405nm.
4. Experimental
4.1. General synthetic methods
Unless otherwise stated, all reactions were performed
under an inert atmosphere with dry reagents and sol-
vents and flame-dried glassware. Analytical thin-layer
chromatography (TLC) was performed using 0.25mm
precoated silica gel Kieselgel 60 F254 plates. Visualiza-
tion of the chromatogram was by UV absorbance,

4986
T. J. Dickerson et al. / Bioorg. Med. Chem. 12 (2004) 4981–4987
4.4. Riboflavin equilibrium dialysis
3.06 (s, 3H), 3.22 (m, 2H), 3.97(d, J=6.2Hz, 1H), 5.26
(m, 1H). 13C NMR (D2O): d 27.2, 29.8, 34.4, 44.3, 62.9,
73.4, 167.0, 171.9, 176.5, 207.7. ESI-MS: Found 247
(M+Na+). MALDI-FTMS for C10H13N2O4 (M+H+)
calculated 224.0797, found 224.0799.
Equilibrium dialysis was performed using [3H]-ribofla-
vin as ligand and the following antibodies: TD1-10E8
and TD1-36H10. The antibodies were serially diluted
in PBS (pH7.4), starting from 50lM, and added to wells
in a 96-well microtiter plate (150lL/well). Wells in a sec-
ond microtiter plate were filled with [3H]-riboflavin
(150lL, 5nM, 0.5lCi) in PBS (pH7.4). The two plates
were tightly connected with filled wells facing each other
and separated with a dialysis membrane (molecular
weight cutoff 6000–8000Da). The plates were attached
vertically to a shaker and were shaken at 24h at room
temperature, after which they were carefully separated.
The membrane was discarded and from each well,
100lL was transferred to a scintillation vial. To each
solution, scintillation fluid (5mL) was added and the
radiation counted for 5min/sample. The experiment
was repeated twice for each serum sample. The average
in differences in DPM (disintegrations per minute)
between opposite wells was determined for each mAb
concentration.
4.8. Nicotine oxidation using ultraviolet light
Antibody stock concentrations were determined by
UV–Vis spectroscopy using 1 OD=1.25mg/mL at
280nm and 1mg/mL=6.7lM. High-intensity photolysis
was performed in a Rayonet photochemical reactor
with RPR-3500 bulbs. The assay was initiated by the
addition of a solution of antibody (20lM in PBS) to
a solution of nicotine (200lM) in buffer (10mM PBS,
pH7.4) in 4mL glass vials. High-intensity photolysis
was performed for 10min intervals, and at times
throughout the assay, aliquots (50lL) were removed
and directly injected onto the analytical HPLC system
described above.
4.9. Nicotine oxidation using riboflavin as a visible light
sensitizer
4.5. Oxidation of nicotine by chemically supplied ozone
A solution of nicotine (200lM), riboflavin (60lM), and
antibody (20lM) in PBS (pH7.4) was irradiated on a
white light transilluminator (2.8mW/cm2) at 4°C in
duplicate in a standard polystyrene ELISA plate. The
final volume of each reaction was 200lL. At given time
points throughout the assay, aliquots (50lL) were
removed and directly injected onto the analytical HPLC
system described above. The following control reactions
were also performed: nicotine, riboflavin, and antibody in
the dark; nicotine, riboflavin, and visible light; nicotine
and riboflavin in the dark; antibody, riboflavin, nicotine,
and visible light in degassed PBS. In all cases, no product
was observed in any control reaction.
Nicotine (50mg, 0.31mmol) was dissolved in CH2Cl2
(10mL). The reaction mixture was then cooled to
À78°C and treated with ozone for 30min. After this
time, the reaction mixture was concentrated to yield
68mg of a brown thick tar. A sample (ꢀ1mg) of the
tar was removed, dissolved in PBS (2mL, 10mM,
pH7.4) and analyzed by HPLC using the system de-
scribed above.
4.6. Isolation and spectral characterization of hydroxyl-
amine 1
Nicotine ozonation reactions were performed with
chemically supplied ozone as described above (vide
supra). A sample from the resulting product mixture
(ꢀ4mg) was dissolved in phosphate buffer (10mM,
pH7.4) and injected onto the ion-exchange HPLC sys-
tem described above. The peak with retention time
4.2min was isolated and the solvent removed by lyophil-
4.10. Reactive oxygen inhibition experiments
A solution of nicotine (200lM), riboflavin (60lM),
antibody (20lM), and methionine (1mM) or superoxide
dismutase (150U/mL) in PBS (pH7.4) was irradiated on
a white light transilluminator (2.8mW/cm2) at 4°C in
duplicate in a standard polystyrene ELISA plate. The
final volume of each reaction was 200lL. At given time
points throughout the assay, aliquots (50lL) were
removed and directly injected onto the analytical HPLC
system described above.
1
ization. H NMR (D2O): d 1.92 (m, 2H), 2.03 (m, 1H),
2.34 (m, 1H), 3.20 (m, 1H), 3.48 (m, 1H), 3.77 (d,
J=6.7Hz, 1H), 7.19 (m, 1H), 7.84 (m, 1H), 8.48 (m,
1H), 8.55 (m, 1H). 13C NMR (D2O): d 24.3, 31.5,
59.0, 71.9, 123.5, 134.5, 137.1, 148.0, 149.2. ESI-MS:
Found 187(M+Na +). MALDI-FTMS for C9H13N2O
(M+H+) calculated 165.1028, found 165.1027.
Acknowledgements
4.7. Isolation and spectral characterization of polyhy-
droxylated cotinine 2
We thank Drs. Richard Lerner and Paul Wentworth, Jr.
for discussions and disclosure of their unpublished
results in the area of riboflavin–antibody interactions.
We also thank Dr. Michael Meijler for assistance with
the equilibrium dialysis studies. This work was finan-
cially supported by The Skaggs Institute for Chemical
Biology and the National Institute on Drug Abuse
(R01 DA015700 to K.D.J. and F31 DA015973 to
T.J.D.). The Eisai Co. Ltd is gratefully acknowledged
for sabbatical support to N.Y.
Nicotine ozonation reactions were performed with
chemically supplied ozone as described above (vide
supra). A sample of the resulting product mixture
(ꢀ4mg) was dissolved in phosphate buffer (10mM,
pH7.4) and injected onto the ion-exchange HPLC sys-
tem described above. The peak with retention time
5.2min was isolated and the solvent removed by lyoph-
ilization. 1H NMR (D2O): d 2.09 (m, 2H), 2.68 (m, 2H),

T. J. Dickerson et al. / Bioorg. Med. Chem. 12 (2004) 4981–4987
4987
14. Carrera, M. R. A.; Ashley, J. A.; Hoffman, T. Z.; Isomura,
S.; Wirsching, P.; Koob, G. F.; Janda, K. D. Bioorg. Med.
Chem. 2004, 12, 563.
15. Crooks, P. A.; Pool, W. F.; Damani, L. A.; Winkle, S. A.
Chem. Ind. 1988, 3, 95.
References and notes
1. Nelson, D. E.; Kirkendall, R. S.; Lawton, R. L.; Chris-
mon, J. H.; Merritt, R. K.; Arday, D. A.; Giovino, G. A.
Morbid. Mortal. Wkly. Rep. 1994, 43, 1.
16. Park, T.-K.; Kim, K.-H.; Huh, T.-S. J. Korean Chem. Soc.
1984, 28, 323.
17. Taylor, M. J.; Hoffman, T. Z.; Yli-Kauhaluoma, J. T.;
Lerner, R. A.; Janda, K. D. J. Am. Chem. Soc. 1998, 120,
12783.
18. Dickerson, T. J.; Tremblay, M. R.; Hoffman, T. Z.; Ruiz,
D. I.; Janda, K. D. J. Am. Chem. Soc. 2003, 125, 15395.
19. Isomura, S.; Wirsching, P.; Janda, K. D. J. Org. Chem.
2001, 66, 4115.
20. Carrera, M. R. A.; Ashley, J. A.; Parsons, L. H.;
Wirsching, P.; Koob, G. F.; Janda, K. D. Nature 1995,
378, 727.
2. US Dept. Health Hum. Serv. Reducing the Health
Consequences of Smoking. 25 Years of Progress. A
Report of the Surgeon General; Public Health Services:
Rockville, MD, 1989.
3. Kyerematen, G. A.; Vesell, E. S. Drug Metab. Rev. 1991,
23, 3.
4. Oelschlager, H.; Al Shaik, M. In Biological oxidation of
nitrogen in organic molecules; Gorrod, J. W., Damani, L.
A., Eds.; Ellis Horwood: Chinchester, 1985; p 43.
5. Meijler, M. M.; Matsushita, M.; Wirsching, P.; Janda, K.
D. Curr. Drug Disc. Technol. 2004, 1, 77, and references
cited therein.
21. Kohler, G.; Milstein, C. Nature 1975, 256, 495.
22. Murahashi, S.-I.; Mitsui, H.; Shiota, T.; Tsuda, T.;
Watanabe, S. J. Org. Chem. 1990, 55, 1736.
23. Wentworth, P., Jr.; Wentworth, A. D.; Zhu, X.; Wilson, I.
A.; Janda, K. D.; Eschenmoser, A.; Lerner, R. A. Proc.
Natl. Acad. Sci. U.S.A. 2003, 100, 1490.
24. McDunn, J. E.; Dickerson, T. J.; Janda, K. D. In Catalytic
Antibodies; Keinan E., Ed., Wiley: New York; in press.
25. Kanofsky, J. R.; Sima, P. D. Arch. Biochem. Biophys.
1995, 316, 52.
26. Hollfelder, F.; Kirby, A. J.; Tawfik, D. S. Nature 1996,
383, 60.
27. Hustad, S.; McKinley, M. C.; McNulty, H.; Schneede, J.;
Strain, J. J.; Scott, J. M.; Ueland, P. M. Clin. Chem. 2002,
48, 1571.
28. Janoff, E. N.; Hardy, W. D.; Smith, P. D.; Wahl, S. M. J.
Immunol. 1991, 147, 2130.
29. Kennan, R. M.; Katsukami, D. K.; Pentel, P. R.;
Thompson, T. N.; Grillo, M. A. Clin. Pharmacol. Ther.
1994, 55, 581.
30. Bardo, M. T.; Green, T. A.; Crooks, P. A.; Dwoskin, L. P.
Nornicotine is Self-Administered Intravenously by Rats.
Psychopharmacology 1999, 146, 290.
6. Xu, Y.; Yamamoto, N.; Janda, K. D. Bioorg. Med. Chem.,
in press.
7. Wirsching, P.; Ashley, J. A.; Lo, C.-H. L.; Janda, K. D.;
Lerner, R. A. Science 1995, 270, 1775.
8. Wentworth, A. D.; Jones, L. H.; Wentworth, P., Jr.;
Janda, K. D.; Lerner, R. A. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 10930.
9. Wentworth, P., Jr.; Jones, L. H.; Wentworth, A. D.; Zhu,
X.; Larsen, N. A.; Wilson, I. A.; Xu, X.; Goddard, W. A.,
III; Janda, K. D.; Eschenmoser, A.; Lerner, R. A. Science
2001, 293, 1806.
10. Wentworth, P., Jr.; McDunn, J. E.; Wentworth, A. D.;
Takeuchi, C.; Nieva, J.; Jones, T.; Bautista, C.; Ruedi, J.
M.; Gutierrez, A.; Janda, K. D.; Babior, B. M.; Esc-
henmoser, A.; Lerner, R. A. Science 2002, 298,
2195.
11. Babior, B. M.; Takeuchi, C.; Ruedi, J. M.; Gutierrez, A.;
Wentworth, P., Jr. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
3031.
12. Wentworth, P., Jr.; Nieva, J.; Takeuchi, C.; Galve, R.;
Wentworth, A. D.; Dilley, R. B.; DeLaria, G. A.; Saven,
A.; Babior, B. M.; Janda, K. D.; Eschenmoser, A.; Lerner,
R. A. Science 2003, 302, 1053.
31. Dickerson, T. J.; Janda, K. D. Proc. Natl. Acad. Sci.
U.S.A. 2002, 99, 15084.
13. Innis, W. S. A.; McCormick, D. B.; Merrill, A. H., Jr.
Biochem. Med. 1985, 34, 151.