[CANCER RESEARCH 51. 4371-4377. August 15. 1991]
Frequency of Urination and Its Effects on Metabolism, Pharmacokinetics, Blood
Hemoglobin Adduci Formation, and Liver and Urinary Bladder DNA Adduct
Levels in Beagle Dogs Given the Carcinogen 4-Aminobiphenyl1
Fred F. Kadlubar, Kenneth L. Dooley, Candee H. Teitel, Dean W. Roberts, R. Wayne Benson, Mary Ann Butler,
John R. Bailey, John F. Young, Paul W. Skipper, and Steven R. Tannenbaum
The National Center for Toxicological Research, Jefferson, Arkansas 72079 fF. F. K., K. L. D., C. H. T., D. W. R., R. W. B., M. A. B., J. R. B., J. F. Y.J, and the
Massachusetts Institute of Technology, Cambridge, Massachusetts 02178 [P. W. S., S. R. T.]
ABSTRACT
likely to be a direct consequence of the hepatic /V-oxidation of ABP to
N-OH-ABP and its entry into the circulation; and the measurement of
The human urinary bladder carcinogen, 4-aminobiphenyl (ABP), is
known to undergo hepatic metabolism to an A-hydroxy arylamine and its
corresponding /V-glucuronide. It has been proposed that these metabolites
are both transported through the blood via renal filtration to the urinary
bladder lumen where acidic pH can facilitate the hydrolysis of the \glucuronide and enhance the conversion of A"-hydroxy-4-aminobiphenyl
hemoglobin adduct levels can be regarded as valid index of both exposure
and metabolic activation of this carcinogen in humans. In addition, these
studies provide strong evidence for the role of unconjugated urinary NOH-ABP in carcinogen-DNA adduct formation in the carcinogen-target
tissue and suggest that the frequency of urination should be a critical
determinant in human urinary bladder carcinogenesis.
(N-OH-ABP)
to a reactive electrophile that will form covalent adducts
with urothelial DNA. Blood ABP-hemoglobin adducts, which have been
used to monitor human exposure to ABP, are believed to be formed by
reactions within the erythrocyte involving N-OH-ABP that has entered
the circulation from the liver or from reabsorption across the urothelium.
To test these hypotheses directly, experimental data were obtained from
female beagles given [3H)ABP (p.o., i.V., or intraurethrally), |3H|N-OHABP (i.v. or intraurethrally), or |3H|N-OH-ABP
W-glucuronide (i.V.).
Analyses included determinations of total ABP in whole blood and
plasma, ABP-hemoglobin
adducts in blood erythrocytes, ABP and V
OH-ABP levels (free and A'-glucuronide) in urine, urine pH, frequency
of urination (controlled by urethral catheter), rates of reabsorption of
ABP and N-OH-ABP across the urothelium, and apparent volumes of
distribution in the blood/tissue compartment. The major ABP-DNA
adduct, AHguan-8-yl)-4-aminobiphenyl,
was also measured in urothelial
and liver DNA using a sensitive immunochemical method. An analog/
digital hybrid computer was then utilized to construct a multicompartmental pharmacokinetic model for ABP and its metabolites that sepa
rates: (a) absorption; (b) hepatic metabolism and distribution in blood
and tissues; (c) ABP-hemoglobin adduct formation; (d) hydrolysis and
reabsorption in the urinary bladder lumen; and (e) excretion. Using this
model, cumulative exposure of the urothelium to free N-OH-ABP was
simulated from the experimental data and used to predict ABP-DNA
adduct formation in the urothelium.
The results indicated that exposure to N-OH-ABP and subsequent
ABP-DNA adduct formation are directly dependent on voiding frequency
and to a lesser extent on urine pH. This was primarily due to the finding
that, after p.o. dosing of ABP to dogs, the major portion of the total NOH-ABP entering the bladder lumen was free N-OH-ABP (0.7% of the
dose), with much lower amounts as the acid-labile A'-glucuronide (0.3%
of the dose). The fraction of free/conjugated
N-OH-ABP
then decreased
with greater voiding intervals, as a consequence of the rapid reabsorption
of free N-OH-ABP across the urothelium. Accordingly, the model cor
rectly predicted a 17-fold decrease in the level of urothelial DNA adducts
as average voiding frequency was increased from 3 times a day to a
continuous voiding for the first 12 h after dosing. This was also consistent
with data from intraurethral instillation of N-OH-ABP, which showed a
>60-fold higher DNA adduct level than similarly administered ABP. In
contrast, blood ABP-hemoglobin adduct levels, which accounted for 810% of the p.o. dose, were independent of voiding interval; and the
amounts of free N-OH-ABP available for reabsorption in the bladder
lumen were insufficient to account for the high ABP-hemoglobin adduct
levels observed.
Thus, these data indicate that hemoglobin-ABP
adduct formation is
INTRODUCTION
ABP2 is strongly carcinogenic for the urinary bladder of both
dogs and humans (1), and evidence for two separate metabolic
activation pathways has been presented. The first involves a
hepatic cytochrome P-450IA2-catalyzed /V-oxidation (2, 3) to
yield N-OH-ABP, which can be subsequently conjugated by
hepatic glucuronyltransferases to form an /V-hydroxy arylamine
/V-glucuronide (4). Both N-OH-ABP and its /V-glucuronide may
enter the circulation, undergo renal filtration, and thus enter
the lumen of the urinary bladder for subsequent excretion. In
the blood, a major proportion of the free N-OH-ABP appears
to be further oxidized within the erythrocyte to 4-nitrosobiphenyl which then forms a covalent addition product with a
single cysteine on hemoglobin, accounting for up to 10% of the
ABP dose administered (5, 6). In the urinary bladder lumen,
infrequent voiding and an acidic urine would be expected to
promote the hydrolysis of the A'-glucuronide and the release of
more N-OH-ABP, followed by its reabsorption and covalent
binding to urothelial DNA (7). N-OH-ABP and other /V-hy
droxy arylamines have long been shown to be direct-acting
mutagens, to react spontaneously with nucleic acids and pro
teins under slightly acidic conditions, and to induce tumor
formation at sites of application (reviewed in Refs. 8 and 9).
Accordingly, in a previous study (10), urine pH and voiding
interval were proposed as important pharmacokinetic determi
nants for arylamine-induced urinary bladder carcinogenesis. In
addition, a three-compartment pharmacokinetic model was de
veloped, based on data from 2-naphthylamine-treated rats, that
was predictive of individual and species differences in suscep
tibility to urinary bladder carcinogenesis by 2-naphthylamine.
A second metabolic activation pathway for arylamine bladder
carcinogens has been suggested to involve the peroxidation of
ABP directly in the carcinogen-target tissue by the action of
urothelial prostaglandin hydroperoxidase. This peroxidase,
which is present at relatively high levels in the bladder epithe
lium of both dogs (11) and humans (12), mediates the arachidonic acid-dependent cooxidation of several arylamines to
DNA-bound adducts (13). For benzidine, 2-aminofluorene, and
2-naphthylamine, specific diimine or iminoquinone metabolites
have been implicated as reactive intermediates leading to the
Received 1/2/91; accepted 5/30/91.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
2The abbreviations used are: ABP, 4-aminobiphenyl; N-OH-ABP, iV-hydroxy1Supported in part by NIH Grants ES-00-597 and ES-02-109 to P. W. S. and
4-aminobiphenyl; HPLC, high pressure liquid chromatography.
S. R. T.
4371
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.
URINARY FREQUENCY AND ADDUCT FORMATION
formation of DNA adducts that are unique to the peroxidative
pathway (14-16). Studies with ABP have provided similar
results and indicate that the major peroxidase-derived adduci,
although not yet identified, is chromatographically
distinct
from the major adducts formed by reaction of N-OH-ABP with
DNA (17). The predominant DNA adducts from N-OH-ABP
have been characterized previously as yV-(deoxyguanosin-S-yl)ABP, ./V-ideoxyguanosin-A^-yO-ABP, and A/-(deoxyadenosin-8yl)-ABP (8, 18). While the peroxidative pathway appears to be
primarily responsible for urothelial DNA adduct formation
from benzidine in the dog (14, 19), only about 20% of the DNA
adducts formed from 2-naphthylamine (16) and about 10% of
those formed from ABP (17) in the dog can be attributed to
peroxidative activation.
Therefore, the reaction of urinary N-OH-ABP with urothelial
DNA remains a probable mechanism to account for the major
DNA adducts formed from ABP in vivo. In this regard, N(deoxyguanosin-S-yl)-ABP is not only the predominant adduct
formed from N-OH-ABP in vitro but has also been shown to
represent 65-70% of the total adducts found in the dog bladder
epithelium and to be persistent in the target tissue (17, 18). In
order to test this hypothesis further and to assess the relative
exposure of the bladder epithelium to N-OH-ABP as a function
of voiding interval and urine pH, experimental studies were
undertaken in the dog and a complex pharmacokinetic model
was developed to determine the effect of these variables on
urothelial ABP-DNA adduct formation. In addition, ABPblood hemoglobin adduct levels, which have been shown to be
an index of human exposure to ABP (6), were determined
concomitantly in order to examine the relative contributions of
initial hepatic TV-oxidation and subsequent urinary N-OH-ABP
reabsorption to total hemoglobin adduct formation.
MATERIALS
AND METHODS
Chemicals. [2,2'-'H]ABP (37 mCi/mmol) and unlabeled ABP were
obtained from Chem-Syn Science Laboratories, Lenexa, KS, and from
the Aldrich Chemical Co., Milwaukee, WI, respectively. Both were
purified further by silica gel (grade 923; Davison Chem. Co., Baltimore,
MD) column chromatography using benzene:ethyl acetate (9:1) as the
eluent and by subsequent crystallization from n-hexane. Their purity
was judged to be >99% as determined by HPLC (see below). [2,2'-3H]
N-OH-ABP (92 mCi/mmol) and unlabeled N-OH-ABP were freshly
prepared by ammonium bisulfide reduction (20) of [3H]4-nitrobiphenyl
(Chem-Syn) and 4-nitrobiphenyl (Aldrich), respectively. Their purity
was judged to be >95% as determined by HPLC. [2,2'-3H]N-OH-ABP
A'-glucuronide (107 mCi/mmol) was prepared biosynthetically from
[3H]N-OH-ABP (Chem-Syn) and characterized as described previously
(4), except that solutions were extracted with diethyl ether (and de
gassed) just prior to animal treatment to remove any aglycone formed.
Animal Studies. Female beagles (7-11 kg) were obtained from Mar
shall Farms USA, Inc., North Rose, NY, and were housed in dog
metabolism cages (Unifab Corp., Kalamazoo, MI) fitted with a flowthrough pH meter that simultaneously recorded urine pH and time of
urination (21). Urine samples were obtained either by normal micturi
tion or from an indwelling Foley urethral catheter. When required for
implacement of urinary catheters and multiple blood collections, dogs
were sedated with a short-acting anesthetic (10 mg/kg ketamine and
0.5 mg/kg diazepam i.v.) at 15 min prior to carcinogen dosing and were
placed in a restraining harness (Alice King Chatham Medical Arts, Los
Angeles, CA). No effects on carcinogen metabolism, pharmacokinetics,
or adduct formation were observed as a consequence of sedation.
For p.o. dosing, [3H]ABP (3.7 mCi/mmol) was administered p.o. at
5 mg/kg in a gelatin capsule. For i.v. injections to determine the
volumes of distribution, [3H]ABP (37 mCi/mmol), [3H]N-OH-ABP (92
mCi/mmol),and [3HjN-OH-ABP/V-glucuronide(107 mCi/mmol) were
AFTER ABP
dissolved in 1.6 ml of 60% dimethyl sulfoxide-water and administered
at 0.04, 0.04, and 0.01 mg/kg, respectively. Intraurethral doses were
instilled by catheter in argon-saturated, phosphate-buffered (pH 6)
saline-0.5 IHMEDTA (5 ml/kg) containing [3H]ABP (37 mCi/mmol)
or [3H]N-OH-ABP (92 mCi/mmol) at a delivered dose of 0.4 mg/kg.
After 1 h, the bladder lumen was drained with the catheter and washed
with warm saline (1 ml/kg), and recovery of the administered dose was
determined by scintillation counting to be nearly 40% of the instilled
dose for both ABP and N-OH-ABP. Thus, the actual dose absorbed
across the urothelium was calculated to be 0.25 mg/kg.
Heparinized whole blood, plasma, washed erythrocytes, and urine
were collected and analyzed immediately at the specified intervals for
total radioactivity by liquid scintillation counting (7). Urine and erythrocyte samples were then quickly frozen over dry ice and stored at
—20°C.
ABP-hemoglobin adduct levels in erythrocytes were determined
by gas chromatography/mass spectrometry (6) and analyses of each
urine sample for ABP, N-OH-ABP, and their A'-glucuronides were
carried out by HPLC, diode-array spectrophotometry, and flow-through
scintillation counting (2, 22, 23) as detailed previously.
Upon terminal sacrifice at 24 h, DNA was isolated from the urothe
lium and liver of dogs treated p.o. with ABP or intraurethrally with
ABP or N-OH-ABP; and ABP-DNA adducts were estimated by a
specific immunochemical method that measures the major ABP-base
adduct, jV-(guan-8-yl)-ABP (24).
Pharmacokinetic Modeling. A multicompartmental pharmacokinetic
model was constructed from these data and from data on the rates of
hydrolysis of N-OH-ABP A'-glucuronide at pH 5, 6, 7, and 8.3 The
results obtained from i.v. administration of ABP, N-OH-ABP, and NOH-ABP A'-glucuronide were used to simplify the modeling procedure
by subdividing this multicompartment into several of its components.
PCNONLIN, a pharmacokinetics program for personal computers
(Version 3.0; SCI Software, Lexington, KY), was utilized to derive
initial estimates of the rate constants from a two-term exponential
equation. These estimates were then used in a two-compartment model
on an analog/digital hybrid computer (PACER 500; Electronics Asso
ciates, Inc., West Long Branch, NJ) to obtain least square iterative fits
of each i.v. data set. These results were subsequently used to set and
hold constant these portions of the multicompartmental model in order
to allow simulation of the remaining components that were applicable
to the p.o. and intraurethral routes of administration. After the individ
ual rate constants were determined for the complete model, bladder
exposure to free N-OH-ABP in the lumen was simulated as a function
of pH (5.0, 6.0, 7.0, and 8.0) and voiding interval (1, 2, 3, 4, 6, 8, and
10 h). These values encompass the range of observations seen in the
dogs that were studied.
RESULTS
AND DISCUSSION
Effect of Voiding Interval and Urine pH on the Pharmacodynamics of ABP and Its A/-Hydroxy Metabolites in Dogs. In order
to examine the effect of different voiding intervals on the
absorption, distribution, metabolism, and excretion of ABP, NOH-ABP, and N-OH-ABP A'-glucuronide, dogs were allowed
to exhibit normal voiding patterns or were fitted with a urinary
catheter to allow emptying of the bladder lumen at specified
intervals. In dogs given [3H]ABP p.o. at 5 mg/kg (Fig. 1),
absorption of ABP occurred rapidly and the percentage of the
dose in whole blood and plasma reached a maximum at 2-4 h
after treatment. Plasma levels then decreased rapidly whereas
the percentage of dose in whole blood reached a plateau at 6 h.
This occurred concomitantly with the formation of ABP-he
moglobin adducts, which increased steadily from 2 to 12 h after
treatment until the percentage of dose bound to hemoglobin
approximated the levels of total ABP in whole blood (8-11 %).
Urinary excretion of ABP and its metabolites increased steadily
over this time period and was essentially complete at 24 h after
3 V. Woolen and F. Kadlubar, unpublished studies.
4372
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.
URINARY FREQUENCY AND ADDUCT FORMATION AFTER ABP
60
a.
m
< 20
50
12
16
20
Time (hrs)
-60
12
12
16
Time (hrs)
16
Time (hrs)
Fig. 1. Metabolic profiles of four dogs with different voiding intervals showing the time course distribution of radioactivity in whole blood, plasma, bound
hemoglobin (Hb-ABP), and urine after p.o. administration of 5 mg/kg [3H]ABP. (A) The dog voided normally with urine samples taken at 10, 13 and 24 h. (B) The
dog voided normally with urine samples taken at 1,2, 4, 12, and 24 h. (C) The dog voided at 2-h intervals by urinary catheter, with samples taken for analysis for up
to 12 h, followed by a 24-h sampling. (D) The dog voided continuously by catheter for up to 12 h with urine samples taken for analysis every 2 h, followed by a 24-h
sampling. Blood samples were taken on a sampling schedule that was independent of urine sample collections.
treatment, ranging from 38 to 59% of the dose. Fecal excretion
accounted for an additional 35-40% of the dose.
In these dogs, the average voiding interval varied widely and
ranged from 8 h (Fig. IA) to 4.8 h (Fig. IB) to 3.4 h (Fig. 1C)
to continuous voiding for the first 12 h after dosing (Fig. 10).
Interestingly, voiding interval had no discernible effect on the
overall metabolic profile of ABP or on ABP-hemoglobin adduct
levels. These data clearly indicate that hemoglobin adduct for
mation, which arises from the reaction of hemoglobin with NOH-ABP in the erythrocyte (25, 26), occurs independently of
micturition and appears to coincide with initial metabolism of
ABP in the liver and the apparent release of free N-OH-ABP
into the circulation.
To examine this possibility further, each urine collection was
analyzed by HPLC (cf. " Materials and Methods") to determine
the relative amounts of ABP, N-OH-ABP, their 7V-glucuronides, and other major metabolites. Cumulatively, the predom
inant urinary metabolites of ABP in the dog were the sulfate
and glucuronide conjugates of 3-hydroxy-ABP and 4'-hydroxyABP (27), with ABP and N-OH-ABP (and their 7V-glucuronides) accounting for only about 2 and 1% of the dose, respec
tively. In contrast to the widely held hypothesis that the Nglucuronide is the major transport form for yV-hydroxy arylamines (4), HPLC analyses of urine collected continuously by
catheter for the first 12 h indicated that 68-89% of the N-OHABP entering the bladder lumen was in the unconjugated form.
Moreover, the relative proportion of the unconjugated N-OHABP vs. its 7V-glucuronide metabolite was inversely proportional
so
•Ë
70
60
x
O
50
'c _
O
30
20
Continuous
12 hrs
3456
Average Voiding Interval (hrs)
Fig. 2. Effect of voiding interval on the proportion of urinary N-OH-ABP
excreted as the unconjugated metabolite. Average voiding interval for each dog
was determined as the number of voided samples obtained by micturition or by
urethral catheter as described in Fig. 1 for Dogs A-D.
to the voiding interval (Fig. 2), ranging from 22 to 73% uncon
jugated N-OH-ABP, as the average daily voiding interval was
decreased from 8 h to continuous voiding for the first 12 h after
dosing. For each dog, 70-90% of the total amounts of urinary
N-OH-ABP were excreted during the first 10-12 h after dosing.
These data indicate that the majority of N-OH-ABP entering
4373
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.
URINARY FREQUENCY AND ADDUCT FORMATION AFTER ABP
lower amounts of this conjugate indicate that total bladder
exposure to N-OH-ABP should be much more dependent on
voiding interval than on urine pH. In these studies, urine pH
was found to vary markedly from sample to sample (pH 5-8).
Moreover, it did not correlate with any metabolic parameter
and appeared to become more alkaline with the greater degree
of physical activity exhibited by the animal during the
experiment.
For comparison to the p.o. route of administration and for
purposes of pharmacokinetic modeling (see below), metabolic
profiles were also determined for ABP, N-OH-ABP, and NOH-ABP /V-glucuronide given i.v. at 0.04, 0.04, and 0.01 mg/
kg, respectively (Fig. 3). After injection, blood levels of these
compounds rapidly decreased and their volumes of distribution
were calculated to be 0.26, 0.16, and 0.19 liter/kg, respectively.
Although the percentage of dose in urine at 24 h for each
compound was similar to that obtained after p.o. dosing of
ABP, the kinetics and extent of ABP-hemoglobin adduci for
mation differed appreciably. As might be expected, i.v. injection
of N-OH-ABP resulted in rapid hemoglobin adduci formation;
ABP given i.v. formed adducts similarly to the p.o. dose, while
N-OH-ABP jV-glucuronide showed the slowest rate and extent
of ABP-hemoglobin adduci formation, as well as a more rapid
excretion inlo urine.
In order to approximate the rate of absorption of urinary
ABP and N-OH-ABP across the urothelium after p.o. dosing
with ABP, dogs were given both compounds by intraurelhral
calheler at 0.40 mg/kg. After 1 h, the bladder luminal contenls
were removed and the absorbed dose was determined lo be 0.25
mg/kg (cf. "Materials and Methods"), which corresponded to
50
about 60% of Ihe administered dose. This rale of reabsorplion
12
16
20
24
Time (hrs)
Fig. 3. Metabolic profiles of dogs showing the time course distribution of
radioactivity in whole blood, plasma, bound hemoglobin (Hb-ABP), and urine
after i.v. injection of ABP (A), N-OH-ABP (B). or N-OH-ABP N-glucuronide
(C).
Ihe bladder lumen is unconjugaled and ihus immedialely avail
able for reabsorption and for covalent binding lo urolhelial
DNA. On Ihe olher hand, the tolal amounl of N-OH-ABP that
can be reabsorbed (»1%of the dose) is insufficient to account
for Ihe high levels of ABP-hemoglobin adducts that are formed
(~ 10% of the dose). Therefore, il would appear lhal aboul 11%
of Ihe dose of ABP indeed enters the circulation from the liver
(and perhaps other tissues) as free N-OH-ABP, of which Ihe
majority (~10% of the dose) is absorbed inlo Ihe erylhrocyte
and becomes bound to hemoglobin, with only about 0.7 and
0.3% of the dose being filtered by the kidney into the bladder
lumen as free N-OH-ABP and its /V-glucuronide, respectively.
Although the A'-glucuronide may hydrolyze at acidic pH in the
bladder lumen and release more N-OH-ABP, Ihe relalively
o
12
16
Time (hrs)
Fig. 4. Metabolic profiles of dogs showing the time course distribution of
radioactivity in whole blood, plasma, bound hemoglobin (Hb-ABP), and urine
after intraurethral instillation of ABP (A) and N-OH-ABP (B).
4374
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.
URINARY FREQUENCY AND ADDUCI
ABP
(10%)
BLOOD
(+ TISSUES)
N-OH-ABP
N-OH-ABP
ABP
'ABP-M\LUMEN
N-uiucuromaeIÕT~ABP-M
^^
i^
BLADDER'î
ABPURINE
\^
N-OH-ABP«f
N-OH-ABP
xABP(2%)\
AFTER ABP
prior to each urination. The amounts in excreted urine also
increased in a stepwise manner, while the total exposure of the
urothelium to N-OH-ABP (the integral of N-OH-ABP levels
in the lumen) increased as a curvilinear function for the first 8
h. These simulations were then conducted for each of the pH
values and voiding intervals indicated above and values for the
integral of N-OH-ABP were tabulated.
These relative exposure values were subsequently plotted as
a function of voiding interval for each pH simulated (Fig. 7).
The results show that exposure to the carcinogenic N-OH-ABP
metabolite is directly proportional to the length of time the
urine resides in the bladder lumen. Decreased urine pH did
increase the amounts of N-OH-ABP formed from its /V-glucu
ronide, but the increase in exposure was only about 10% and
clearly not as profound as that derived from unconjugated NOH-ABP as a consequence of prolonged urine retention.
Hb-ABP
ABSORPTION
FORMATION
£-
N-OH-ABPABP-M
>N-Glucuronide4*'
N.Glucuronide(-50%)
N-OH-ABP
(0.7%)
,n QO/I
Feces
(-35%)
Fig. 5. Multicompartmental pharmacokinetic model for the distribution of
ABP and its metabolites in the dog. A'umftprs in parentheses, percentage of dose
excreted in the urine or bound to hemoglobin (Hb) after a 5-mg/kg p.o. dose of
ABP. The values for N-OH-ABP and N-OH-ABP /V-glucuronide were selected
from a dog with the shortest average voiding interval (12-h continuous). ABP-M
is used to designate all metabolites of ABP other than N-OH-ABP or its Nglucuronide.
was comparable to that reported previously (7) for reabsorption
of N-OH-ABP across the rat urinary bladder. The metabolic
profiles (Fig. 4) of these dogs indicated a rapid absorption of
the intraurethral dose, followed by a steady decrease in whole
blood and plasma levels, with increasing levels of ABP-hemoglobin adducts and cumulative excretion into the urine. As
observed with i.v. dosing, intraurethral instillation of N-OHABP resulted in an appreciably higher rate and extent of he
moglobin adduct formation.
For each route of administration of ABP, the covalent bind
ing of circulating N-OH-ABP to hemoglobin was about 10-fold
higher than the amount of N-OH-ABP entering the urinary
bladder lumen. Thus, ABP-hemoglobin adduct formation can
also be regarded as a major detoxification pathway.
Pharmacokinetic Model to Predict Exposure of the Urinary
Bladder to N-OH-ABP as a Function of Voiding Interval and
Urine pH. From these data, a multicompartmental pharmaco
kinetic model was designed to assess the role of the two phys
iological variables, micturition and urine pH, in determining
the exposure of the urothelium to N-OH-ABP, an ultimate
carcinogenic metabolite of ABP. This model (Fig. 5) allows for:
(a) the absorption of a p.o. dose of ABP into blood and body
tissues; (b) the conversion of ABP to N-OH-ABP, N-OH-ABP
/V-glucuronide, and other metabolites, primarily in the liver; (c)
the movement of ABP and its metabolite through the circula
tion, the reaction of N-OH-ABP with hemoglobin, and the
deposition of these metabolites in the urinary bladder lumen;
(d) the acidic hydrolysis of urinary N-OH-ABP /V-glucuronide
to N-OH-ABP and the reabsorption of N-OH-ABP and ABP
across the urothelium; and (e) the removal of ABP and its
metabolites from these compartments by excretion into urine
and feces. Rate constants for these processes are summarized
in Table 1.
The accumulation of N-OH-ABP in the bladder lumen and
consequently its contact with the urothelium was simulated
with this model on an analog/digital hybrid computer using pH
values from 5 to 8 and voiding intervals from 1 to 10 h. A
typical simulation is presented in Fig. 6 for a urine pH of 6.0
and a voiding interval of 4 h. As shown in the simulation, the
levels of total N-OH-ABP in the lumen gradually increased
Table 1 Rate constants associated with the multicompartmental pharmacokinetic
model shown in Fig. 5"
Process
Rate constant
(min)-'
Absorption
ABP. p.o.
ABP. intraurethral
N-OH-ABP, intraurethral
0.060
0.060
0.020
Distribution in blood/tissues
ABP -. ABP-M*
ABP —N-OH-ABP
N-OH-ABP -. ABP-M
N-OH-ABP ^Hb-ABP
N-OH-ABP —N-OH-ABP A'-glucuronide
N-OH-ABP A'-glucuronide —N-OH-ABP
0.030
0.0039
0.020
0.012
0.0067
0.0025
Excretion from blood/tissues to bladder lumen
ABP
ABP-M
N-OH-ABP
N-OH-ABP A'-glucuronide
0.00071
0.011
0.0033
0.00067
Fecal excretion
ABP-M
0.0087
°Elimination from the urinary bladder lumen into the urine was set faster than
any other rate constant and was chosen to completely empty the bladder at the
time of voiding. Rate constants for the hydrolysis of N-OH-ABP A'-glucuronide
to N-OH-ABP were pH dependent in the bladder lumen and were set in accord
ance with the urine pH measured at each micturition.
* ABP-M, ABP converted to other metabolites.
Time (hrs)
Fig. 6. Computer-simulated levels of: (A) total ABP in whole blood (x 2); (B)
total free N-OH-ABP in the bladder lumen (x 200); (C) free N-OH-ABP excreted
in urine (x 30); and (D) the integral of N-OH-ABP concentration or cumulative
exposure in the bladder lumen (x 400). The simulation was done for a urine pH
of 6.0 and a voiding interval of 4 h. Note that Cunes B and C are stepwise
functions that always decrease and increase, respectively.
4375
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.
URINARY FREQUENCY AND ADDUCT FORMATION
Effect of Voiding Interval on the Formation of ABP-DNA
Adducts in the Liver and Urothelium of Dogs. The predictive
value of this model (cf. Fig. 7) was tested by estimation of the
ABP-DNA adducts formed in the bladders of ABP-treated dogs
(5 mg/kg), whose average daily voiding intervals was varied (cf.
"Materials and Methods") from 8 h to continuous voiding for
the first 12 h after dosing (Fig. 1). As shown in Fig. 8, when
voiding interval was decreased, the level of jV-(guan-8-yl)-ABP
adducts in the urothelium increased over a 17-fold range; while
levels in the liver varied less than 3-fold. The potent reactivity
of urinary N-OH-ABP toward bladder epithelial DNA was
further demonstrated by direct instillation of N-OH-ABP (0.25
mg/kg) into the dog urinary bladder, which gave levels of
10
468
VOIDING INTERVAL (MRS)
Fig. 7. Computer-simulated estimations of the effect of voiding interval and
urine pH on cumulative exposure of the bladder epithelium to urinary N-OHABP. Relative exposure is calculated by assigning a zero value to the integral of
the free N-OH-ABP concentration in the bladder lumen (i.e., cumulative expo
sure) when the voiding interval is zero. The points on the curve are relative
exposure values obtained from computer simulations of cumulative exposure for
voiding intervals of I-10 h and urine pH values of 5. 6, 7, and 8. Urine pH was
held constant for each simulation of exposure to N-OH-ABP; each point repre
sents cumulative exposure values (as shown by Cum D in Fig. 6) obtained in
separate computer simulations.
AFTER ABP
urothelial yV-(guan-8-yl)-ABP adducts that were comparable to
that found in dogs given 20-fold higher p.o. doses of ABP. In
contrast, intraurethral administration of ABP (0.25 mg/kg) did
not give detectable binding levels.
Conclusions. These studies provide strong evidence that ac
cumulation of N-OH-ABP in the urinary bladder is primarily
responsible for ABP-DNA adduci formation in the carcinogentarget tissue. Moreover, the construction of a multicompartmental pharmacokinetic model that accurately simulates blad
der exposure to N-OH-ABP as a function of voiding interval
and urine pH provides an excellent predictive tool for assess
ment of individual differences in susceptibility to arylamineinduced urinary bladder carcinogenesis. Thus, it would be useful
to collect data on both urine pH and voiding intervals in
epidemiological studies in humans, where jV-(guan-8-yl)-ABP
has now been shown to be present as major adduci in the
urolhelium of cigarelle smokers (28).
Moniloring of humans for blood Hb-ABP levels should also
provide a useful estimation of both exposure and metabolic
aclivation. The metabolic activation pathways for ABP in hu
mans and dogs appear to be very similar (23) and previous
sludies in humans have shown that Hb-ABP adduci levels are
a reliable measure of exposure to ABP in cigaretle smoke (24,
29). The pharmacokinelic studies reported herein clearly indi
cate that Hb-ABP adducts are formed from circulating levels
of N-OH-ABP in blood; and receñÃsludies in vivo in rals using
cylochrome P-450IA2 inhibitors (30) provide direct evidence
that Hb-ABP adduct formalion is derived from the hepatic 7Voxidalion of ABP lo N-OH-ABP. Finally, Ihe use of Hb-ABP
adducts as a predictive tool in humans is further supported by
our recent observation lhal smoking-relaled DNA adducls pres
ent in exfoliated urolhelium of differenl individuals are significanlly correlaled wilh measured blood levels of Hb-ABP.4
ACKNOWLEDGMENTS
We express our sincere appreciation to Professor Gerard Mulder of
the University of Leiden for his helpful advice and stimulating discus
sions throughout the course of this study.
REFERENCES
N-OH-ABPIntraurethral,
ABP or
mg/kgB
0.25
[-1-B
0.8
0.6
0.4
0.2
BLADDER
| Continuous:!-'3.4 Hfl'.
UVER/3
4.6 HH'
8 HI»
LAI1I1I¡DCR
LIVER!.•»1210.80.60.40.20•
A hi'
. W.OH. ABP
Fig. 8. ABP-DNA adduci levels in the urothelium and liver (x Vj) of six dogs:
as a function of voiding interval after p.o. administration of ABP ( 11: or after
intraurethral instillation of ABP or N-OH-ABP (B). For each determination.
DNA samples were hydrolyzed and analyzed in triplicate by a competitive avidinbiotin-amplified enzyme-linked immunoassay (24). The percentage coefficient of
variation was<10%.
1. Garner, R. C., Martin, C. N., and Clayson, D. B. Carcinogenic aromatic
amines and related compounds. In: C. E. Searle (ed.). Chemical Carcinogens,
Ed. 2, pp. 175-276. ACS Monograph 182. Washington, DC: American
Chemical Society, 1984.
2. Butler, M. A., Iwasaki, M., Guengerich, F. P., and Kadlubar, F. F. Human
cytochrome P-450A (P-450IA2), the phenacetin O-deethylase, is primarily
responsible for the hepatic 3-demethylation of caffeine and A/-oxidation of
carcinogenic arylamines. Proc. Nati. Acad. Sci. USA, 86: 7696-7700, 1989.
3. Kamataki, T., Maeda. K., Yamazoe, Y., Matsuda, N., Matsuda, K., Ishii, K.,
and Kato, R. A high-spin form of cytochrome P-450 highly purified from
polychlorinated biphenyl-treated rats. Mol. Pharmacol., 24: 146-155, 1983.
4. Kadlubar, F. F., Miller, J. A., and Miller, E. C. Hepatic microsomal Nglucuronidation and nucleic acid binding of jV-hydroxy arylamines in relation
to urinary bladder carcinogenesis. Cancer Res., 37: 805-814, 1977.
5. Green, L. C., Skipper, P. L., Turesky, R. J., Bryant, M. S., and Tannenbaum,
S. R. In vivo dosimetry of 4-aminobiphenyl in rats via a cysteine adduct in
hemoglobin. Cancer Res., 44:4254-4259, 1984.
6. Bryant, M. S., Skipper, P. L., Tannenbaum, S. R., and Madure, M. Hemo
globin adducts of 4-aminobiphenyl in smokers and nonsmokers. Cancer Res.,
4 7:602-608, 1987.
7. Oglesby, L. A., Flammang, T. J., TullÃ-s,D. L., and Kadlubar, F. F. Rapid
absorption, distribution, and excretion of carcinogenic A'-hydroxy-arylamines
after direct urethral instillation into the rat urinary bladder. Carcinogenesis
(Lond.). 2: 15-20, 1981.
* G. Talaska. M. Schamer, P. Skipper, S. Tannenbaum, N. Caporaso, L.
Unruh. F. Kadlubar, H. Bartsch, C. Malavielle, and P. Vineis. Detection of
carcinogen-DNA adducts in exfoliated urothelial cells of cigarette smokers:
association with smoking, hemoglobin adducts. and urinary mutagenicity, sub
mitted for publication.
4376
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.
URINARY FREQUENCY AND ADDUCT FORMATION AFTER ABP
8. Beland, F. A., Beranek, D. T., Dooley, K. L., Heflich, R. H., and Kadlubar,
F. F. Arylamine-DNA adducts in vitro and in vivo: their role in bacterial
mutagenesis and urinary bladder carcinogenesis. Environ. Health Perspect.,
49:125-134, 1983.
9. Kadlubar, F. F., and Beland, F. A. Chemical properties of ultimate carcino
genic metabolites of arylamines and arylamides. In: R. G. Harvey (ed.),
Polycyclic Hydrocarbons and Carcinogenesis, ACS Symposium Series 283,
pp. 341-370. Washington, DC: American Chemical Society, 1985.
10. Young, J. F., and Kadlubar, F. F. A pharmacokinetic model to predict
exposure of the bladder epithelium to urinary A'-hydroxyarylamine carcino
gens as a function of urine pH, voiding interval and résorption.Drug Metab.
Dispos., ÃŽO.-641-644,1982.
11. Wise, R. W., Zenser, T. V., Kadlubar, F. F., and Davis, B. B. Metabolic
activation of carcinogenic aromatic amines by dog bladder and kidney prostaglandin H synthase. Cancer Res., 44: 1893-1897, 1984.
12. Flammang, T. J., Yamazoe, Y., Benson, R. W., Roberts, D. W., Potter, D.
W., Chu, D. Z. J., Lang, N. P., and Kadlubar, F. F. Arachidonic aciddependent peroxidative activation of carcinogenic arylamines by extrahepatic
human tissue microsomes. Cancer Res., 49: 1977-1982, 1989.
13. Kadlubar, F. F., Frederick, C. B., Weis, C. C., and Zenser, T. V. Prostaglandin
endoperoxide synthetase-mediated metabolism of carcinogenic aromatic
amines and their binding to DNA and protein. Biochem. Biophys. Res.
Commun., 70«:253-258, 1982.
14. Kadlubar, F. F., Yamazoe, Y., Lang, N. P., Chu, D. Z. J., and Beland, F. A.
Carcinogen-DNA adduct formation as a predictor of metabolic activation
pathways and reactive intermediates in benzidine carcinogenesis. In: J. J.
Kocsis, D. J. Jollow, C. M. Witmer, J. O. Nelson, and R. Snyder (eds.),
Biologically Reactive Intermediates III, pp. 537-549. New York: Plenum
Publishing Corp., 1986.
15. Krauss. R. K., and Eling, T. E. Formation of unique arylamine:DNA adducts
from 2-aminofluorene activated by prostaglandin H synthase. Cancer Res.,
45:1680-1686, 1985.
16. Yamazoe, Y., Miller, D. W., Weis, C. C., Dooley, K. L., Zenser, T. V.,
Beland, F. A., and Kadlubar, F. F. DNA adducts formed by ring-oxidation
of the carcinogen 2-naphthylamine with prostaglandin H synthase in vitro
and in the dog urothelium ¡nvivo. Carcinogenesis (I.ond.l. 6: 1379-1387,
1985.
17. Kadlubar, F. F., Butler, M. A., Hayes, B. E., Beland, F. A., and Guengerich,
F. P. Role of microsomal cytochromes P-450 and prostaglandin H synthase
in 4-aminobiphenyl-DNA adduct formation. In: J. Miners, D. J. Birken, R.
Drew, and M. McManus (eds.), Microsomes and Drug Oxidations, pp. 380-
389. London: Taylor and Francis, 1988.
18. Beland, F. A., and Kadlubar, F. F. Formation and persistence of arylamine
DNA adducts in vivo. Environ. Health Perspect., 62: 19-30, 1985.
19. Yamazoe, Y., Zenser, T. V., Miller, D. W., and Kadlubar, F. F. Mechanism
of formation and structural characterization of DNA adducts derived from
peroxidative activation of benzidine. Carcinogenesis (Lond.), 9: 1635-1641,
1988.
20. Willstätter, R., and Kubli, H. Überdie Reduktion von Nitroverbindungen
nach der Methode von Zinin. Chem. Ber., 41:.\936-1940, 1908.
21. West, R. W., Stanley, J. W., and Newport, G. D. Single-mouse urine
collection and pH monitoring system. Lab. Anim. Sci., 28: 343-345, 1978.
22. Frederick, C. B., Mays, J. B., and Kadlubar, F. F. A Chromatographie
technique for the analysis of oxidized metabolites: application to carcinogenic
W-hydroxyarylamines in urine. Anal. Biochem., 118: 120-125, 1981.
23. Butler, M. A., Guengerich, F. P., and Kadlubar, F. F. Metabolic oxidation
of the carcinogens 4-aminobiphenyl and 4,4'-methylenebis(2-chloroaniline)
by human hepatic microsomes and by purified rat hepatic cytochrome P-450
monooxygenases. Cancer Res., 49: 25-31, 1989.
24. Roberts, D. W., Benson, R. W., Groopman, J. D., Flammang, T. J., Nagle,
W. A., Moss, A. J., and Kadlubar, F. F. Immunochemical quantitation of
DNA adducts derived from the human bladder carcinogen 4-aminobiphenyl.
Cancer Res., 48: 6336-6342, 1988.
25. Ringe, D., Turesky, R. J., Skipper, P. L., and Tannenbaum, S. R. Structure
of the single stable hemoglobin adduct formed by 4-aminobiphenyl in vivo.
Chem. Res. Toxicol., /: 22-24, 1988.
26. Haugen, D. A. Charge-shift strategy for isolation of hemoglobin-carcinogen
adducts formed at the 093 cysteine sulfhydryl groups. Chem. Res. Toxicol.,
2: 379-385, 1989.
27. Gorrod, J. W., and Manson, D. The metabolism of aromatic amines. Xeno
biotica, 16: 933-955, 1986.
28. Talaska, G. Al-Juburi, A. Z. S. S., and Kadlubar, F. F. Smoking-related
carcinogen-DNA adducts in biopsy samples of human urinary bladder: iden
tification of JV-(deoxyguanosin-8-yl)-4-aminobiphenyl as a major adduct.
Proc. Nati. Acad. Sci. USA, SS: 5350-5354, 1991.
29. Coghlin, J., Gann, P. H., Hammond, S. K., Skipper, P. L., Taghizadeh, K.,
Paul, M., and Tannenbaum, S. R. 4-Aminobiphenyl hemoglobin adducts in
fetuses exposed to tobacco smoke carcinogen in utero. J. Nati. Cancer Inst.,
«3:275-280, 1991.
30. Hammons, G. J., Dooley, K. L., and Kadlubar, F. F. 4-Aminobiphenylhemoglobin adduct formation as an index of in vivo A'-oxidation by hepatic
cytochrome P-450IA2. Chem. Res. Toxicol., 4: 144-147, 1991.
4377
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.
Frequency of Urination and Its Effects on Metabolism,
Pharmacokinetics, Blood Hemoglobin Adduct Formation, and
Liver and Urinary Bladder DNA Adduct Levels in Beagle Dogs
Given the Carcinogen 4-Aminobiphenyl
Fred F. Kadlubar, Kenneth L. Dooley, Candee H. Teitel, et al.
Cancer Res 1991;51:4371-4377.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/51/16/4371
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1991 American Association for Cancer Research.