![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vol. 305, Issue 2, 755-764, May 2003
Department of Psychology and Program in Neuroscience, University of Illinois at Urbana-Champaign, Champaign, Illinois (S.G.W., C.D.B., W.R.L., M.S.O., E.J.C., J.S.M.); Department of Physiology and Pharmacology, School of Medical Sciences, University of New South Wales, Sydney, Australia (B.E.G.); and Department of Psychology, McGill University, Montreal, Canada (J.S.M.)
 |
Abstract |
|---|
 Top
 Abstract
 Introduction
Materials and Methods
Results
Discussion
Conclusions
References
|
|---|
Chromosomal loci containing genes affecting antinociceptive sensitivity to morphine have been identified, but virtually nothing is known about the genetic mediation of sensitivity to over-the-counter analgesics. Such knowledge would be of great clinical interest, as prodigious interindividual variability has been noted in the efficacy of these ubiquitously used drugs. In the present study, we assessed heritability and genetic correlations among three over-the-counter analgesics in mice of 12 inbred mouse strains on the 0.9% acetic acid (i.p.) writhing test. Analgesics included the centrally acting analgesic, acetaminophen (150 mg/kg, s.c.), and the nonsteroidal anti-inflammatory drugs (NSAIDs), indomethacin (40 mg/kg, s.c.) and lysine-acetylsalicylic acid (800 mg/kg, s.c.). Significant strain differences in sensitivity to each of the drugs were observed, with narrow-sense heritability estimates ranging from 23 to 45%. Similar strains were sensitive and resistant, respectively, to the two NSAIDs (rs = 0.64). In contrast, a completely different pattern of sensitivities was observed for acetaminophen, implying genetic dissociation (rs = 0.29 and 0.02) compared with the NSAIDs. Additional experiments were performed on two strains, C57BL/6 and DBA/2, with extreme sensitivities to acetaminophen. Plasma acetaminophen levels in these strains were not significantly different during the time of antinociception assessment, suggesting the existence of genetic factors affecting acetaminophen pharmacodynamics rather than pharmacokinetics.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
Conclusions
References
|
|---|
Acetaminophen
and nonsteroidal anti-inflammatory drugs (NSAIDs) are among the most
widely used medications in the world by both prescription and over the
counter. The physiological basis of their anti-inflammatory,
antinociceptive/antihyperalgesic, and antipyretic actions has been
debated for some time. The classical explanation that these drugs work
via inhibition of peripheral prostaglandin synthesis has been seriously
challenged by evidence for central actions (see Yaksh et al., 1998
).
The discovery of multiple cyclooxygenase enzymes (see Smith and DeWitt,
1996), i.e., the constitutively expressed cyclooxygenase-1 and the
inducible cyclooxygenase-2, represented a great advance in the
understanding of NSAID action. The recent discovery of
cyclooxygenase-3, a splice variant of the cyclooxygenase-1 gene
(Chandrasekharan et al., 2002), may explain acetaminophen action.
However, it remains unknown whether any cyclooxygenase isoform is
necessary and/or sufficient for the antinociceptive actions of
over-the-counter analgesics. Evidence from transgenic knockout mice
lacking these genes has failed to converge (see Wallace, 1999; Ballou
et al., 2000; Guhring et al., 2002), and thus, a new approach may be of value.
Considerable variability in both the antinociceptive and
anti-inflammatory responses to individual over-the-counter drugs has
been reported (e.g., Huskisson et al., 1976; Scott et al., 1982;
Bellamy, 1985; Day et al., 1988; Walker et al., 1994). This variability
has rendered it all but impossible to rank such drugs in terms of
therapeutic efficacy. In clinical practice, failure to obtain relief
with one drug leads to rotation among the common NSAIDs until a
satisfactory response is achieved or a switch is made to narcotic
analgesics. The cause of this variability is unknown but is likely
unrelated to disease variability, as it can be demonstrated in
controlled experimental situations (e.g., Walker et al., 1994).
Although pharmacokinetic variations among subjects are a favored
explanation of over-the-counter antinociceptive variability, several
studies have failed to demonstrate such a relationship (see Walker,
1995). Other proposed explanations of response variability to
over-the-counter drugs include sex differences (Walker and Carmody,
1998) and the strength of the placebo effect (Amanzio et al., 2001).
Genetic factors have never been seriously considered as an
explanation of variable sensitivity to over-the-counter drugs, although
antinociceptive sensitivity to opioids is known to be strongly affected
by inherited genetic factors (see Mogil, 1999). The chromosomal
locations of genes underlying morphine sensitivity in mice are known
(see Mogil, 1999; Bergeson et al., 2001), and pharmacological evidence
supports the candidacy of the Oprm (µ-opioid receptor) and
Htr1b (serotonin-1B receptor) genes as contributing to
variable morphine antinociception in sensitive (DBA/2J) versus resistant (C57BL/6J) strains (see Mogil, 1999). Very little is known
about genetic factors underlying variable responses to any other
analgesic drug. In a recent study, we demonstrated a surprisingly high
genetic correlation between antinociceptive sensitivity to morphine and
four neurochemically distinct analgesics: the 
-opioid agonist
U50,488, the cannabinoid WIN55,212-2, the nicotinic agonist epibatidine, and the 
2-adrenergic agonist
clonidine (Wilson et al., 2003). In that study, the same 12 strains
were tested for their antinociceptive response to the five drugs, and
similar strains were found to be sensitive and resistant, respectively, to each one. This finding implicates the same gene(s) in the mediation of sensitivity to all of the drugs. In addition, for all drugs, a high
correlation was obtained between initial nociceptive sensitivity (on
the 49°C tail-withdrawal test) and subsequent drug response (Wilson
et al., 2003).
Thus, the purposes of the present study were 2-fold. First, we
wished to investigate whether sensitivity to over-the-counter antinociception is heritable in the mouse, exhibiting inbred strain mean differences. Second, we wished to investigate whether sensitivity to different over-the-counter drugs would correlate genetically, as we
had found previously for other centrally acting drugs (Wilson et al.,
2003), or would show genetic dissociation, which might be predicted
from the clinical reality described above. Acetaminophen, indomethacin,
and lysine-acetylsalicylic acid (aspirin) were chosen for this study,
as they are known to differ greatly in their anti-inflammatory efficacy
(McCormack and Brune, 1991), cyclooxygenase enzyme selectivity (e.g.,
Meade et al., 1993), mode of antagonistic action on cyclooxygenase enzymes (see Smith and DeWitt, 1996), and effects on
prostaglandin-independent signaling mechanisms (see Tegeder et al.,
2001). To these ends, we determined the antinociceptive sensitivity of
11 inbred mouse strains to acetaminophen (150 mg/kg, s.c.),
indomethacin (40 mg/kg, s.c.), and lysine-aspirin (800 mg/kg, s.c.) on
the 0.9% acetic acid abdominal constriction (writhing) test. For each
drug, full dose-response relationships were then examined in
extreme-responding strains. For one drug, acetaminophen, follow-up
experiments were performed to establish the generalizability of the
strain differences and to investigate whether those differences were
due to pharmacodynamic or pharmacokinetic factors.
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
Conclusions
References
|
|---|
Subjects. Male and female breeders of the outbred strains ND4 Swiss-Webster (Hsd:ND4) or CD-1 (Hsd:ICR) were obtained from Harlan (Indianapolis, IN). Male and female breeders of 11 inbred strains (129P3, A, BALB, C3H/He, C57BL/6, C57BL/10, CBA, DBA/2, and RIIIS, all "J" substrains) were obtained from The Jackson Laboratory (Bar Harbor, ME). All mice used in the studies were bred in our temperature-controlled (20 ± 2°C) vivarium under a 12/12-h light/dark cycle (lights on at 07:00). Mice were weaned at 18 to 21 days of age, housed with same-sex littermates in groups of two to five, and given ad lib access to Purina mouse chow (Purina, St. Louis, MO) and water. Both sexes were used in these experiments.
All mice were tested in a quiet room just meters away from the vivarium. Every attempt was made to counterbalance strains across testing schedules, but the vagaries of breeding success prevented true counterbalancing. Nonetheless, all mice were handled equivalently, and all testing was performed on at least two strains simultaneously. Also, to control against day-to-day variability exerting an undue influence on these data, we tested saline-treated mice along with drug-treated mice in every session. Each mouse was used only once and given a single dose of one drug. Sample sizes ranged from n = 5 to 24 per dose per strain (depending on strain availability) in all experiments.Drugs. Acetaminophen and indomethacin (Sigma-Aldrich, St. Louis, MO) were dissolved in a solution of physiological saline and 1,2-propanediol in a ratio of 87.5:12.5% and titrated to pH 7 with NaOH. Acetaminophen was injected subcutaneously (10 ml/kg) in doses ranging from 0 to 400 mg/kg and indomethacin in doses ranging from 0 to 160 mg/kg. A pilot study performed in both an outbred (ND4 Swiss-Webster) and an inbred (DBA/2) strain confirmed that this vehicle did not produce antinociception on the tests used (data not shown). We were unable, however, to solubilize or adequately suspend aspirin in any vehicle that did not itself produce antinociception. For this reason, we used water-soluble lysine-aspirin in these experiments. We are unaware of any published difference in the pharmacodynamic mechanisms underlying the antinociceptive effects of lysine-aspirin relative to the reference compound. Lysine-aspirin (Aspegic; Synthelabo, France) was dissolved in normal saline and injected s.c. in doses ranging from 0 to 1600 mg/kg.
Pilot experiments using outbred mice revealed linear and dose-dependent antinociception from each drug (data not shown), with the following half-maximal antinociceptive doses (AD50 values; see below) for each drug: acetaminophen (116 mg/kg), indomethacin (41 mg/kg), and lysine-aspirin (460 mg/kg). We chose 150 mg/kg acetaminophen and 40 mg/kg indomethacin as "probe" doses for the inbred strain survey. For lysine-aspirin, a sex difference was observed, such that the AD50 for males was 386 mg/kg and for females was 1255 mg/kg. For this reason, we decided on 800 mg/kg as an appropriate probe dose. It should be noted that this sex difference has been subsequently found to be unreliable, as we were unable to replicate it in ND4 Swiss-Webster mice or any other inbred strain.Nociceptive Assessment.
Mice were assessed for nociceptive
sensitivity using the writhing test as described by Koster et al.
(1959). A similar model has been shown to be useful in correlating the
ED50 values of NSAID antinociception in mice with
those in humans (Pong et al., 1985). Although this assay features
considerable variability (e.g., a non-negligible percentage of
nonresponders), which is particularly problematic for the experiments
described herein, it is the only common nociceptive assay in the mouse
to feature reliable sensitivity to weak over-the-counter analgesics
(see Wilson and Mogil, 2001). Although over-the-counter drugs are
reported to be effective in the tonic phase of the formalin test (see
Yaksh et al., 1998), we did not observe convincing antinociception at
nontoxic doses in pilot studies. We have been able to reduce the
percentage of nonresponding mice in the writhing test by using a higher
acetic acid concentration, 0.9%, than is commonly used in this assay (0.6%).
). Mice were placed on a metal surface
maintained at 50.0 ± 0.2°C (IITC hot plate analgesia meter
model PE34M-HC; IITC, Inc., Woodland Hills, CA). Locomotion of
the mouse on the plate was constrained by 20 cm-high Plexiglas walls to
an area 14 × 14 cm. Latency to respond to the heat stimulus was
measured to the nearest 0.1 s. Mice remained on the plate until
they performed either of two behaviors regarded as indicative of
nociception: hindpaw lick or hindpaw shake/flutter. Subjects were
tested immediately before, and at various time points after, injection
of saline or acetaminophen; they remained in their home cages between
tests. A cutoff latency of 150 s was imposed to prevent the
possibility of tissue damage.
Data Analysis.
In writhing experiments, where within-animal
baselines cannot be obtained, antinociception was quantified by
reference to the saline-treated control group for each strain. Percent
antinociception was calculated as [(saline-treated mean writhes 
drug-treated writhes)/saline-treated mean writhes] × 100, with the
saline-treated mean writhes value being calculated separately for each
strain in each experiment. In hot-plate experiments, percent
antinociception was quantified with respect to the subject's own
baseline as [(postinjection latency baseline
latency)/(150 baseline latency)] × 100. Half-maximal AD50 values, potency ratios, and associated 95%
confidence intervals were calculated from percent antinociception data
by the method of Tallarida and Murray (1981).
level of 0.05 was employed. A datum from one
mouse receiving acetaminophen was removed from the data set after being
identified as a statistical outlier (Studentized residual = 3.7).
Because there are no dominance effects in inbred strains, narrow-sense
heritability (h2) can be estimated
from the between-strain variance (Va)
and the within-strain/error variance
(Ve) using the formula
h2 = Va/(Va + Ve), which is based on the
population intraclass correlation coefficient. The variance components
were estimated in our sample using PROC VARCOMP (version 8.2; SAS
Institute, Inc., Cary, NC). Type I sums of squares estimation (in which
observed mean squares are equated to expected mean squares and solved
for the appropriate variance component) was used, because this is more
robust to potential violation of normality assumptions. A fixed-effect
of dose was considered in each ANOVA. Because strains were chosen
without respect to the traits under investigation, these values are
likely to be reflective of the variability present in the overall mouse population (Hegmann and Possidente, 1981).
Genetic correlations among traits (vehicle-treated sensitivity and
acetaminophen, indomethacin, and aspirin antinociception) were assessed
using correlation coefficients applied to the relevant strain means.
Since the joint distribution of the phenotypes was not normal,
Spearman's correlation (rs) for rank
data was employed. Full dose-response curves were collected in
extreme-responding strains and used to calculate the half-maximal
AD50 for each inbred mouse strain.
Acetaminophen Pharmacokinetics. To assess whether strain differences in antinociception from acetaminophen were due to pharmacokinetic or pharmacodynamic factors, we assessed antinociceptive sensitivity and plasma acetaminophen levels in C57B/6 and DBA/2 mice at 30, 60, or 120 min postinjection. Immediately following behavioral assessment of thermal nociceptive sensitivity at one of the three postinjection time points, mice were decapitated, and trunk blood was collected and placed on ice. The blood was centrifuged, and plasma (about 0.1 ml) was removed and freeze dried.
Plasma concentrations of acetaminophen were determined by a modification of published HPLC procedures (Granados-Soto et al., 1993).
Briefly, 50 µl of 0.5 mg/ml phenacetin (internal standard) was added
to 100 µl of freeze-dried mouse plasma, and the mixture was extracted
with 5 ml of ethyl acetate. The solvent was then evaporated and the
residue redissolved in 50 µl of methanol. Samples (10-µl) were
injected onto a HPLC system, which consisted of a C8 HPLC column (Platinum EPS C8 100A,
5 µM; 150 × 4.6 mm; Alltech Associates Pty. Ltd., Australia)
and eluted with a mixture of 0.05 M sodium acetate buffer (pH 4.0) with
acetonitrile (92.5:7.5) at a constant flow of 1 ml/min. The effluent
from the column was recorded by UV detection at 254 nm using a Shimadzu
UV Detector (SPD10Avp; Shimadz, Kyoto, Japan). Retention times for
acetaminophen and the internal standard were 3.4 and 11.3 min,
respectively. The limit of determination of acetaminophen was 0.5 µg/ml, and the interassay coefficient of variation was 6%.
Concentration-time data were analyzed using nonlinear mixed effects
modeling implemented in P-PHARM version 1.3 (Gomeni et al., 1994). The
population approach examines fixed (e.g., pharmacokinetic model
parameters, such as clearance and volume of distribution) and random
(e.g., interanimal variance of pharmacokinetic parameters and residual
variability) effects (Aarons, 1991).
Preliminary analysis of the pooled plasma concentration-time data
for acetaminophen indicated that a one-compartment pharmacokinetic model best described the disposition of this drug in the mouse. The
parameters of the combined pharmacokinetic model were the apparent
volume of distribution (V/F; where F
is the fraction of the dose that gets absorbed and was assumed to be
unity), first-order absorption rate constant
(Ka), and apparent clearance (CL).
Random effects are considered to consist of interindividual variability
in each pharmacokinetic parameter, with the remaining variability being
termed the residual or unexplained variability within animals (Gomeni
et al., 1994). In this study, the interanimal variability in
pharmacokinetic parameters was best described using a normal
distribution, and the residual variability, which encompasses measurement error and model misspecification, was assumed to be constant across the study population.
P-PHARM generates the population mean of pharmacokinetic parameters and
an estimate of the interanimal variability in this parameter (expressed
as a percent coefficient of variation). Posterior Bayesian parameter
estimates for each animal are also generated. This allows the
concentration-time profile in plasma to be determined despite only one
sample being available in each animal.
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
Conclusions
References
|
|---|
Strain Differences in Baseline Sensitivity on the Writhing
Test.
As expected, inbred strain differences were observed in the
sensitivity to 0.9% acetic acid in vehicle-treated mice. In all three
experiments, there was a significant main effect of strain (acetaminophen, p < 0.05; indomethacin,
p < 0.001; and lysine-aspirin, p < 0.001). Estimates of the heritability of writhing test sensitivity calculated from these three data sets range from
h2 = 0.22 to 0.61. Although the three
experiments were performed separately and by two different
experimenters, significant genetic correlations were observed between
responses to 0.9% acetic acid in all cases
(rs = 0.69-0.74, p < 0.001). Also, significant genetic correlations
(rs = 0.73-0.90, p < 0.001) were observed between vehicle-treated strain means in this
experiment and strain means of mice given no s.c. injection and
receiving 0.6% acetic acid in a previous study (Mogil et al., 1999),
suggesting that neither the vehicle injection nor the increased acetic
acid concentration markedly affected responses.
Strain Differences in Sensitivity to Acetaminophen.
Strain means of mice treated with vehicle and 150 mg/kg acetaminophen
are shown in Fig. 1A; percent
antinociception scores calculated from these data are shown in Fig. 1B.
ANOVA revealed a significant main effect of strain on acetaminophen
antinociception (F10,100 = 3.02, p < 0.005), corresponding to a heritability estimate of h2 = 0.24. The main effect of sex
and sex × strain interaction were not significant.
|
2.2
to 23%. Thus, to confirm the reliability of these findings, we chose
one strain from each category, C57BL/6 (responder) and DBA/2 (weak responder), and compiled full dose-response curves (0-800 mg/kg). As
shown in Fig. 1C, the strain difference persists over a wide range of
acetaminophen doses, with AD50 values differing
in these strains by approximately 5-fold.
Strain Differences in Sensitivity to Indomethacin.
Strain
means of mice treated with vehicle and 40 mg/kg indomethacin are shown
in Fig. 2A; percent antinociception
scores calculated from these data are shown in Fig. 2B. ANOVA revealed
a significant main effect of strain on acetaminophen antinociception
(F10,62 = 1.99, p = 0.05), corresponding to a heritability estimate of h2 = 0.23. The main effect of sex was
not significant, but the sex × strain interaction was highly
significant (F10,62 = 3.02, p < 0.005). This interaction manifested itself as
significantly greater indomethacin antinociception in males of two
strains relative to females (129P3 and AKR) and significantly greater
antinociception in females of two strains relative to males (BALB/c and
RIIIS). However, caution should be applied, since sex-specific sample sizes were too low to engender confidence in these findings. Overall conclusions regarding genetic correlations were unchanged when male and
female indomethacin data were considered separately.
|
Strain Differences in Sensitivity to Lysine-Aspirin.
Strain
means of mice treated with vehicle and 800 mg/kg lysine-aspirin are
shown in Fig. 3A; percent antinociception
scores calculated from these data are shown in Fig. 3B. ANOVA revealed a significant main effect of strain on lysine-aspirin antinociception (F10,58 = 5.73, p < 0.001), corresponding to a heritability estimate of
h2 = 0.45. The main effect of sex and
sex × strain interaction were not significant, although the main
effect of sex approached significance (p = 0.11; males,
59.3 ± 5.7% and females, 50.1 ± 5.2%).
|
Genetic Correlations.
There appeared to be essentially no
correlation between strain means of vehicle-treated mice (i.e.,
baseline sensitivity on the 0.9% writhing test) and antinociceptive
sensitivity to acetaminophen, indomethacin, or lysine-aspirin
(rs = 0.12, 0.23, and 0.19, respectively). This lack of genetic correlation was also observed when
experiment-specific vehicle means were used instead of combined vehicle
means (data not shown).
0.02 and 0.29, respectively),
confirming the impression that different strains were sensitive to
acetaminophen than to the two NSAIDs. The same conclusion was arrived
at when Pearson correlation coefficients were applied (data not shown).
|
), with Spearman correlation
coefficients ranging from rs = 0.34
to 0.50 and averaging rs = 0.14.
Sensitivity of C57BL/6 and DBA/2 Mice to Acetaminophen on the
50°C Hot-Plate Test.
To establish the generalizability of these
findings to a nociceptive assay other than the writhing test, we tested
C57BL/6 and DBA/2 mice for their sensitivity to acetaminophen
antinociception on the 50°C hot-plate test. As shown in Fig.
5A, a strain difference in the same
direction as in the writhing test was observed. Calculation of precise
AD50 values was complicated by incomplete
efficacy in both strains, especially DBA/2, in which 50%
antinociception was never achieved at any dose. Doses higher than 800 mg/kg could not be tested because of lethal toxicity.
|
Acetaminophen Pharmacokinetics in C57BL/6 and DBA/2
Mice.
The mean plasma acetaminophen concentration over time
following a 400 mg/kg (s.c.) acetaminophen injection in the same two mouse strains is shown in Fig. 5B. The temporal profiles of
acetaminophen concentrations in C57BL/6 and DBA/2 are very similar,
despite the fact that C57BL/6 mice display robust antinociception on
the hot-plate test at this dose and DBA/2 mice display no
antinociception whatsoever (see Fig. 5A). This result was confirmed by
the pharmacokinetic modeling analysis, which indicated that the
clearance, volume of distribution, and absorption rate constant were
not different, with any apparent differences being small (see Table
1).
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
Conclusions
References
|
|---|
To our knowledge, this study is the first to examine the
genetic mediation of over-the-counter antinociceptive efficacy,
although a large literature exists concerning genetic factors
contributing to the toxicity of these compounds (see Giri, 1993). As
expected based on our previous experience with multiple analgesics
(Wilson et al., 2003), all three compounds tested here displayed
mild-to-moderate heritability. This work represents the first step
toward identification of genes whose inheritance may predict drug
efficacy. Information about the pharmacogenetics of this drug class may
lead to more efficient and successful management of a number of pain
conditions. These findings may also have more immediate application to
the management of postoperative pain in laboratory animals (see Liles and Flecknell, 1992).
The efficacy of over-the-counter analgesics on the mouse writhing test
is well known (see Taber, 1974). Given the variable stimulus intensity
of different nociceptive assays and the presently demonstrated strain
differences in sensitivity, it is difficult to comment on the relation
between antinociceptive magnitudes observed herein and those in the
literature. We do note, however, that the ratio of
acetaminophen/indomethacin/lysine-aspirin AD50 values in ND4 Swiss-Webster mice, for which complete dose-response curves were compiled for all three drugs, was 116:41:460 mg/kg. This
ratio is comparable with similar ratios collected previously in mice
(Pong et al., 1985; Hunskaar and Hole, 1987). We do not believe that
our current demonstration of acetaminophen antinociception on the
hot-plate test should be interpreted as contradicting the general
notion that over-the-counter drugs are ineffective on acute, thermal
assays of nociception. In our hands, only a sensitive strain displayed
robust acetaminophen antinociception against a weak thermal stimulus
(50°C), at a very high dose (400 mg/kg), and with submaximal efficacy
(see Fig. 5).
Genetic Correlations among Over-the-Counter Analgesics.
We
observed presently a moderate-to-large genetic correlation between
sensitivity to lysine-aspirin and indomethacin
(rs = 0.64) but no correlation between
either of these drugs and acetaminophen (rs = 0.02 and 0.29, respectively).
Since genetic correlation implies overlapping physiological mediation
(see Mogil, 2000), it is worth a brief mention of the known
commonalities and dissociations among these drugs. Acetaminophen has
always been the outlier among over-the-counter drugs, displaying very
weak anti-inflammatory effects (at least peripherally) (McCormack and
Brune, 1991; but see Honore et al., 1995). It has therefore been argued
that acetaminophen, unlike NSAIDs, may work in the central nervous
system (e.g., Piletta et al., 1991). However, evidence for central
actions of NSAIDs too is now overwhelming (see Yaksh et al., 1998),
with dose-dependent suppression of pain behavior in animal models
produced by intrathecal or intracerebroventricular injections of a
number of NSAIDs at doses far lower than those required for peripheral
effects. There exist many published dissociations between aspirin and
indomethacin action as well. Although aspirin inhibits inducible nitric
oxide synthase, transcription factor NF-B, and Erk protein kinase in murine cell lines (independent of cyclooxygenase-2 inhibition), indomethacin (and acetaminophen) do not (see Tegeder et al., 2001). Aspirin can block neurogenic inflammation when injected centrally, whereas indomethacin and steroid anti-inflammatory drugs cannot (Catania et al., 1991). Even the mode of cyclooxygenase antagonism differs between aspirin, a noncompetitive antagonist producing irreversible acetylation of the cyclooxygenase site, and indomethacin, which causes a conformational change (see Smith and DeWitt, 1996).
). This
possibility is being investigated at the present time using
quantitative trait locus analysis (see Mogil, 1999), but our recent
experience cautions against the assumption that antinociceptive
variation need be explained by allelic variation at the gene coding for
the molecular binding site of the drug. As mentioned above,
surprisingly high genetic correlations were observed for
antinociceptive sensitivity to five neurochemically distinct compounds
(Wilson et al., 2003). Such a finding all but eliminates the
possibility that genetic variability in the level or functionality of
the receptor of each drug was responsible for variability in
antinociceptive response. Indeed, variability in antinociceptive
efficacy may derive from allelic variation at any gene producing a
protein involved in 1) the neural circuit mediating drug
antinociception, from binding site to effector site, 2) signal
transduction mechanisms within any neuron in that circuit, and/or 3)
pharmacokinetic mechanisms impacting drug concentration at relevant
binding sites. With regard to the latter, it appears that at least for
acetaminophen, the relevant genes are playing a pharmacodynamic role,
since strain-dependent hot-plate antinociception was not accompanied by
evidence of strain-dependent pharmacokinetic parameters (see Table 1)
or plasma acetaminophen concentration (see Fig. 5).
Genetic Correlations between Writhing Test Sensitivity and
Over-the-Counter Drug Antinociception.
Our previous investigation
into genetic mediation of antinociception revealed, in addition to
genetic correlation among disparate analgesics, that antinociceptive
sensitivity was correlated with initial nociceptive sensitivity (Wilson
et al., 2003). This phenomenon was not observed presently, with genetic
correlations between writhing test sensitivity and acetaminophen,
indomethacin, and lysine-aspirin antinociception of
rs = 0.12, 0.23, and 0.19, respectively. It is difficult to speculate on whether the contradictory findings are secondary to the nociceptive assays being used
(tail-withdrawal and formalin tests versus writhing test) or the
different drug classes being tested. It is also the case that genetic
correlations between the drugs tested in our previous study (morphine,
U50,488, WIN55,212-2, epibatidine, and clonidine) and those tested
presently were almost all very low (rs = 0.34-0.50; mean rs = 0.12; all corrected for sign). This low correlation may indicate genetic independence of over-the-counter analgesics from those known to activate descending pain-modulatory systems, but the lack of
correlation may also be explained by the different nociceptive assays
used in each case (e.g., Elmer et al., 1997).
| |
Conclusions |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
Conclusions
References
|
|---|
The finding of genotype-dependent sensitivity to over-the-counter
drug antinociception represents the first step in the identification of
genes associated with such variability. This, in turn, may shed new
light on the mechanism of action of these important therapeutics. Given
the known spinal and supraspinal actions of over-the-counter drugs (see
Yaksh et al., 1998) and their myriad cyclooxygenase-independent actions
and interactions with central opioidergic and serotonergic circuitry
(e.g., Bjorkman, 1995), it is far from clear that the antinociceptive
effects of this drug class can be adequately explained by
cyclooxygenase inhibition alone. Thus, the fact that the cyclooxygenase genes are now well known does not preclude the important involvement of
any number of additional genes. For example, an intriguing new
hypothesis (Guhring et al., 2002) is that NSAIDs, by virtue of blocking
cyclooxygenase, may liberate arachadonic acid to be transformed by an
unknown pathway to anandamide, which has been shown to produce
antinociception when released endogenously (Walker et al., 1999).
| |
Acknowledgments |
|---|
We thank Brenda Edwards and Galina Cotton for excellent animal care and Dr. Michael Ossipov for the use of his dose-response data analysis macro. We also thank Dr. Andrew McLachlan for help with the population pharmacokinetic analysis and Dr. Sandra Rodriguez-Zas for consultation regarding heritability calculation.
| |
Footnotes |
|---|
Accepted for publication February 10, 2003.
Received for publication December 9, 2002.
1 S.G.W. and C.D.B. contributed equally to this work.
This work was supported by Public Health Service Grants DA11394 and DE12735 (J.S.M.) and by the Canada Foundation for Innovation and the Canada Research Chairs program. S.G.W. was supported by NRSA Award DA6000.
DOI: 10.1124/jpet.102.047902
Address correspondence to: Dr. Jeffrey S. Mogil, Dept. of Psychology, McGill University, 1205 Dr. Penfield Ave., Montreal, QC H3A 1B1, Canada. E-mail: jeffrey.mogil{at}mcgill.ca
| |
Abbreviations |
|---|
NSAID, nonsteroidal anti-inflammatory drug; aspirin, acetylsalicylic acid; ANOVA, analysis of variance; HPLC, high-performance liquid chromatography; U50,488, (trans)-3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]benzeneacetamide methane-sulfonate hydrate; WIN 55,212-2, (R)-(+)-[2,3-dihydro-5-methyl-3-(4-morpholinylmethyl)pyrrolo[1,2,3-de]-1,4-benzoxazin-6-yl]-1-naphthalenylmethanone.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
Conclusions
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  R. W. Hurley and M. C. B. Adams Sex, Gender, and Pain: An Overview of a Complex Field Anesth. Analg., July 1, 2008; 107(1): 309 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Chanda and J. S. Mogil Sex differences in the effects of amiloride on formalin test nociception in mice Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2006; 291(2): R335 - R342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Ayoub, R. M. Botting, S. Goorha, P. R. Colville-Nash, D. A. Willoughby, and L. R. Ballou Acetaminophen-induced hypothermia in mice is mediated by a prostaglandin endoperoxide synthase 1 gene-derived protein PNAS, July 27, 2004; 101(30): 11165 - 11169. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
, significantly different from the other strain,
p < 0.05.
