The antipyretic effect of paracetamol occurs

independent of transient receptor potential

ankyrin 1–mediated hypothermia and is

associated with prostaglandin inhibition in the

brain Elahe Mirrasekhian, Johan L. Å. Nilsson, Kiseko Shionoya, Anders Blomgren, Peter

M. Zygmunt, David Engblom, Edward D. Högestätt and Anders Blomqvist

The self-archived postprint version of this journal article is available at Linköping

University Institutional Repository (DiVA):

http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-148562

N.B.: When citing this work, cite the original publication. Mirrasekhian, E., Nilsson, J. L. Å., Shionoya, K., Blomgren, A., Zygmunt, P. M., Engblom, D., Högestätt, E. D., Blomqvist, A., (2018), The antipyretic effect of paracetamol occurs independent of transient receptor potential ankyrin 1–mediated hypothermia and is associated with prostaglandin inhibition in the brain, The FASEB Journal. https://doi.org/10.1096/fj.201800272R

Original publication available at: https://doi.org/10.1096/fj.201800272R

Copyright: Federation of American Society of Experimental Biology (FASEB) http://www.faseb.org/

Mirrasekhian, Nilsson et al., p. 1

The antipyretic effect of paracetamol occurs independently of transient receptor

potential ankyrin 1 mediated hypothermia and is associated with prostaglandin

inhibition in the brain

Elahe Mirrasekhian1,*, Johan L. Å. Nilsson2,*, Kiseko Shionoya1, Anders Blomgren2, Peter M.

Zygmunt2, David Engblom1, Edward D. Högestätt2#, Anders Blomqvist1

1Department of Clinical and Experimental Medicine, Linköping University, S-581 85

Linköping, Sweden, and 2Division of Clinical Chemistry and Pharmacology, Department of

Laboratory Medicine, Lund University, S-221 85 Lund, Sweden

*These authors contributed equally to this work

#Deceased

Running title: Paracetamol antipyresis is TRPA1 independent

19 pages, 4 figures, 1 supplementary figure

Correspondence to: Dr. Anders Blomqvist, Division of Neurobiology, Department of Clinical

and Experimental Medicine, Linköping University, S-581 85 Linköping, Sweden. E-mail:

anders.blomqvist@liu.se; Dr. Peter M. Zygmunt, Division of Clinical Chemistry and

Pharmacology, Department of Laboratory Medicine, Lund University, S-221 85 Lund,

Sweden. E-mail: peter.zygmunt@med.lu.se

Mirrasekhian, Nilsson et al., p. 2

Nonstandard Abbreviations

Cox cyclooxygenase

KO knock-out

LPS lipopolysaccharide

NAC N-acetyl cysteine

NAPQI N-acetyl-p-benzoquinoneimine

NSAID nonsteroidal anti-inflammatory drug

PG prostaglandin

TRPA1 transient receptor potential ankyrin 1

TRPV1 transient receptor potential vanilloid 1

TX thromboxane

WT wild-type

Mirrasekhian, Nilsson et al., p. 3

Abstract

The mode of action of paracetamol (acetaminophen), which is widely used for treating pain

and fever, has remained obscure, but may involve several distinct mechanisms, including

cyclooxygenase inhibition and transient receptor potential ankyrin 1 (TRPA1) channel

activation, the latter being recently associated with paracetamol’s propensity to elicit

hypothermia at higher doses. Here we examined if the antipyretic effect of paracetamol was

due to TRPA1 activation or cyclooxygenase inhibition. Treatment of wild-type and TRPA1

knock-out mice rendered febrile by immune challenge with lipopolysaccharide (LPS) with a

dose of paracetamol that did not produce hypothermia (150 mg/kg), but known to be

analgetic, abolished fever in both genotypes. Paracetamol completely suppressed the LPS

induced elevation of prostaglandin E2 in the brain, and also reduced the levels of several

other prostanoids in brain as well as in blood. The hypothermia induced by paracetamol was

abolished in mice treated with the electrophile scavenger N-acetyl cysteine. We conclude that

paracetamol’s antipyretic effect in mice is dependent on inhibition of cyclooxygenase

activity, including the formation of pyrogenic prostaglandin E2, whereas paracetamol-

induced hypothermia likely is mediated by the activation of TRPA1 by electrophilic

metabolites of paracetamol, similar to its analgesic effect in some experimental paradigms.

Key words: Paracetamol, fever, TRPA1, hypothermia, N-acetyl cysteine, prostaglandins

Mirrasekhian, Nilsson et al., p. 4

Introduction

Paracetamol (acetaminophen) is one of the most commonly used over-the-counter drugs to

treat pain and fever. However, its mechanism of action has remained a matter of discussion.

Similar to nonsteroidal anti-inflammatory drugs (NSAIDs), paracetamol has been shown to

inhibit prostaglandin synthesis (1, 2), but has been suggested to do so not by interfering with

the active site of prostaglandin synthesizing cyclooxygenases, as done by NSAIDs, but by

reducing their active oxidized form to their resting form (3). Paracetamol mediated

cyclooxygenase inhibition would therefore be more effective under conditions of low

peroxide concentration, such as in the brain, but inactive under conditions of high peroxidase

concentrations, such as in an inflammatory site. Such a mechanism would be consistent with

the known tissue selectivity of acetaminophen and explain why it is antipyretic and anti-

nociceptive while possessing only minor anti-inflammatory properties (3, 4).

During recent years, the idea that paracetamol exerts its action by cyclooxygenase

inhibition has been challenged. It has been shown that the anti-nociceptive effect of

paracetamol may be executed by the formation of paracetamol metabolites, such as the N-

acylphenolamine derivative N-arachidonoylaminophenol (AM404) and N-acetyl-p-

benzoquinoneimine (NAPQI) activating the TRP ion channels transient receptor potential

vanilloid 1 (TRPV1) and transient receptor potential ankyrin 1 (TRPA1), respectively (5-8).

Paracetamol given at a high dose is known to induce hypothermia both in mice and humans

(9), and it was recently demonstrated that this effect is dependent on TRPA1 but not on

TRPV1 (10). In the same study it was also reported that the same high dose of paracetamol

that produced hypothermia in wild-type (WT) mice normalized body temperature in TRPA1

knock-out (KO) mice in a model of yeast induced fever, suggesting that paracetamol

produces antipyretic and hypothermic effects through distinct and independent mechanisms

(10).

Here we examined the antipyretic effect of paracetamol, using a well-established

lipopolysaccharide (LPS)-induced fever paradigm (11-13). We titrated a dose of paracetamol

that does not produce hypothermia when given to WT mice, and after having verified that

paracetamol-induced hypothermia is TRPA1 mediated, we administered the sub-hypothermic

dose to TRPA1 KO and WT mice that had been rendered febrile by peripheral injection of

bacterial cell wall lipopolysaccharide. We report that paracetamol exerts an antipyretic effect

in both genotypes, i.e. it is independent of TRPA1, and we show that the antipyretic effect is

associated with cyclooxygenase inhibition and normalized prostaglandin E2 levels in the

Mirrasekhian, Nilsson et al., p. 5

brain. Furthermore, we show that the glutathione precursor N-acetyl cysteine protects WT

mice from the hypothermic effect of paracetamol, indicating that the paracetamol-induced

hypothermia, similar to what was shown for its analgesic effect in some experimental

paradigms, is mediated by the activation of TRPA1 by paracetamol’s electrophilic

metabolites, such as NAPQI (5).

Materials and Methods

Animals

Adult TRPA1 KO and WT mice of both sexes were used, and experimental groups were

balanced with respect to sex and age. The KO mice were originally provided by Dr. David

Julius (14) (RRID: MGI:3696956) and backcrossed for over 10 generations onto C57BL/6

mice (Taconic, Denmark) in the laboratory of Högestätt and Zygmunt. Homozygous breeding

was then used for 2-3 generations to create KO and WT mice for experiments. The mice were

kept at a 12:12 h light/dark cycle (light on at 7.00 a.m.) with free access to food and water.

All injections were given early during the light period. Between and after injections animals

were left undisturbed to avoid eliciting activity related temperature responses. All

experiments were approved by the Linköping and Lund Animal Ethics Committees, and were

in accordance with the European Union directive for the use of animals for scientific

purposes (2010/63).

Drugs

In the experiments monitoring body temperature, LPS from Escherichia coli (100 µg/kg;

Sigma, St. Louis, MO, USA; 0111:B4), paracetamol (100-200 mg/kg; Sigma), and parecoxib

(10 mg/kg, Dynastat®; Pfizer, Sollentuna, Sweden) were all diluted in saline and injected

intraperitoneally (i.p.; ~ 0.1 ml). N-acetyl cysteine (1g/kg; Meda, Solna, Sweden) was diluted

in saline and injected subcutaneously (s.c.; ~0.3 ml).

Telemetric temperature recordings

The mice were briefly anesthetized with 1 % isoflurane (Abbot Scandinavia, Solna, Sweden)

and implanted i.p. with a transmitter that records core body temperature (Data Sciences

International, New Brighton, MN, USA; model TA11TAF10). Buprenorphine (Temgesic;

100 µg/kg, in about 0.1 ml saline i.p.; RB Pharmaceuticals, Sough, Berkshire, UK) was given

for post-operative analgesia. Immediately after the surgery, the mice were transferred to a

room in which the ambient temperature was set to 29°C, providing near-thermoneutral

conditions (13). The animals were allowed to recover for at least 1 week before any

Mirrasekhian, Nilsson et al., p. 6

recordings were made. This time period was selected to permit e.g. the replenishment of

cytochrome P450 2E1 (CYP2E1), the liver enzyme which metabolizes paracetamol to its

electrophilic metabolite NAPQI, which activates TRPA1 (5), and which is known to be

inhibited by isoflurane and other fluorinated anesthetics (15). Prior to drug injection, the

basal temperature of each mouse was recorded for at least 24 h to assure that they displayed

normal body temperature with normal circadian variation.

Mass spectrometry analyses of prostanoids

LPS (100 µg/kg i.p.) or saline was administered 4 h prior to i.p. injection of acetaminophen

(150 mg/kg) or vehicle [Ringer’s acetate solution, Fresenius Kabi AB, Uppsala, Sweden (Na+

131 mM, Cl- 112 mM, Acetate 30 mM, K+ 4 mM, Ca2+ 2 mM, Mg 2+ 1 mM)]. Forty minutes

after the acetaminophen injection, the animals were anesthetized with isoflurane and

decapitated. The brain was removed, and the diencephalon was promptly dissected, snap

frozen in liquid nitrogen, and stored at -80°C. Blood was collected in test tubes containing

100 µl buffered citrate and ibuprofen (500 µM), frozen on dry ice, and stored at -80°C.

Blood and diencephalon were homogenized in 1 ml ice-cold 10 mM Tris buffer, to

which had been added 1 mM EDTA, 300 µM ascorbic acid, 10 µM methyl arachidonyl

fluorophosphonate (Cayman Chemical Co., Ann Arbor, MI, USA) and 100 µM ibuprofen to

prevent in vitro prostanoid formation and degradation. Aliquots (200 µl) of homogenates

(blood and diencephalon) were precipitated with 1 ml ice-cold acetone, containing 300 µM

ascorbic acid and 10 nM PGE1-d4 (Cayman Chemical), as internal standard. After

centrifugation at 17960 g for 10 min at 4°C, all supernatants were vacuum evaporated. Each

extraction residual was reconstituted in 200 µl of a mixture of 80 % methanol, 19.5 % H2O

and 0.5 % acetic acid. The protein content of the pellet was determined by Coomassie protein

assay (Pierce, Illinois, USA), using bovine serum albumin as standard reference.

All samples were analyzed on a Shimadzu HPLC system (Shimadzu, Kyoto, Japan)

coupled to a Sciex 5500 tandem mass spectrometer (Applied Biosystems/MDS Sciex,

Darmstadt, Germany). PGE2 (Sigma-Aldrich, Schnelldorf Germany), PGF2α (Sigma-

Aldrich), 6-keto-PGF1α (a stable metabolite of prostaglandin I2; Cayman Chemical) and

TXB2 (a stable metabolite of thromboxane A2; Biomol GmbH, Hamburg, Germany), were

quantified by LC-MS/MS in negative mode. Sample aliquots of 5 µl were injected onto a

Waters CSH C18 column [50 × 2 mm, 3 µm (Waters, Milford, USA)] held at 50°C. All

mobile phases were water-acetonitrile gradients, containing 0.1 % formic acid and using a

flow rate of 500 µl/min. The acetonitrile content was initially 20 %, then increased linearly to

Mirrasekhian, Nilsson et al., p. 7

100 % over 3 min and kept at this level for 3 min 15 s, followed by an equilibration period at

20 % for 45 s. The electrospray interface was operated in the positive ion mode at 550°C and

the ion spray voltage was set to ~ 4500 volts. The mass transitions (m/z) were as follows:

351.2/315.3 for PGE2, 357.2/321.2 for PGE1-d4, 369.2/169.1 for TXB2, 353.4/309.3 for

PGF2α, and 369.4/163.1 for 6-keto-PGF1α. The declustering potentials and the collision

energies were between -105 and -120 volts and between -17 and -36 volts, respectively. All

samples were normalized to an internal standard (PGE1-d4). The detection limits were

calculated as the concentrations corresponding to three times the standard deviation of the

blanks. When levels were below detection limit (diencephalon: PGE2, 493,26 pmol/g protein;

TXB2, 162,88; 6-keto-PGF1α, 264,30; PGF2α, 1874,68; blood: PGE2, 99,33; TXB2, 21,57),

numerical values of 1/4 of the detection limit were used in the calculations.

Statistical analyses

Analysis of temperature differences (nadir temperatures) between treatments of WT mice

with different doses of paracetamol or of differences in WT mice in prostanoid

concentrations in blood and brain following saline + vehicle, LPS + vehicle or LPS +

paracetamol treatment was done with a one-way ANOVA, followed by Dunnett’s or Tukey’s

multiple comparisons test, respectively. Analysis of treatment effects (paracetamol) on LPS

induced temperature responses of WT and KO mice were done with a two-way ANOVA,

with genotype and treatment as main factors, followed by Tukey’s multiple comparisons test.

All statistical analyses were done in Graph Pad Prism (GraphPad Software, La Jolla, CA,

USA).

Results

TRPA1 KO mice display normal circadian temperature variation

We monitored basal temperature of WT and TRPA1 KO mice during a 48-h period. Both WT

and TRPA1 KO mice showed normal circadian temperature rhythm, and there was no

significant difference between the genotypes (Fig. 1). However, for both genotypes, female

mice displayed a slightly higher average temperature than male mice (37.4°C vs 36.8°C for

WT mice, and 37.3°C vs 36.8 for KO mice).

Paracetamol given at 200 mg/kg, but not at 150 mg/kg, produces TRPA1-dependent

hypothermia

We titrated a dose of the paracetamol that did not produce hypothermia. We found that 200

Mirrasekhian, Nilsson et al., p. 8

mg/kg body weight injected i.p. to WT mice resulted in pronounced hypothermia, whereas

100 and 150 mg/kg did not (1-way ANOVA: F3,14 = 9.292, P = 0.0012 at nadir temperature;

P = 0.0153 for 200 mg vs 150 mg paracetamol; Fig. 2A, A1). This observation is consistent

with that previously reported by Li et al. (16), who found that 160 mg/kg paracetamol

produced hypothermia, whereas doses <160 mg/kg did not. It is also in line with the

observation by Gentry et al. (10) that 300 mg/kg, but not 100 mg/kg produced hypothermia.

We then verified that the hypothermia induced by the higher dose of paracetamol (200

mg/kg) indeed was TRPA1 mediated. We found that whereas WT mice displayed

hypothermia, as in the previous experiment (Fig. 2A), the body temperature of TRPA1 KO

remained normal [2-way ANOVA (genotype x treatment): F1,17 = 4.795, P = 0.0428; P =

0.0463 for paracetamol treatment of WT and KO mice; Fig. 2B, B1].

N-acetyl cysteine abolishes paracetamol induced hypothermia

Because the analgesic effect of paracetamol may be mediated by the binding to TRPA1 of its

electrophilic metabolites, such as NAPQI (5), we examined if NAPQI also was involved in

paracetamol induced hypothermia. We therefore treated WT mice with N-acetyl cysteine

(NAC), which replenishes the glutathione needed for metabolizing NAPQI into an inactive

conjugate (17). We found that mice treated with 1g/kg NAC, given s.c. 20 min prior to

paracetamol (200 mg/kg), did not develop hypothermia, in contrast to mice treated with

vehicle [2-way ANOVA (treatment): F1,20 = 10.41, P = 0.0042; P = 0.0229 for NAC vs

saline treatment of paracetamol treated mice; Fig. 2C, C1].

TRPA1 KO mice display LPS induced fever, similar to WT mice, and in both genotypes 150

mg/kg paracetamol is antipyretic

Having established that a dose of 150 mg/kg paracetamol did not produce hypothermia in

non-immune challenged mice, we used this dose, known to produce analgesia (7, 18), to treat

LPS induced fever. We injected WT and TRPA1 KO mice with LPS i.p. (100 µg/kg) during

the beginning of the light period. This procedure yielded a body temperature elevation of

about 1°C compared to that of mice that were injected with saline [2-way ANOVA

(treatment): F1,51 = 21.23 and P < 0.0001; P = 0.0446 and P = 0.0002 for LPS vs NaCl

treatment of WT and KO mice, respectively; Fig. 3A, A1]. Four hours after the LPS/saline

injection, i.e. at a time point when the LPS treated mice had displayed a sustained fever for

about 2 h (Fig. 3A), we injected the mice i.p. with paracetamol (150 mg/kg) or saline. In mice

of both genotypes that were first injected with saline, the second injection, irrespective of

whether it was paracetamol or saline, yielded a handling stress induced hyperthermia (as did

Mirrasekhian, Nilsson et al., p. 9

the first injection; Fig. 3A). Similarly, mice of both genotypes that first had been injected

with LPS and as a second injection received saline showed a stress induced increase of their

body temperature, but from an already elevated level (Fig. 3A). Importantly, mice first

injected with LPS and which in a second injection was given paracetamol displayed a

prominent but transient temperature fall, which for TRPA1 KO mice normalized their body

temperature [2-way ANOVA (treatment): F1,36 = 54.47 and P < 0.0001; P < 0.0001 and P =

0.0008 for paracetamol vs NaCl treatment of WT and KO mice respectively], hence

demonstrating that TRPA1 is not necessary for the antipyretic effect of paracetamol (Fig. 3A,

A2). In WT mice, however, paracetamol not only normalized the body temperature, but

produced an additional hypothermia [2-way ANOVA (genotype): F1,36 = 5.116 and P =

0.0298; P = 0.0285 for paracetamol treated WT and KO mice; Fig. 3A, A2]. The nadir of the

temperature fall appeared at approximately 1 h after the paracetamol injection and lasted for

about 2 ½ h. These observations fit well with published data on the pharmacokinetics of

paracetamol in mice. After intraperitoneal injection of 200 mg/kg, maximal plasma

concentration was reached already at 15 min and half-life was about 60 min (19). There was

no statistically significant difference between male and female mice, with WT mice of both

sexes displaying hypothermia after paracetamol, whereas neither male or female KO mice did

so.

To determine if the different sensitivity to 150 mg/kg paracetamol between febrile WT

and TRPA1 KO mice was dependent on paracetamol induced activation of TRPA1, we

repeated the experiment with antipyretic treatment of LPS-induced fever, but this time with

the cyclooxygenase-2 (Cox-2) inhibitor parecoxib, for which no interaction with TRPA1

should be expected. Injection of parecoxib 4 h after LPS resulted in a temperature depression

that was almost identical in magnitude between TRPA1 KO and WT mice (Fig. 3B),

suggesting that the differences seen between genotypes in the response to paracetamol were

TRPA1 dependent.

Based on these observations we hypothesized that the increased sensitivity of febrile

mice to the hypothermic effect of paracetamol was due to reduced glutathione stores and

hence reduced capacity to neutralize paracetamol’s electrophilic metabolites. We tested this

hypothesis by treating febrile WT mice with the cysteine prodrug NAC, which replenishes

the intracellular levels of the antioxidant glutathione, before they were injected with

paracetamol. As shown in Fig. 3C, NAC strongly attenuated the paracetamol induced

hypothermia.

Mirrasekhian, Nilsson et al., p. 10

Paracetamol inhibits LPS-induced prostanoid synthesis in the brain

Finally, we examined to what extent paracetamol inhibited prostanoid synthesis when given

during LPS-induced fever. WT mice were injected with LPS followed by either paracetamol

or vehicle 4 h later. A control group was given saline followed 4 h later by vehicle. Animals

were killed 40 min after the last injection and brain and blood were collected for analysis.

This time point is about midway between Tmax for paracetamol (19) and nadir of its

antipyretic effect (Fig. 3A, A2). In selecting this time point we reasoned that the most

prominent suppression of prostaglandin levels in the brain (including the levels of PGE2)

would be seen prior to nadir of the temperature response when temperature still was falling

rapidly, because at nadir, PGE2 levels should be rising to be able to reverse the temperature

fall.

As shown in Fig. 4A, LPS treatment induced PGE2 in the brain, and this induction was

completely suppressed by paracetamol (1-way ANOVA: F2,15 = 36.31, P < 0.0001; P <

0.0001 for LPS vs vehicle and LPS vs LPS + paracetamol). Notably, PGE2 levels were below

detection limit in blood (not shown). Although there was no or only modest increase in the

levels of other analyzed prostanoids (TXB2, PGF2α and 6-keto-PGF1α; Fig. 4B-E) in the LPS

treated mice, paracetamol decreased the levels of these prostanoids in the brain, pointing to a

general suppression of prostanoid synthesis by paracetamol [TXB2 in diencephalon: 1-way

ANOVA: F2,15 = 6.616, P = 0.0087; P = 0.0068 for LPS vs LPS + paracetamol (Fig. 4B);

PGF2α in diencephalon: F2,15 = 4.127, P = 0.0373; P = 0.0349 for LPS vs LPS + paracetamol

(Fig. 4D); 6-keto-PGF1α in diencephalon: F2,15 = 21.05, P < 0.0001; P = 0.0092 for LPS vs

vehicle; P < 0.0001 for LPS vs LPS + paracetamol; and P = 0.02 for LPS + paracetamol vs

vehicle (Fig. 4E)]. Paracetamol treatment also yielded very low TXB2 levels in blood (Fig.

4C); however, the differences vs the non-paracetamol treated groups were not statistically

significant.

Discussion

The present findings show that a dose of paracetamol that does not elicit hypothermia when

given to naïve mice, still has the capacity to normalize the body temperature of mice rendered

febrile by peripheral injection of LPS. The abolishment of fever was seen both in WT mice

and in TRPA1 KO mice, demonstrating that the antipyretic effect of paracetamol is not

dependent on TRPA1. Furthermore, the antipyretic effect of paracetamol was temporally

associated with a complete suppression of the increased PGE2 levels in the brain that was

seen in mice given LPS but treated with vehicle instead of paracetamol. Because induced

Mirrasekhian, Nilsson et al., p. 11

production of PGE2 is critical for fever (20-22), the present findings, taken together, indicate

that paracetamol exerts its antipyretic effect by inhibiting prostaglandin synthesis, a

mechanism that hence is different from the mechanism by which paracetamol, at high doses,

evokes hypothermia (10). The latter mechanism depends on TRPA1 and seems to involve

paracetamol metabolites, such as NAPQI, because, as shown here, it is abolished by NAC, a

substance that promotes inactivation of NAPQI. Hence, it is similar to the mechanism by

which paracetamol is anti-nociceptive, which has been shown also to depend on TRPA1, at

least in some experimental paradigms (5).

The present study also shows that LPS-induced fever is independent of TRPA1 as LPS

evoked the same temperature elevation in both WT and TRPA1 KO mice. In contrast, it has

been shown that pain and acute vascular reactions caused by LPS are primarily dependent on

TRPA1 activation in nociceptive sensory neurons (23) and that LPS-evoked mechanical

hypersensitivity is dependent on TRPA1 via the endogenous activator hydrogen sulfide (24-

26). LPS-induced fever is attenuated by genetic deletion in brain endothelial cells of IL-6

receptors (12) and of prostaglandin synthesizing enzymes (27), implying a mechanism by

which humorally released proinflammatory cytokines act on the brain, which is distinct from

the mechanisms by which LPS elicits pain. These data are well in line with two distinct mode

of actions by which paracetamol is antipyretic and anti-nociceptive, respectively.

We previously demonstrated that the antipyretic action of paracetamol was dependent

on the gene dose of the prostaglandin synthesizing enzyme Cox-2. We showed that mice

heterozygous for the Cox-2 gene, and hence displaying lower levels of Cox-2, were more

sensitive to the fever reducing effect of paracetamol than WT mice (28). Specifically, we

demonstrated that the largest paracetamol dose that had no effect on LPS induced fever in

WT mice strongly attenuated fever in Cox-2 heterozygous mice, whereas no difference was

seen between mice homozygous and heterozygous, respectively, for the terminal PGE2

isomerase microsomal prostaglandin E synthase-1 (mPGES-1), implying that paracetamol

indeed exerted its action by inhibiting Cox-2.

Although Flower and Vane already in 1972 suggested that the antipyretic activity of

paracetamol could be explained by its inhibition of prostaglandin synthase in the brain (2),

surprisingly few studies have demonstrated that paracetamol blocks pyrogen-induced

elevation of brain PGE2. Feldberg et al. (29) demonstrated in cats that elevation of PGE1-like

activity in the cerebrospinal fluid (and fever), evoked by intracerebral injection of bacterial

pyrogen, was abolished when animals were given an intraperitoneal injection of paracetamol,

and Li et al. (16) demonstrated in mice that paracetamol, albeit at a dose that yields

Mirrasekhian, Nilsson et al., p. 12

hypothermia, normalized PGE2 levels in animals that had been immune-challenged with

intravenous injection of LPS. However, studies have shown that paracetamol reduces brain

PGE2 levels also in non-immune challenged mice (7, 30, 31), suggesting that it may act both

on constitutive and inducible PGE2-production. Although the ability of paracetamol to reduce

constitutive levels of PGE2 was not directly tested in the present study, it should be noted that

the average PGE2 level after paracetamol treatment of mice rendered febrile by LPS injection

was in fact lower than that seen in control mice that had been given two consecutive

injections of saline and vehicle only (Fig. 4A); however, the difference was not statistically

significant. Furthermore, the present data showing that paracetamol not only blocked induced

PGE2, implying an effect on Cox-2, but also reduced the levels of other prostaglandins and

TXB2 (Fig. 4B-E), indicate that paracetamol inhibits both Cox-1 and Cox-2. This idea is

consistent with the hypothesis that paracetamol interacts with the peroxidase site of the

cyclooxygenase enzymes (32), present in both isoforms. This effect would preferentially be

seen in tissues with low peroxide concentration, such as the brain.

It should also be noted that there was no induction of PGE2 in blood, with PGE2 levels

being under the detection limit in all examined treatment groups. Blood-borne PGE2, through

diffusion or transport into the brain, has been suggested to be involved in immune-induced

fever (33, 34). However, it is unlikely to play such a role at the time point (280 min post

injection) examined in the present study.

Interestingly, there seems to be a critical dose between 150 mg/kg and 200 mg/kg for

paracetamol-induced hypothermia in mice. Li et al. (16) defined the dose to 160 mg/kg, and

in the present study 200 mg/kg yielded hypothermia, whereas 150 mg did not, and Gentry et

al. (10) reported hypothermia (in WT but not in TRPA1 mice) when 300 mg/kg was given

s.c., whereas 100 mg/kg (given intrathecally) had no effect on body temperature. We found

that WT mice treated with 150 mg/kg paracetamol when having been rendered febrile by a

previous injection of LPS, displayed hypothermia, whereas TRPA1 KO mice did not (Fig.

3A, A2). Since the same treatment with a selective Cox-2 inhibitor instead of paracetamol

resulted in a normalized temperature in both TRPA1 KO and WT mice without eliciting

hypothermia (Fig. 3B), the findings indicated that paracetamol at the given dose indeed

interacted with TRPA1. TRPA1 is a unique chemosensor strictly controlled by the redox

status (35, 36). The reason for the increased sensitivity of the febrile mice to the hypothermic

effect of paracetamol is therefore probably explained by reduced glutathione stores as a

consequence of the oxidative stress associated with inflammation and fever (37) and hence

reduced capacity to neutralize the TRPA1-binding of paracetamol’s electrophilic metabolites.

Mirrasekhian, Nilsson et al., p. 13

This idea was supported by the finding that the cysteine prodrug NAC, which replenishes the

intracellular levels of the antioxidant glutathione and itself is a scavenger of tissue damaging

thiol reactive agents such as NAPQI (17), strongly attenuated the hypothermia elicited by 150

mg/kg paracetamol in the febrile mice (Fig. 3C).

In conclusion, we show that paracetamol exerts an antipyretic effect that is TRPA1-

independent and that this effect is associated with inhibition of the synthesis of PGE2 (as well

as of other prostanoids) in the brain. Taken together with previous findings, these data

support evidence that paracetamol exerts its antipyretic effect by inhibiting cyclooxygenase

enzymes. However, paracetamol also activates TRPA1 through its electrophilic metabolite

NAPQI. This activation elicits hypothermia when paracetamol is given at high doses, but

may also contribute to the antipyretic effect of lower paracetamol doses, especially during

febrile conditions.

Acknowledgements

This study was supported by the Swedish Medical Research Council (#20725 to DE, #07879

to AB and 2014-3801 to PMZ and EDH), the European Research Council (ERC-starting

grant to DE), the Knut and Alice Wallenberg foundation (DE), the Swedish Brain Foundation

(DE and AB), the Swedish Cancer Foundation (#213/692 to AB), the County Council of

Östergötland (DE and AB), the Medical Faculty of Lund University (ALF) (EDH and PMZ)

and AFA insurance (EDH and PMZ). We thank Anna Eskilsson for help with the

illustrations.

Author contributions

E.D.H., and P.M.Z. conceived the study. A. Blomqvist, D.E., E.D.H. and J.N. designed and

directed research; E.M. and K.S. conducted the temperature measurements; J.N. conducted

the biochemical experiments; A. Blomgren performed the mass spectrometry analyses; A.

Blomqvist, D.E., E.D.H., E.M. and J.N. analyzed data; A. Blomqvist drafted the manuscript.

All authors discussed the results and revised or commented on the manuscript.

Mirrasekhian, Nilsson et al., p. 14

References

1. Flower, R. J., Gryglewski, R., Herbaczynska-Cedro, K., and Vane, J. R. (1972)

Effects of anti-inflammatory drugs on prostaglandin biosynthesis. Nat. New Biol. 238,

104-106

2. Flower, R. J., and Vane, J. R. (1972) Inhibition of prostaglandin synthetase in brain

explains the anti- pyretic activity of paracetamol (4-acetamidophenol). Nature 240,

410-411

3. Ouellet, M., and Percival, M. D. (2001) Mechanism of acetaminophen inhibition of

cyclooxygenase isoforms. Arch. Biochem. Biophys. 387, 273-280

4. Boutaud, O., Aronoff, D. M., Richardson, J. H., Marnett, L. J., and Oates, J. A. (2002)

Determinants of the cellular specificity of acetaminophen as an inhibitor of

prostaglandin H(2) synthases. Proc. Natl. Acad. Sci. U S A 99, 7130-7135

5. Andersson, D. A., Gentry, C., Alenmyr, L., Killander, D., Lewis, S. E., Andersson,

A., Bucher, B., Galzi, J.-L., Sterner, O., Bevan, S., Högestätt, E. D., and Zygmunt, P.

M. (2011) TRPA1 mediates spinal antinociception induced by acetaminophen and the

cannabinoid Δ9-tetrahydrocannabiorcol. Nat. Commun. 2, 551

6. Hogestatt, E. D., Jonsson, B. A., Ermund, A., Andersson, D. A., Bjork, H., Alexander,

J. P., Cravatt, B. F., Basbaum, A. I., and Zygmunt, P. M. (2005) Conversion of

acetaminophen to the bioactive N-acylphenolamine AM404 via fatty acid amide

hydrolase-dependent arachidonic acid conjugation in the nervous system. J. Biol.

Chem. 280, 31405-31412

7. Mallet, C., Barriere, D. A., Ermund, A., Jonsson, B. A., Eschalier, A., Zygmunt, P.

M., and Hogestatt, E. D. (2010) TRPV1 in brain is involved in acetaminophen-

induced antinociception. PLoS One 5, e12748

8. Zygmunt, P. M., Chuang, H., Movahed, P., Julius, D., and Hogestatt, E. D. (2000)

The anandamide transport inhibitor AM404 activates vanilloid receptors. Eur. J.

Pharmacol. 396, 39-42

9. Walker, R. M., Massey, T. E., McElligott, T. F., and Racz, W. J. (1981)

Acetaminophen-induced hypothermia, hepatic congestion, and modification by N-

acetylcysteine in mice. Toxicol. Appl. Pharmacol. 59, 500-507

10. Gentry, C., Andersson, D. A., and Bevan, S. (2015) TRPA1 mediates the hypothermic

action of acetaminophen. Sci. Rep. 5, 12771

Mirrasekhian, Nilsson et al., p. 15

11. Engström, L., Ruud, J., Eskilsson, A., Larsson, A., Mackerlova, L., Kugelberg, U.,

Qian, H., Vasilache, A. M., Larsson, P., Engblom, D., Sigvardsson, M., Jönsson, J.-I.,

and Blomqvist, A. (2012) Lipopolysaccharide-induced fever depends on

prostaglandin E2 production specifically in brain endothelial cells. Endocrinology

153, 4849-4861

12. Eskilsson, A., Mirrasekhian, E., Dufour, S., Schwaninger, M., Engblom, D., and

Blomqvist, A. (2014) Immune-induced fever is mediated by IL-6 receptors on brain

endothelial cells coupled to STAT3-dependent induction of brain endothelial

prostaglandin synthesis. J. Neurosci. 34, 15957-15961

13. Rudaya, A. Y., Steiner, A. A., Robbins, J. R., Dragic, A. S., and Romanovsky, A. A.

(2005) Thermoregulatory responses to lipopolysaccharide in the mouse: dependence

on the dose and ambient temperature. Am. J. Physiol. Regul. Integr. Comp. Physiol.

289, R1244-1252

14. Bautista, D. M., Jordt, S.-E., Nikai, T., Tsuruda, P. R., Read, A. J., Poblete, J.,

Yamoah, E. N., Basbaum, A. I., and Julius, D. (2006) TRPA1 mediates the

inflammatory actions of environmental irritants and proalgesic agents. Cell 124, 1269-

1282

15. Tateishi, T., Watanabe, M., Nakura, H., Tanaka, M., Kumai, T., Aoki, T., and

Kobayashi, S. (1998) A comparison of the inhibitory effects of four volatile

anaesthetics on the metabolism of chlorzoxazone, a substrate for CYP2E1, in rabbits.

Acta Anaesthesiol. Scand. 42, 1028-1032

16. Li, S., Dou, W., Tang, Y., Goorha, S., Ballou, L. R., and Blatteis, C. M. (2008)

Acetaminophen: antipyretic or hypothermic in mice? In either case, PGHS-1b (COX-

3) is irrelevant. Prostag. Oth. Lipid M. 85, 89-99

17. Atkuri, K. R., Mantovani, J. J., Herzenberg, L. A., and Herzenberg, L. A. (2007) N-

Acetylcysteine--a safe antidote for cysteine/glutathione deficiency. Curr. Opin.

Pharmacol. 7, 355-359

18. Doherty, N. S., Beaver, T. H., Chan, K. Y., Dinerstein, R. J., and Diekema, K. A.

(1990) The antinociceptive activity of paracetamol in zymosan-induced peritonitis in

mice: the role of prostacyclin and reactive oxygen species. Br J Pharmacol 101, 869-

874

19. Raskovic, A., Cvejic, J., Stilinovic, N., Golocorbin-Kon, S., Vukmirovic, S., Mimica-

Dukic, N., and Mikov, M. (2014) Interaction between different extracts of Hypericum

Mirrasekhian, Nilsson et al., p. 16

perforatum L. from Serbia and pentobarbital, diazepam and paracetamol. Molecules

19, 3869-3882

20. Engblom, D., Saha, S., Engstrom, L., Westman, M., Audoly, L. P., Jakobsson, P. J.,

and Blomqvist, A. (2003) Microsomal prostaglandin E synthase-1 is the central

switch during immune-induced pyresis. Nat. Neurosci. 6, 1137-1138

21. Nilsberth, C., Hamzic, N., Norell, M., and Blomqvist, A. (2009) Peripheral

lipopolysaccharide administration induces cytokine mRNA expression in the viscera

and brain of fever-refractory mice lacking microsomal prostaglandin E synthase-1. J.

Neuroendocrinol. 21, 715-721

22. Saha, S., Engstrom, L., Mackerlova, L., Jakobsson, P. J., and Blomqvist, A. (2005)

Impaired febrile responses to immune challenge in mice deficient in microsomal

prostaglandin E synthase-1. Am. J. Physiol. Regul. Integr. Comp. Physiol. 288,

R1100-1107

23. Meseguer, V., Alpizar, Y. A., Luis, E., Tajada, S., Denlinger, B., Fajardo, O.,

Manenschijn, J.-A., Fernández-Peña, C., Talavera, A., Kichko, T., Navia, B.,

Sánchez, A., Señarís, R., Reeh, P., Pérez-García, M. T., López-López, J. R., Voets, T.,

Belmonte, C., Talavera, K., and Viana, F. (2014) TRPA1 channels mediate acute

neurogenic inflammation and pain produced by bacterial endotoxins. Nat. Commun. 5,

3125

24. Andersson, D. A., Gentry, C., and Bevan, S. (2012) TRPA1 as a key role in the

somatic pro-nociceptive actions of hydrogen sulfide. PLoS ONE 7, e46917

25. Eberhardt, M., Dux, M., Namer, B., Miljkovic, J., Cordasic, N., Will, C., Kichko, T.

I., de la Roche, J., Fischer, M., Suárez, S. A., Bikiel, D., Dorsch, K., Leffler, A.,

Babes, A., Lampert, A., Lennerz, J. K., Jacobi, J., Martí, M. A., Doctorovich, F.,

Högestätt, E. D., Zygmunt, P. M., Ivanovic-Burmazovic, I., Messlinger, K., Reeh, P.,

and Filipovic, M. R. (2014) H(2)S and NO cooperatively regulate vascular tone by

activating a neuroendocrine HNO–TRPA1–CGRP signalling pathway. Nat. Commun.

5, 4381

26. Streng, T., Axelsson, H. E., Hedlund, P., Andersson, D. A., Jordt, S. E., Bevan, S.,

Andersson, K. E., Hogestatt, E. D., and Zygmunt, P. M. (2008) Distribution and

function of the hydrogen sulfide-sensitive TRPA1 ion channel in rat urinary bladder.

Eur. Urol. 53, 391-399

27. Wilhelms, D. B., Kirilov, M., Mirrasekhian, E., Eskilsson, A., Kugelberg, U. Ö., Klar,

C., Ridder, D., Herschman, H. R., Schwaninger, M., Blomqvist, A., and Engblom, D.

Mirrasekhian, Nilsson et al., p. 17

(2014) Deletion of prostaglandin E2 synthesizing enzymes in brain endothelial cells

attenuates inflammatory fever. J. Neurosci. 34, 11684-11690

28. Engström Ruud, L., Wilhelms, D. B., Eskilsson, A., Vasilache, A. M., Elander, L.,

Engblom, D., and Blomqvist, A. (2013) Acetaminophen reduces lipopolysaccharide-

induced fever by inhibiting cyclooxygenase-2. Neuropharmacology 71, 124-129

29. Feldberg, W., and Gupta, K. P. (1973) Pyrogen fever and prostaglandin-like activity

in cerebrospinal fluid. J. Physiol. 228, 41-53

30. Ayoub, S. S., Botting, R. M., Goorha, S., Colville-Nash, P. R., Willoughby, D. A.,

and Ballou, L. R. (2004) Acetaminophen-induced hypothermia in mice is mediated by

a prostaglandin endoperoxide synthase 1 gene-derived protein. Proc. Nat. Acad. Sci.

U S A 101, 11165-11169

31. Ayoub, S. S., Colville-Nash, P. R., Willoughby, D. A., and Botting, R. M. (2006) The

involvement of a cyclooxygenase 1 gene-derived protein in the antinociceptive action

of paracetamol in mice. Eur. J. Pharmacol. 538, 57-65

32. Anderson, B. J. (2008) Paracetamol (Acetaminophen): mechanisms of action. Pediatr.

Anesth. 18, 915-921

33. Steiner, A. A., Chakravarty, S., Rudaya, A. Y., Herkenham, M., and Romanovsky, A.

A. (2006) Bacterial lipopolysaccharide fever is initiated via Toll-like receptor 4 on

hematopoietic cells. Blood 107, 4000-4002

34. Steiner, A. A., Ivanov, A. I., Serrats, J., Hosokawa, H., Phayre, A. N., Robbins, J. R.,

Roberts, J. L., Kobayashi, S., Matsumura, K., Sawchenko, P. E., and Romanovsky, A.

A. (2006) Cellular and molecular bases of the initiation of fever. PLoS Biol. 4, e284

35. Moparthi, L., Kichko, T. I., Eberhardt, M., Hogestatt, E. D., Kjellbom, P., Johanson,

U., Reeh, P. W., Leffler, A., Filipovic, M. R., and Zygmunt, P. M. (2016) Human

TRPA1 is a heat sensor displaying intrinsic U-shaped thermosensitivity. Sci. Rep. 6,

28763

36. Zygmunt, P. M., and Hogestatt, E. D. (2014) TRPA1. Handb. Exp. Pharmacol. 222,

583-630

37. Bar-Or, D., Bar-Or, R., Rael, L. T., and Brody, E. N. (2015) Oxidative stress in severe

acute illness. Redox Biol. 4, 340-345

Mirrasekhian, Nilsson et al., p. 18

Legends to figures

Fig. 1. Diurnal body temperature variations in wild-type (WT) and TRPA1 knock-out (KO)

mice. Solid lines represent mean, and dotted lines SEM. Dark periods are indicated by

horizontal bars on the abscissa. n = 6.

Fig. 2. Temperature responses to paracetamol in wild-type (WT) and TRPA1 knock-out (KO)

mice, and effect of N-acetyl cysteine. A. Paracetamol at a dose of 200 mg/kg (injected at time

zero; arrow) elicits hypothermia, whereas a dose of 150 mg/kg or less does not. Note that the

initial temperature peak is due to the handling stress associated with the intraperitoneal

injection. Solid lines represent mean, and dotted lines SEM. A1 shows nadir temperatures

(means ± SEM) at 80 min after the paracetamol/saline injection (vertical line in A). *

indicates P < 0.05. n = 4-6. B. TRPA1 KO mice, in contrast to WT mice, do not display

hypothermia after treatment with 200 mg/kg paracetamol (injected at time zero; arrow). Solid

lines represent mean, and dotted lines SEM. B1 shows nadir temperatures (means ± SEM) at

80 min after the paracetamol/saline injection (vertical line in B). * and ** indicate P < 0.05

and P < 0.01, respectively. n = 5-6. C. N-acetyl cysteine (NAC) blocks paracetamol induced

hypothermia. NAC (1g/kg body weight) or vehicle was injected s.c. 40 min prior to the

paracetamol injection (200 mg/kg, at time zero). C1 shows values at 106 min (vertical line in

C) after paracetamol/saline injection. * indicates P < 0.05. n = 7 for paracetamol treated

animals and n = 5 for vehicle treated animals. While the initial temperatures differed between

groups, there were no differences in the mean temperature of the different groups during 24 h

preceding the experiments (not shown).

Fig. 3. Antipyretic effect of paracetamol. A. Intraperitoneal injection of LPS (100 µg/kg at

time zero; left arrow) results in fever in both wild-type (WT) and TRPA1 knock-out (KO)

mice. 150 mg/kg paracetamol given during the LPS induced fever (at 240 min; right arrow)

normalizes body temperature in TRPA1 KO mice, but results in hypothermia in WT mice.

Solid and dashed lines represent mean, and dotted lines SEM. Note that half of the WT and

KO mice (dashed blue and magenta lines) at 240 min were injected with either paracetamol

(solid dark blue and magenta lines) or saline (solid light blue and pink lines). n = 20 and 10

before and after 240 min, respectively. Solid green line represent WT and KO mice injected

with saline + saline and saline + paracetamol. n = 15. The temperature curves for the

individual groups are shown in Supplementary Fig. 1. A1 shows the febrile response to LPS

of WT and KO mice at 180 min after LPS injection (left vertical line in A). * and ** indicate

P < 0.05 and P < 0.01, respectively. n = 20 for LPS treated mice and n = 7-8 for saline treated

Mirrasekhian, Nilsson et al., p. 19

mice. Note that saline treated groups include for each genotype mice that at 240 min after

LPS injection were treated with either paracetamol or saline, and the color of these

temperature traces as shown in Supplementary Fig. 1 is represented by the pattern color and

fill color, respectively, of each bar. A2 shows nadir temperatures (means ± SEM) 70 min after

treatment with 150 mg/kg paracetamol (right vertical line) that was given at 240 min post

LPS injection. Horizontal dashed line indicate nadir temperature for WT and KO mice treated

with saline + saline (i.e. the ”normal” temperature in this experimental setting). * and ***

indicate P < 0.05 and P < 0.001, respectively. n = 10. B. Parecoxib, 10 mg/kg, given 240 min

after LPS injection (at time zero) normalizes LPS induced fever in both WT and TRPA1 KO

mice. Solid lines represent mean, and dotted lines SEM. n = 8. C. N-acetyl cysteine (1g/kg

s.c.) given 30 min prior to 150 mg/kg paracetamol in an experimental set-up similar to that

shown in A (100 mg/kg LPS injected i.p. 4 h prior to paracetamol) attenuates the paracetamol

induced hypothermia of WT mice. *** indicates P < 0.001. n = 8 (male mice).

Fig. 4. Concentrations of prostanoids in brain and blood of mice immune challenged with

LPS (100 µg/kg), with or without treatment with paracetamol (150 mg/kg). A. PGE2 levels in

the diencephalon in mice treated with saline and vehicle, LPS and vehicle, or LPS and

paracetamol. PGE2 levels were also analyzed in blood, but was below detection level (not

shown). B. TXB2 levels in the diencephalon. C. TXB2 levels in blood. D. PGF2α levels in the

diencephalon. E. 6-keto-PGF1α levels in the diencephalon. *, **, *** indicate P < 0.05, 0.01

and 0.001, respectively. n = 6.

12 18 24 30 36 42 4835

36

37

38

39 WT TRPA1 KO

tem

pera

ture

(o C)

60time (h)

Figure 1

-60 60 120 180 240 300 36033.534.034.535.035.536.036.537.037.538.0

time (min)

tem

pera

ture

(o C)

paracetamol 100 mg/kgparacetamol 150 mg/kg

salineparacetamol 200 mg/kg

A

0 saline 100 mg 150 mg 200 mg34

35

36

37

paracetamol

A1

tem

pera

ture

(o C)

*

-60 60 120 180 240 300 36034.535.035.536.036.537.037.538.038.5

time (min)

WT paracetamol 200 mg/kgKO paracetamol 200 mg/kgWT salineKO saline

B

tem

pera

ture

(o C)

WT KO

35

36

37

38

paracet saline paracet saline

B1

0

**

*

tem

pera

ture

(o C)B1

A1

Figure 2

para

ceta

mol

/sal

ine

para

ceta

mol

/sal

ine

-60 60 120 180 240 300 360 42035.0

35.5

36.0

36.5

37.0

37.5

time (min)

NAC + paracetamol 200 mg/kgsaline + paracetamol 200 mg/kgNAC + salinesaline + saline

0

tem

pera

ture

(o C)

C

C1

para

ceta

mol

/sal

ine

NAC

/sal

ine

NAC paracet

NaClparacet

NAC saline

NaClsaline

35.0

35.5

36.0

36.5

37.0

37.5

tem

pera

ture

(o C)

*

C1

35.5

36.0

36.5

37.0

37.5

A1

WT KOLPS saline LPS saline

* **

WT KO

34

35

36

37

38

paracetsaline saline paracet

A2*** ***

tem

pera

ture

(o C)

tem

pera

ture

(o C)

Figure 3

-60 60 120 180 240 300 360 420 480 54035.035.536.036.537.037.538.038.539.0

time (min)

WTKO

0

B

tem

pera

ture

(o C)

pare

coxi

b

LPS

C

*

-60 0 60 120 180 240 300 360 420 480 54034.034.535.035.536.036.537.037.538.0

time (min)

tem

pera

ture

(°C)

WT LPS + paracetamol 150 mg/kg

KO LPS + paracetamol 150 mg/kgsaline/paracetamol controls

KO LPS

WT LPS

KO LPS + saline

WT LPS + salineLP

S/sa

line

para

ceta

mol

/sal

ine

A1 A2

A

-60 0 60 120 180 240 300 360 420 48033.534.034.535.035.536.036.537.037.5

time (min)

vehicleNAC ∗∗∗

LPS

NAC

/sal

ine

para

ceta

mol

tem

pera

ture

(o C)

PGE2 diencephalonpm

ol/g

pro

tein

saline + LPS + LPS + paracetamol

0

1000

2000

3000

4000

A TXB2 diencephalon

0

500

1000

1500

pmol

/g p

rote

in

B

0

100

200

300

saline + LPS + LPS + paracetamol

saline + LPS + LPS + paracetamol

TXB2 blood

0

2000

4000

6000

PGF2a diencephalon

pmol

/g p

rote

in

pmol

/g p

rote

in

saline + LPS + LPS + paracetamol

0

500

1000

1500

2000

2500

pmol

/g p

rote

in

6-keto-PGF1a diencephalon

saline + LPS + LPS + paracetamol

C D

E

*** ***

** ***

*

**

*

vehiclevehicle vehicle vehicle

vehicle vehicle vehicle vehicle

vehicle vehicle

Figure 4

-60 60 120 180 240 300 360 420 480 54034.5

35.0

35.5

36.0

36.5

37.0

37.5

38.0

time (min)

WT LPS + paracetamol 150 mg/kg

KO LPS + paracetamol 150 mg/kg

WT LPS + saline

KO LPS + saline

WT saline + paracetamol 150 mg/kg

KO saline + paracetamol 150 mg/kg

WT saline + saline

KO saline + saline

0

tem

pera

ture

(o C)

para

ceta

mol

/sal

ine

LPS/

salin

e

Supplementary Fig. 1. Temperature traces for each group of mice in the experiment reported in Fig. 3A. The upper 4 traces (deep blue and light blue for WT mice, and magenta and pink for KO mice) seen between about 120 and 240 min represent mice that were given LPS, and the lower 4 traces (dark green and green for WT mice, and brown and yellow for KO mice) represent mice that instead were given saline. At about 300 min, the upper two traces (light blue and pink) represent LPS treated mice WT or KO, respectively, that were given saline at 240 min, whereas the deep blue and magenta traces represent LPS treated WT and KO mice that were given paracetamol at 240 min. Dark green, green, brown and yellow traces represent WT and KO animals that first were injected with saline (instead of LPS) and at 240 min given either paracetamol (dark green and brown) or saline (green and yellow). n = 10 for LPS treated mice and n = 3-4 for mice given saline instead of LPS.