PHYSIOLOGICAL RESEARCH • ISSN 0862-8408 (print) • ISSN 1802-9973 (online)
© 2009 Institute of Physiology v.v.i., Academy of Sciences of the Czech Republic, Prague, Czech Republic
Fax +420 241 062 164, e-mail: firstname.lastname@example.org, www.biomed.cas.cz/physiolres
Physiol. Res. 58: 419-425, 2009
Increased Gene Expression and Production of Spinal
Cyclooxygenase 1 and 2 during Experimental Osteoarthritis Pain
M. PROCHÁZKOVÁ1, P. ZANVIT2, T. DOLEŽAL1, L. PROKEŠOVÁ2, M. KRŠIAK1
1Department of Pharmacology, Third Faculty of Medicine, Prague, 2Institute of Immunology and
Microbiology, First Faculty of Medicine, Charles University, Prague, Czech Republic
Received October 29, 2007
Accepted April 2, 2008
On-line July 18, 2008
Knowledge on the involvement of spinal COX-1 and COX-2 in
pain due to osteoarthritis could be useful for better
understanding of its pathogenesis and therapy. In this study we
have investigated a long-term pattern of expression and
production of spinal COX-1 and COX-2 in the model of
osteoarthritis induced in rats by injection of monoiodoacetate
(MIA) into the knee joint. MIA injection produced thermal
hyperalgesia (assessed by the plantar test) and tactile allodynia
(measured with von Frey hairs). The pain measures reached
maximum on the 5th day, then remained relatively stable. The
expression of spinal COX-2 mRNA reached maximum on day 5
(5.2 times; P<0.001) and remained increased until day 31 (4.9
times; P<0.001). Expression of spinal COX-1 mRNA increased
gradually reaching maximum on the day 31 (4.5 times; P<0.001)
when the relative expression of both genes was almost equal.
The production of both proteins was almost similar at the
beginning of the experiment. The highest production of COX-2
protein was observed on day 5 after the induction of
osteoarthritis (increased 3.9 times). The levels of COX-1 protein
increased gradually with maximum on day 31 (3.4 times). The
present findings indicate that not only expression of COX-2
mRNA but also that of COX-1 mRNA is significantly increased in
the spine during osteoarthritis pain. Thus, in contrast to
inflammatory pain, the upregulation of spinal COX-1 may be
important in osteoarthritis pain.
Osteoarthritis • Cyclooxygenase (COX) • Pain • Spinal cord
M. Procházková, Department of Pharmacology, 3rd Faculty of
Medicine, Charles University, Ruská 87, 100 34 Prague 10, Czech
Republic. E-mail: email@example.com, Fax: +420
267 102 461.
the conversion of arachidonic acid to prostaglandins
which play an important role in inflammation and pain.
There are two identified cyclooxygenase (COX)
isoenzymes: COX-1 and COX-2. COX-1 is the „house-
keeping“ form, expressed by a wide variety of cells.
COX-2 is highly inducible in response to inflammatory
and noxious stimuli. Both cyclooxygenases are
constitutively expressed in the spinal cord (Kaufmann et
The involvement of spinal COX-1 and COX-2 in
various pain states is not fully understood. COX-2 gene
expression and production in the spinal cord was
significantly increased in
inflammation induced by intraplantar injection of
Freund´s complete adjutant (Beiche et al. 1996, Hay et al.
1997, Beiche et al. 1998). No increase of COX-1 gene
expression and production in the spinal cord was found in
this model which was associated with swelling,
hyperalgesia and allodynia (Beiche et al. 1996, Hay et al.
1997, Beiche et al. 1998). In agreement with these
findings, spinal COX-2 mRNA were markedly increased
and spinal levels of COX-1 mRNA were not significantly
altered in peripheral inflammation induced by intraplantar
injection of carrageenan (Procházková et al. 2006).
Rats with streptozotocin-induced
exhibited significantly increased levels of spinal COX-2
protein and activity along with hyperalgesia (Ramos et al.
2007). Intrathecal administration of COX-2 inhibitors has
an anti-hyperalgesic effect on streptozotocin-induced
Cyclooxygenases are the enzymes that catalyze
rats with peripheral
420 Procházková et al.
mechanical hyperalgesia (Matsunaga et al. 2007). A
sharp upregulation of spinal COX-2 was reported in the
mouse model of amyotrophic lateral sclerosis (McGeer
and McGeer 2002).
On the other hand, the expression of COX-2
mRNA in the spine was less dominant in postoperative
pain model than in inflammatory pain, while expression
of spinal COX-1 mRNA was significantly increased in
postoperative pain (Procházková et al. 2006). The
important role of COX-1 in the model of postoperative
pain was also shown by Zhu et al. (2003).
Spinal COX-1 also appears to play a role in the
model of neuropathic pain. Spinal COX-1 expression was
increased after partial peripheral nerve transsection (Zhu
and Eisenach 2003). The inhibition of COX-1 prevented
the development of allodynia and hyperalgesia after
peripheral nerve ligation (Hefferan et al. 2003).
One of the widespread painful disorders is
osteoarthritis (OA), which is a degenerative joint disease.
The joints are characterized by progressive degeneration
of articular cartilage leading to inflammation and pain. In
an attempt to peruse
osteoarthritis, experimental models that mimic human
disease have been developed (Pritzker 1994).
One of the most frequently used and one of the
best characterized models of osteoarthritis is that of
monoiodoacetate (MIA) induced OA. This model was
first described twenty years ago by Kalbhen (1987).
Monoiodoacetate inhibits the activity of glyceraldehyde-
producing degeneration of the cartilage.
Knowledge on the involvement of spinal COX-1
and COX-2 in pain due to OA could be useful both for
better understanding of its pathogenesis and for its
therapy. In this study, therefore, we have investigated a
long-term pattern of expression and production of spinal
COX-1 and COX-2 in the model of monoiodoacetate-
induced osteoarthritis. In particular, we have attempted to
correlate changes in spinal cord production of COX-1 and
COX-2 and expression of mRNA for COX-1 and COX-2
with the development of thermal hyperalgesia and tactile
Male Wistar albino rats (weight 200-220 g)
obtained from VÚFB Konárovice (Czech Republic) were
used in all experiments. The animals were housed under
the pathophysiology of
in chondrocytes and
standard laboratory conditions (in a temperature-
controlled (21±1 ºC) room with a normal 12-h light/dark
cycle). Animals were fed a standard pelleted rat chow
(ST-1; Velaz, Czech Republic) with water ad libitum
throughout the whole experiment. Rats were acclimated
to their surroundings over one week to eliminate the
effect of stress before the experiment.
All experiments were approved by the
Committee for Protection of Laboratory Animals of the
3rd Faculty of Medicine at Charles University and were
concordant with IASP Committee for Research and
Ethical Issues requirements (Zimmermann 1983).
Induction of osteoarthritis
For induction of osteoarthritis, rats were
anesthetized with halothane (Narcotan, Zentiva). Eight
rats per group received
monoiodoacetate (2 mg) (Sigma–Aldrich) into the right
knee joint in a total volume of 25 µl. Control animals
(n=8) were injected 25 µl of vehicle into the right knee
joint under the same conditions.
Paw withdrawal testing
The response to noxious thermal stimulus was
determined using thermal plantar device (Ugo Basile,
Italy) according to the procedure described by Hargreaves
et al. (1988) before and in defined times during 31 days
after the injection of monoiodoacetate. Rats were placed
to opaque plastic chambers (22 cm in width x 17 cm in
length x 14 cm in height) for 10 min prior to the start of
the each experiment. This lets the animals accommodate
to their new environment before testing. Movable
infrared radiant heat source was placed directly under the
plantar surface of the hind paw and the time taken for
hind paw withdrawal was monitored. A cut-off time of
20 s was used in all experiments. Three tests were carried
out at 10 min intervals and then the mean value was taken
as the nociceptive threshold.
Following three baseline measurements, rats
received intraarticular injection of monoiodoacetate or
saline. In the defined times after injection of
monoiodoacetate or saline, paw withdrawal latencies
von Frey hairs
Tactile allodynia was measured with von Frey
hairs (Ugo Basile, Italy). Animals were placed into wire
mesh bottom cages and allowed to acclimatize prior the
start of the experiment. Tactile allodynia was tested by
single injection of
Expression and Production of COX-1,2 during Osteoarthritis 421
touching the plantar surface of the animal's hind paw with
von Frey hairs in ascending order of force until a paw
withdrawal response was elicited. Each von Frey hair was
applied to the paw for 5 s or until a response occurred.
Once a withdrawal response was established, the paw was
retested. The lowest amount of force required to elicit a
response was recorded as withdrawal threshold in grams.
Following three baseline measurements, rats
received intraarticular injection of monoiodoacetate or
saline. Paw withdrawal thresholds were measured in
defined times after the injection of monoiodoacetate or
In four different times after monoiodoacetate or
saline injection, animals (eight per group) were euthanized
in halothane anesthesia. Lumbar section of the spinal cord
was removed and given in RNAlater solution (Qiagen).
Disruption and homogenization of small parts of
the lumbar section of spinal cord weighing approximately
100 mg stabilized with the RNAlater (Qiagen) was
performed using Ultra-Turrax (Ika). Total RNA was
isolated with the RNeasy lipid tissue isolation kit
(Qiagen) according to the manufacturer’s instruction.
RNA integrity was determined by gel electrophoresis in
2 % agarose gel stained with ethidium bromide. The
purity of the RNA was assessed by the ratio of
absorbance at 260 nm and 280 nm. RNA was stored in
aliquots at –70 °C until used for reverse transcription.
RNA samples were reverse transcribed using RT
buffer, 25 mM MgCl2, 10 mM dNTPs (2.5 mM of each),
50 µM random hexamers, RNase inhibitor (20 U/µl) and
reverse transcriptase (50 U/µl), all from Applied
Biosystems. The mix was aliquoted into individual tubes
and RNA was added. Samples were incubated for 10 min
at 25 ºC, 30 min at 48 ºC and then for 5 min at 95 ºC.
A reaction mix for real-time PCR was made with
TaqMan Universal PCR master mix, water and Assays on
Demand gene expression products (all Applied
Reaction mix was aliquoted to the wells on a
real-time PCR plate. Each sample was made in duplicate.
A volume of 5 µl of cDNA was added to each well. A no-
template control contained water instead of cDNA. PCR
reaction was run on ABI PRISM 7300 (ABI PRISM 7300
SDS analytical cycler, Applied Biosystems) using
Expression of COX-1
normalized to RNA loading for each sample using the
β2-microglobulin as an internal standard. The quantity of
mRNA was given as 2-ΔΔct. ΔΔct was calculated as follows:
ΔΔct = Δct (gene of interest) – Δct (endogenous control).
For detection of antigens, sandwich enzyme-
linked immunosorbent assay (ELISA) was used.
Microtiter NUNC plates (Schoeller) were coated with
monoclonal antibody specific for cyclooxygenase 1
(Alpha Diagnostic) and for cyclooxygenase 2 (Kamiya
Biomedical Company) diluted in coating buffer and
After 24 h the plates were washed two times
with washing buffer PBS (phosphate buffered saline) and
two times with PBST (PBS containing 0.05 % Tween 20,
Sigma-Aldrich). Each well was then filled with PBST
and incubated for one hour at room temperature to
prevent non-specific adsorption of protein to the well
surfaces. During this time, samples from spinal cord of
monoiodoacetate-injected or control animals were
homogenized in 5 % FBS (fetal bovine serum, Sigma-
Aldrich) and then added to the wells. Two hours after this
incubation at room temperature, the plates were washed
two times with PBS and two times with PBST (PBS
containing 0.05 % Tween 20, Sigma-Aldrich).
The secondary biotinylated antibodies (Acris
Antibodies) were added to the wells and incubated for
next two hours. The plates were then washed, followed
by incubation of 1:1000 dilution of streptavidin
(Beckman Coulter) for 20 min. After washing the coated
well, TMB (tetramethylbenzidine, Sigma-Aldrich) and
citric buffer with peroxide were added.
For development of the color reaction the plates
were incubated in the dark and the reaction was stopped
by the addition of H2SO4 to each well. The color intensity
was determined at 450 nm on Multiscan RC reader.
Results are expressed as stimulation index (OD of
monoiodoacetate treated vs. OD of control animals).
Monoiodoacetate was purchased from Sigma-
Aldrich and halothane (Narcotan) was obtained from
and COX-2 was
422 Procházková et al.
All results are expressed as mean values
± S.E.M. Statistical analysis was carried out using two-
way repeated measures ANOVA with a post-hoc Student-
Newman-Keuls test in the case of repetitive testing of
paw withdrawal. P<0.05 was accepted as significant. The
evaluation of real-time PCR data was done by one-way
ANOVA with a post-hoc Turkey’s test using 2-ΔΔct values
of each samples. P<0.05 value was considered
significant. Data from ELISA method are presented as
stimulation index and were analyzed using unpaired
t-test. The results were considered significant if P value
was less than 0.05.
Effect of monoiodoacetate on paw withdrawal latency
Paw withdrawal latencies were measured
before and in defined times after application of
monoiodoacetate or saline. Intraarticular injection of
monoiodoacetate into the right knee joint produced
marked and significant reduction of paw withdrawal
latencies to noxious radiant heat stimuli. Decreased paw
withdrawal latencies were evident from the first day
following injection of monoiodoacetate, with the
maximum on day 5 after induction of osteoarthritis
(statistically significant at all observed times compared
to baseline and to control animals; P<0.001). The paw
withdrawal latencies remained decreased until day 31
Effect of monoiodoacetate on tactile allodynia
Tactile allodynia was measured with von Frey
hairs before and following intraarticular injection of
monoiodoacetate or saline.
monoiodoacetate into the right knee joint induced
marked allodynia. The onset of allodynia was evident
from the first day
monoiodoacetate. Figure 2 shows that tactile allodynia
in the monoiodoacetate-injected knee joints reaches the
maximum on day 5 and was observed throughout the
experiment (statistically significant at all observed times
compared to baseline and to control animals, P<0.001).
Expression of COX-1 and COX-2 mRNA in the spinal
cord after induction of osteoarthritis
Expression of COX isoenzymes was measured
at four different times. First day after monoiodoacetate
injection, spinal levels of COX-1 mRNA and COX-2
The injection of
following injection of
mRNA were moderately increased (2.3 and 2.6 times,
respectively; P<0.05). The expression of spinal COX-2
mRNA was much higher on day 5 (5.2 times; P<0.001)
and remained increased at this level until the day 31
(4.9 times; P<0.001). On the other hand, expression of
spinal COX-1 mRNA increased gradually during the
whole testing period reaching maximum on the day 31
(4.5 times; P<0.001) when the relative expression of
both genes was almost equal. All results are expressed
in comparison with control animals (Fig. 3).
Fig. 1. Paw withdrawal latencies to a noxious radiant heat
stimulus before and after the induction of osteoarthritis.
Osteoarthritis reduced paw withdrawal latencies in the right
(monoiodoacetate-injected) hind paws. Control animals exhibited
no differences in paw withdrawal latencies. Withdrawal latencies
of the left (uninfected) paws of osteoarthritis rats did not differ
from those of control. The data represent observations from
8 animals per group displayed as means ± S.E.M. (P<0.001 at all
observed times compared to baseline and to control animals).
Fig. 2. Paw withdrawal thresholds to von Frey hairs before and
after the induction of osteoarthritis. Paw withdrawal thresholds to
von Frey hairs were significantly decreased from the first day
after monoiodoacetate injection into the right knee joint. Control
and uninfected (left) paws were not altered during the whole
testing period. Results are expressed as median force in grams ±
S.E.M. required to induce paw withdrawal in 8 animals per group
(P<0.001 at all observed times compared to baseline and to
Expression and Production of COX-1,2 during Osteoarthritis 423
Production of COX-1 and COX-2 proteins after induction
The production of spinal COX proteins was
comparable to the expression results. The production of
COX-1 and COX-2 proteins was almost similar at the
beginning of the experiment (1.9 and 2.0 times,
respectively; P<0.001). The highest production of COX-2
protein was observed on day 5 after the induction of
osteoarthritis (increased 3.9 times; P<0.001). The levels
of COX-1 protein increased gradually from the 5th day
after induction of osteoarthritis (1.7 times; P<0.001) with
maximum on day 31 (3.4 times; P<0.001). Results are
expressed as stimulation index (OD of monoiodoacetate-
treated vs. OD of control animals) (Fig. 4).
measured by von Frey thresholds and paw withdrawal
latencies in the plantar test. The pain measures reached
maximum on the 5th day, then remained relatively stable.
Expression and production of spinal COX-2 was rapidly
increased in parallel to pain measures reaching maximum
on the 5th day and then remained relatively stable. On the
other hand, the expression and production of spinal
COX-1 mRNA increased at a slower pace but 31 days
after the induction of osteoarthritis there was almost no
difference between relative amount of COX-1 and
The present results indicate that osteoarthritis
pain has different patterns of expression of spinal mRNA
for COX-1 and COX-2 compared with chronic
inflammatory or postoperative pain. Expression and
production of spinal COX-2 were significantly increased
22 days after the induction of arthritis by CFA (Complete
Freund's Adjuvant) while spinal COX-1 mRNA and
protein levels remained unchanged at this time (Beiche et
al. 1996, Beiche et al. 1998). In contrast, COX-1 might
play a more important role in postoperative pain.
Expression of spinal COX-1 mRNA raised gradually after
rat paw incision and 6 hours after the surgery there was
no difference between relative amount of COX-1 and
COX-2 mRNA in the spinal cord (Procházková et al.
The increased expression and production of
spinal COX-1 in some types of pain may have therapeutic
implications. The lack of analgesic effect of intrathecally
administered selective inhibitor of COX-2 (NS-398) in
the model of postoperative pain induced by skin incision
suggest that spinal COX-2 might play a less important
role in this type of pain (Yamamoto and Sakashita 1999).
Intrathecal administration of COX-1 inhibitors, but not of
COX-2 inhibitor, dose-dependently reduced pain in the
model of postoperative pain (Zhu et al. 2003).
Perioperative intrathecal administration of COX-1
inhibitors (COX-1 preferring inhibitor ketorolac and
COX-1 selective inhibitor SC-560), but not of selective
COX-2 inhibitor (NS-398), reduced paw incision induced
hypersensitivity in the model of postoperative pain in rats
(Zhu et al. 2005). Intrathecal administration of selective
COX-1 inhibitor (SC-560), but not selective inhibitor of
COX-2 (NS-398), restores normal exploratory activity
(rearing behavior) after laparotomy (Martin et al. 2006).
The increased expression of spinal COX-1
mRNA found in the present study corroborates other
Monoiodoacetate injection produced pain as
Fig.3. Relative expression of cyclooxygenase 1 and cyclooxy-
genase 2 in lumbar section of spinal cord 1, 5, 14 and 31 days
after the induction of osteoarthritis (induced by injection of
monoiodoacetate) as determined by real-time PCR. Each column
represents observations from 8 animals displayed as means
± S.E.M. Asterisks indicate significant difference between
monoiodoacetate injected and control animals at respective time-
points (* P<0.05; ** P<0.01 and *** P<0.001).
Fig. 4. Production of cyclooxygenase 1 and cyclooxygenase 2
proteins in lumbar section of spinal cord 1, 5, 14 and 31 days
after induction of osteoarthritis (induced by injection of
monoiodoacetate) as determined by ELISA method. Results
(mean values ± S.E.M.) are expressed as stimulation index (OD
of monoiodoacetate treated vs. OD of control animals). Asterisks
indicate significant difference between production of COX
proteins in monoiodoacetate-injected and control animals at four
different times (*** P<0.001).
424 Procházková et al.
findings suggesting important role of this enzyme in
osteoarthritis. Knorth et al. (2004) demonstrated the
important role of synovial COX-1 in patients with
The expression of COX-1 and COX-2 mRNA
from synovial tissue from patients with osteoarthritis or
from patients with inflammatory arthritis (rheumatoid
arthritis) was compared by Siegle et al. (1998). The
expression of COX-2 was found to be elevated in both
groups of the patients, but the difference in expression of
both cyclooxygenases was more evident in the group of
patients with rheumatoid arthritis. The amount of COX-2
mRNA was significantly higher in patients with
rheumatoid arthritis compared with patients with
The expression of both, COX-1 and COX-2
mRNA, was detected in cells from synovial fluid of
patients with acute and chronic arthritis. COX-1 was the
BEICHE F, SCHEUERER S, BRUNE K, GEISSLINGER G, GOPPELT-STRUEBE M: Up-regulation of
cyclooxygenase-2 mRNA in the rat spinal cord following peripheral inflammation. FEBS Lett 390: 165-169,
BEICHE F, BRUNE K, GEISSLINGER G, GOPPELT-STRUEBE M: Expression of cyclooxygenase isoforms in the
rat spinal cord and their regulation during adjuvant-induced arthritis. Inflamm Res 47: 482-487, 1998.
HARGREAVES K, DUBNER R, BROWN F, FLORES C, JORIS J: A new and sensitive method for measuring
thermal nociception in cutaneous hyperalgesia. Pain 32: 77-88, 1988.
HAY CH, TREVETHICK MA, WHEELDON A, BOWERS JS, BELLEROCHE JS: The potential role of spinal cord
cyclooxygenase-2 in development of Freund´s complete adjuvant-induced changes in hyperalgesia and
allodynia. Neuroscience 78: 843-850, 1997.
HEFFERAN MP, O'RIELLY DD, LOOMIS CW: Inhibition of spinal prostaglandin synthesis early after L5/L6 nerve
ligation prevents the development of prostaglandin-dependent and prostaglandin-independent allodynia in the
rat. Anesthesiology 99: 1180-1188, 2003.
INIGUEZ MA, PABLOS JL, CARREIRA PE, CABRE F, GOMEZ-REINO JJ: Detection of COX-1 and COX-2
isoforms in synovial fluid cells from inflammatory joint diseases. Br J Rheumatol 37: 773-778, 1998.
KALBHEN DA: Chemical model of osteoarthritis – a pharmacological evaluation. J Rheumatol 14: 130-131, 1987.
KAUFMANN WE, ANDREASSON KL, ISAKSON PC, WORLEY PF: Cyclooxygenases and the central nervous
system. Prostaglandins 54: 601-624, 1997.
KNORTH H, DORFMULLER P, LEBERT R, SCHMIDT WE, WITTENBERG RH, HEUKAMP M, WIESE M,
WILLBURGER RE: Participation of cyclooxygenase-1 in prostaglandin E2 release from synovitis tissue in
primary osteoarthritis in vitro. Osteoarthritis Cartilage 12: 658-666, 2004.
MARTIN TJ, BUECHLER NL, EISENACH JC: Intrathecal administration of a cyclooxygenase-1, but not a
cyclooxygenase-2 inhibitor, reverses the effects of laparotomy on exploratory activity in rats. Anesth Analg
103: 690-695, 2006.
MATSUNAGA A, KAWAMOTO M, SHIRAISHI S, YASUDA T, KAJIYAMA S, KURITA S, YUGE O:
Intrathecally administered COX-2 but not COX-1 or COX-3 inhibitors attenuate streptozocin-induced
mechanical hyperalgesia in rats. Eur J Pharmacol 554: 12-17, 2007.
most abundant isoform with the strong immunostaining
observed in a special fraction of mononuclear cells
(Iniguez et al. 1998).
In summary, the present findings indicate that
not only expression and production of COX-2 but also
that of COX-1 is significantly increased in the spine
during osteoarthritis pain. In contrast to inflammatory
pain, the increased expression and production of spinal
COX-1 might play an important role in osteoarthritis
Conflict of Interest
There is no conflict of interest.
This work was supported by research grants VZ
2009 Download full-text
Expression and Production of COX-1,2 during Osteoarthritis 425
MCGEER PL, MCGEER EG: Inflammatory processes in amyotrophic lateral sclerosis. Muscle Nerve 26: 459-470,
PRITZKER KP: Animal models for osteoarthritis: processes, problems and prospects. Ann Rheum Dis 53: 406-420,
PROCHÁZKOVÁ M, DOLEŽAL T, SLÍVA J, KRŠIAK M: Different patterns of spinal cyclooxygenase-1 and
cyclooxygenase-2 mRNA expression in inflammatory and postoperative pain. Basic Clin Pharmacol Toxicol
99: 173-177, 2006.
RAMOS KM, JIANG Y, SVENSSON CI, CALCUTT NA: Pathogenesis of spinally mediated hyperalgesia in diabetes.
Diabetes 56: 1569-1576, 2007.
SIEGLE I, KLEIN T, BACKMAN JT, SAAL JG, NUSING RM, FRITZ P: Expression of cyclooxygenase 1 and
cyclooxygenase 2 in human synovial tissue: differential elevation of cyclooxygenase 2 in inflammatory joint
diseases. Arthritis Rheum 41: 122-129, 1998.
YAMAMOTO T, SAKASHITA Y: The role of the spinal opioid receptor like1 receptor, the NK-1 receptor, and
cyclooxygenase-2 in maintaining postoperative pain in the rat. Anesth Analg 89: 1203-1208, 1999.
ZHU X, CONKLIN D, EISENACH JC: Cyclooxygenase-1 in the spinal cord plays an important role in postoperative
pain. Pain 104: 15-23, 2003.
ZHU X, CONKLIN DR, EISENACH JC: Preoperative inhibition of cyclooxygenase-1 in the spinal cord reduces
postoperative pain. Anesth Analg 100: 1390-3, 2005.
ZHU X, EISENACH JC: Cyclooxygenase-1 in the spinal cord is altered after peripheral nerve injury. Anesthesiology
99: 1175-1179, 2003.
ZIMMERMANN M: Ethical guidelines for investigations of experimental pain in conscious animals. Pain 16: 109-110,