Chapter 6

Locomotor Activity in a Novel Environment

as a Test of Inflammatory Pain in Rats
David J. Matson, Daniel C. Broom, and Daniel N. Cortright

Abstract
Creating a robust and unbiased assay for the study of current and novel analgesics has been a daunting
task. Traditional rodent models of pain and inflammation typically rely on a negative reaction to various
forms of evoked stimuli to elicit a pain response and are subject to rater interpretation. Recently, models
such as weight bearing and gait analysis have been developed to address these drawbacks while detecting
a drug’s analgesic properties. We have recently developed the Reduction of Spontaneous Activity by
Adjuvant (RSAA) model as a quick, unbiased method for the testing of potential analgesics. Rats, following
prior administration of an activity-decreasing inflammatory insult, will positively increase spontaneous
locomotor exploration when given single doses of known analgesics. The RSAA model capitalizes on a
rat’s spontaneous exploratory behavior in a novel environment with the aid of computer tracking software
to quantify movement and eliminate rater bias.

Key words:  Pain, Inflammation, Behavior, Spontaneous activity, Complete Freund’s adjuvant

1. Introduction

Valid preclinical evaluation of analgesic compounds has historically

been challenging to achieve. Numerous models and behavioral
tests for evaluating pain in rodents have been established and used
widely for the testing of potential analgesics (1). The most
commonly used tests involve evoking a “pain-like” nociceptive
and/or aversive behavior and evaluating the effect of commonly
used and potential analgesics to reverse this behavior. Such testing
can be performed in rodents after the induction of sensory
hypersensitivity (allodynia or hyperalgesia) by nerve injury or
inflammation. Tests include measuring mechanical allodynia
with von Frey (Semmes–Weinstein) monofilaments (2), mechanical

Arpad Szallasi (ed.), Analgesia: Methods and Protocols, Methods in Molecular Biology, vol. 617,
DOI 10.1007/978-1-60327-323-7_6, © Springer Science + Business Media, LLC 2010

67
68 Matson, Broom, and Cortright

hyperalgesia with Randall–Selitto paw pressure apparatus (3), and

thermal hyperalgesia with radiant heat (4).
Current behavioral endpoints present significant challenges
for the development of novel analgesics. Firstly, they all to some
degree rely on the subjectivity of the rater performing the test.
For example, what is classified as a response to the mechanical or
thermal stimulus being applied? This can be overcome somewhat
by blinding the rater to the identity of each individual animal’s
treatment and randomizing the treatment groups throughout
the experiment. Secondly, as these assays rely on an evoked input
resulting in a behavioral outcome, test subjects require habituation
and baseline testing to allow the animal to become accustomed to
the testing procedure. Such activities often require time-consuming
handling and baseline testing sessions resulting in less than optimal
throughput if used in a drug screening paradigm. Thirdly, as these
assays measure an evoked physical response (i.e. a nocifensive
withdrawal of the hindpaw) and therefore an increase in behav-
ioral responding, effects of test drugs on physical activity may
produce changes in responding in the absence of analgesic
activity (5). This is of particular concern as certain classes of
analgesics, notably the opiates, affect locomotor activity in rodents.
Indeed, when measuring a pain-stimulated response (acetic
acid-induced abdominal stretching) and a pain-suppressed behavior
(decreased consumption of a liquid), Stevenson and colleagues
found a false positive effect of haloperidol, an antipsychotic agent,
upon acetic acid induced stretching, but not upon the decreased
consumption of a liquid (6). This result elegantly demonstrates
the potential for falsely identifying locomotor modifying agents
as potential analgesics when measuring pain-stimulating behavior
and suggests a potential benefit from using pain-suppressing
behavior as a measure of potential analgesia. Finally, most standard
models of pain-like behavior, like many behavioral assays, act as
surrogates for the clinical situation and demonstrate analgesia for
currently known active mechanisms. The effect of compounds
that act on novel mechanisms may be less clear in these assays
and there is a chance that a novel compound may therefore be
erroneously labeled as an analgesic when it is not or as a nonanal-
gesic when it actually is. This may partly stem from the problems
of identifying an emotional endpoint in rodents. As clinical
pain has a large subjective, emotional component, the testing
of potentially novel analgesics by using an evoked, quantitative
endpoint may not truly mirror the clinical response and therefore,
may not be truly indicative of the analgesic potential of the specific
compound. As stated above, these tests demonstrate analgesia
with commonly used analgesics such as the opiates, nonsteroidal
antiinflammatory drugs (NSAIDs), and anticonvulsants (gabapentin,
pregabalin), all of which provide strong analgesic responses as
measured by stimulus-evoked endpoints. However, the potential
 Locomotor Activity in a Novel Environment 69

to falsely identify or completely miss new classes of analgesics

remains. The addition of new assays to the existing standards may
help minimize this risk.
In an attempt to address some of these issues, a variety of
new models for evaluating pain-like behavior and the analgesic
properties of compounds have been developed (7). Some models
have evaluated unevoked measures of behavior, thus eliminating
the necessity of a stimulus that may otherwise produce a potentially
artificial behavior from the animal in order to obtain a quantifiable
response. These tests include measures of weight bearing on
injured compared to uninjured paws (8, 9). This assay involves
the placement of each of the animal’s hindpaws on a separate
weight-sensitive pad that measures the weight the animal is placing
on each hindpaw. Although useful and not directly evoked, it
does require extensive habituation and baseline testing. This test
has been pharmacologically validated with current analgesics
providing reproducible effects and is therefore a valuable addition
to the more traditional tests (9, 10). Gait analysis has also been used
for measuring hypersensitivity and certain parameters of an
animal’s gait are modified in both neuropathic and inflammatory
pain-like states in rats (11–15). Interestingly, the results of this test
correlate with the results from the von Frey mechanical allodynia
assay (15) suggesting great potential for the measurement of pain.
However, to date, only 1–2 pharmacological agents have been
evaluated in gait analysis paradigms. The lack of experimental results
with commonly used analgesic agents in this type of test limits the
utility of gait analysis in identifying analgesic drugs (11, 16).
We developed the reduction in spontaneous activity due to
adjuvant (RSAA) assay (17) in an attempt to address the issues of
using an evoked response. We aimed to utilize the spontaneous,
unevoked and natural exploratory behavior of rats when placed in
a novel environment as a way to measure the “comfort” of the rat
in using injured hindlimbs to explore the novel environment.
To do this, we used standard measures of locomotor activity.

2. Materials

2.1. Complete Freund’s 1. Complete Freund’s Adjuvant (CFA) (Sigma, St. Louis, MO)
Adjuvant (CFA) stored at 4°C prior to use.
Procedure 2. 0.9% sterile saline.
3. Male Sprague Dawley rats (Charles River, Kingston, NY)
weighing 320 g at time of CFA injection.
4. One ml BD tuberculin syringes.
5. ½ inch 26-G PrecisionGlide beveled needles.
6. 2″ square gauze.
70 Matson, Broom, and Cortright

7. Towel or pad.
8. Isoflurane anesthesia vaporizer with nose cone and induction
box (Vetequip, Pleasanton, CA).
9. Solid bottom cages with bedding.

2.2. Collagen-Induced 1. Highly purified type II Porcine Collagen (18) dissolved at

Arthritis (CIA) 2 mg/ml in 0.05 M acetic acid. Store at 4°C in the dark for
Procedure less than 1 week.
2. Incomplete Freund’s Adjuvant (IFA) (Difco, Detroit, MI).
3. 0.9% Sterile saline.
4. Female Lewis rats (Harlan, Indianapolis, IN) weighing 150 g
at time of first collagen injection.
5. 1 ml Hamilton syringes with luer lock tip.
6. ½ inch 26-G PrecisionGlide beveled needles.
7. 2″ square gauze.
8. Rodent restrainer.
9. Solid bottom cages with bedding.
10. Homogenizer (diameter: 5 mm or less).
11. 20 ml glass vials.
12. 1 ml BD tuberculin syringes.
13. 5 ml BD tuberculin syringes.
14. 1½ inch 20-G beveled needles.
15. Crushed ice.

2.3. Test Equipment 1. Digiscan-16 Animal activity monitor (model 1300JC/

CCDigi, version 2.4, Omnitech Electronics, Columbus, OH).
The Plexiglas testing box measures (41.25 × 41.25 × 30 cm).
Each box should have at least 48 infrared photocell emitters
and detectors (2.5 cm between sensors). 32 sensors should
surround the perimeter of the box at the base to capture
horizontal movement. 16 separate sensors should be set
13.5 cm above the floor to capture vertical rearing.
2. Cage bedding (thin layer on testing cage floor).
3. Versamax software version 4.0-125E (Accuscan Istruments,
Columbus, OH) for measurement recording and analysis.
4. White noise generator capable of emitting 62 dB
5. Red light (60 W).
6. Testing room. This needs to be able to be completely shut off
from other rooms in the facility. Ideally, it should have sound
dampening materials on any common walls to prevent any
inadvertent distractions during the test. All lights need to be
able to be shut off.
 Locomotor Activity in a Novel Environment 71

3. Methods

3.1. CFA Injection 1. Remove CFA from the refrigerator 30 min prior to injection
Procedure time and allow warming to room temperature.
2. Set up the anesthesia apparatus and charge the induction
chamber (preferably in a fume hood or area with good
ventilation). Set the vaporizer to between 3 and 4  L per
minute with 5% isoflurane (see Note 1).
3. Mix the CFA to form a uniform suspension of the heat-
killed mycobacteria. Be sure not to incorporate air into
the adjuvant.
4. In the 1 ml syringe, draw up the CFA. Do not fill the syringe
past 700 ml (see Note 2). Wipe excess CFA from the outside
of the syringe with gauze.
5. Attach the ½″ 26G needle and remove any air.
6. Place a single animal in the induction chamber and monitor
status until breathing is relaxed and even.
7. Once the animal is under anesthesia, move it onto a towel and
attach the nose cone. The animal should be on its back with
the head pointed to the right (for a right handed person).
8. Place a second animal in the induction chamber. Carefully
monitor animal breathing to be sure that the animal is not in
danger of respiratory distress. Animals should not be left in
the induction chamber for any longer than 5 min.
9. Check the animal to make sure it is properly anesthetized
by pinching the tail and looking for the loss of the reflex
response.
10. With the middle finger and lower part of the thumb on the
left hand, hold the lower leg and flex the tibia-femur joint to
form a 120° angle.
11. With the pointer finger and the fingertip of the thumb on the
left hand, hold the sides of the tibia-femur joint to stabilize
the leg.
12. When the joint is flexed, there will be a small indentation on
either side of the midline. The raised section on the midline is
the ligament of the patella. Avoid damaging this ligament.
13. With the right hand, slide the needle between the bones into
the tibia-femur joint space. You will feel the needle “pop”
into the joint cavity. Be careful not to puncture through the
other side of the joint space (see Notes 3 and 4).
14. Slowly inject 50  ml (50  mg) of CFA or saline into the joint
space (see Notes 5). Use as little pressure on the syringe as
possible (see Note 6).
72 Matson, Broom, and Cortright

15. Remove the needle and apply slight pressure to the injection
site to help minimize any bleeding.
16. Repeat the procedure on the other knee.
17. Return the animal to its home cage and allow to it to fully
waken.
18. Repeat steps 7–18 until all animals are injected.
19. Allow animals to develop inflammation for 48  h prior to
testing.

3.2. CIA Injection 1. Remove the type II collagen from the freezer and allow
Procedure warming to room temperature.
(Slightly Modified 2. Place a 20 ml vial in the ice bath.
from the Chondrex Inc.
3. Add 5 ml of IFA to the vial.
Protocol (18))
4. Place the homogenizer probe into the IFA and mix at a slow
speed.
5. Draw up 5 ml of the type II collagen with the 5 ml syringe
with the 20G needle (see Note 7).
6. Start adding the Collagen to the IFA drop wise while mixing.
Increase the mixing speed slightly while continuing to add
the Collagen due to thickening of the emulsion.
7. Once all of the collagen has been added, slowly increase the
speed to 30,000 rpm. Mix at that speed for 3 min. The end
result will be a thick emulsion.
8. Test the stability of the emulsion. Partially fill a second 20 ml
vial with approximately 10 ml of water. Take a 1 ml syringe
and draw up a small amount of the emulsion. Place one drop
of the emulsion on top of the water. A stable emulsion
will remain as a solid drop on the surface of the water.
If the emulsion breaks down, mix in a few drops of IFA
and repeat.
9. Keep the emulsion on ice until you are ready to inject.
10. Draw up 0.8 ml of the emulsion into the Hamilton syringe.
Be sure to remove any air from the syringe. Cap with a 26G
needle.
11. Place an animal in the restrainer with the tail exposed.
12. Inject 0.4 ml of the emulsion (400 mg of collagen) subcuta-
neously in the tail (see Note 8). This should be about 2–3 cm
from the base of tail. Hold the needle level with the tail, bevel
up. After injection, apply slight pressure to the injection site
to avoid leakage.
13. Place animal back in its home cage and return to the vivarium
for 7 days.
 Locomotor Activity in a Novel Environment 73

14. On day 7 after initial injection, repeat steps 1–10 to make

new collagen for the booster injection.
15. Give each animal a booster injection of the emulsion. Inject
0.2 ml of the emulsion (200 mg of collagen) subcutaneously
in the tail. This should also be about 2–3 cm from the base
of tail.
16. Return animals to their home cages and the vivarium for 14
days.

3.3. Test Procedure 1. Turn on white noise in testing room.

(48 h Post-CFA, 21 2. Turn on red light and turn off all other lights in the testing
days Post-CIA) room.
3. Set up the Versamax software for the first run (see Note 9).
At the very least specify the animal number. Set the total
collection time to 15  min (see Note 10). The overall test
session can be further broken down into bins if desired, but
this is not necessary.
4. Scatter a thin layer of bedding on the bottom of the testing
box and check to be sure the boxes are clean.
5. Transport the animals from the viviarium to a dosing room.
This should be a separate room from the testing room. Allow at
least 30 min to acclimate to the dosing room before testing.
6. Pretreat the animals with any desired test compound in the
dosing room.
7. After the pretreatment time has elapsed, move the animals to
the testing room and place them in their assigned testing box,
one per box. It is permissible to test multiple animals in the
same room at the same time.
8. After placing the animals in the box, flip the start switch to
run on the front of the testing box to begin the test.
9. Leave the room while the test is being conducted.
10. After the 15 min test session is over, reenter the room and flip
the start switches to off. Remove the animals from the testing
room.
11. Reset the computer and proceed with any additional runs as
above.
12. After the last run, remove the bedding from the testing boxes
and replace with fresh bedding.
13. Data is typically presented as means with standard errors of
measure. Overall analysis of the data determined using
one-way analysis of variance. Follow-up post hoc tests are
performed using Fisher’s least significant difference with
p £ 0.05 (see Note 11).
74 Matson, Broom, and Cortright

4. Notes

1. The gas is set high to quickly put the animals under anesthesia.
Once the injection procedure is mastered, animals will only
be exposed to Isoflurane inhalation for less than 1 min. If the
animal appears to be under anesthesia too deeply, turn down
the Isoflurane to 3%. At Isoflurane levels under 2% the animal
might not be deep enough under anesthesia and regain its
reflex response to the injection.
2. With greater volumes, the syringe tends to jam or the plunger
becomes hard to move due to the adjuvant. For the same
reason, a new syringe should be used for each withdrawal.
Only fill one syringe at a time. If syringes are preloaded, the
heat-killed mycobacteria in the adjuvant will settle and result
in a nonuniform suspension. Be sure to mix the CFA stock
suspension before each withdrawal.
3. The entire ½″ needle does not need to be inserted for the tip
of the needle to be within the intra-articular space.
4. Finding the appropriate injection site in the knee can be
difficult. Needle insertion too far proximal or distal will result in
hitting bones. This should be avoided as it can cause unwanted
damage to the joint. The midline should also be avoided to
limit damage to the ligaments as much as possible. Hold the
needle perpendicular to the injection site and insert slowly. If any
resistance is felt, pull back and move slightly toward the center
of the joint. When in the correct location, the needle will
slide easily into the joint space. If uncertain that the needle is
in the correct place, slightly moving the needle in the correct
position will move the joint naturally. Do not move the needle
more than necessary as joint damage may occur.
5. The joint space can easily accommodate 50 ml of volume.
Larger volumes are possible, if desired, but do not exceed
150 ml.
6. Filling the syringe takes time and can be messy as the adjuvant
is “sticky”. For this reason, the syringe is preloaded to inject
seven animals. If too much force is used on the plunger, the
remaining CFA in the syringe will become compressed and
obstruct the end of the syringe. Alternatively, occasionally a
particle of the heat-killed mycobacteria will become stuck in
the needle. If this occurs, gently pull back slightly on the plunger
to move the orientation of the particle. Usually, this slight
movement will allow the particle to pass thought the needle.
Do not try to force the particle through the needle with more
pressure as this might result in the needle separating from
the syringe or clumping of the remaining heat-killed myco-
bacteria in the syringe.
 Locomotor Activity in a Novel Environment 75

7. This amount of collagen is used to dose around 40 animals.

It is necessary to make a good deal more collagen emulsion
than is needed. A lot of emulsion is lost on the outside of the
syringe when wiping it clean after filling. There is also some
emulsion that will not be able to be drawn into the syringe
due to the emulsion sticking to the sides of the vial.
8. The original protocol from Chondrex Inc. recommends
injection of 200 mg of collagen. There can be a large variation
in CIA development between rat stains. This variation is also
apparent in the same strain from different vendors. Under the
conditions in our lab, we found that our suggested volume of
collagen emulsion was needed to produce the expected level
of arthritis.
9. There are many different companies that make locomotor
tracking systems for rodents. The three measures most critical
to the RSAA test are vertical rearing, movement time, and
total distance traveled. Most systems do an adequate job
tracking the horizontal movements and movement time of
rodents. Vertical rearing is captured by the photocells above
the floor of the testing box. On many systems, the height
these photocells are set at are adjustable. The height that we
used in our testing boxes gave us reproducible results on
vertical rearing between studies.
10. Test sessions longer that 15  min are possible. The bulk of
rodent exploratory behavior in a novel open field occurs in
the first 15 min after test onset. Approximately after 30 min,
the level of exploration of a naïve animal is greatly reduced
when compared with the beginning of the test. Any benefit of
exploration beyond 30 min is offset by inadvertent noise and
variability in the data.
11. A typical CFA dose response curve is shown in Fig. 1. CFA
administration greater than 50  mg/bilaterally does not sig-
nificantly increase the therapeutic window for effect and
might increase the dose of analgesic needed to see an analgesic
result. At 50 mg of CFA bilaterally, morphine at 1 mg/kg and
ibuprofen (IB) at 10 mg/kg will reliably attenuate nocicep-
tive behaviors (see Fig. 2a). At doses of morphine greater than
1 mg/kg, its analgesic effect is reduced in association with
an increasingly sedative profile (see Fig. 2b). Morphine and
IB were repeatedly used as positive controls in every study.
A positive control group run with each study is recommended
to ensure that the model is working robustly and helps clarify
any negative result of novel compounds. The RSAA will also
discriminate against compounds that are excitatory in nature.
Amphetamine will increase the locomotor behavior in
measures of horizontal activity, but has no significant effect
on vertical rearing (see Fig. 3a and b).
76

Fig. 1. Examples of the effect of bilateral knee injections of CFA on spontaneous locomo-
tor activity in rats. CFA doses of 50 mg/50 ml/bilaterally and higher produced a signifi-
cant reduction in rears compared with the noninjected group or the 150  mg/150  ml/
unilateral group. Bars, mean rears ± S.E.M. (n = 10/group). *, p < 0.05; **, p < 0.01; or ***,
p < 0.001 compared with the saline or noninjected groups. ###, p < 0.001 compared
with the 150 mg of CFA unilateral injection group. Methods and full results are presented
in Matson et al., 2007 (17). Figure reproduced from Matson et al., 2007 (17) with kind
permission from the publisher

Fig. 2. (a) Effect of morphine on vertical activity in the RSAA model. Morphine dosed s.c.
at 10 mg/kg was not significantly different from the CFA/vehicle group. The CFA/IB group
was used as a positive control. (b) Morphine administered s.c. in a standard model of
locomotor activity in untreated rats. Bars, mean rears ± S.E.M. (n = 10/group). *, p < 0.05;
**, p < 0.01; or ***, p < 0.001 compared with the noninjected/vehicle or vehicle group.
+++, p < 0.001 compared with the CFA/vehicle group. Methods and full results are pre-
sented in Matson et al., 2007 (17). Figure reproduced from Matson et al., 2007 (17) with
kind permission from the publisher
 Locomotor Activity in a Novel Environment 77

Fig. 3. The effect of amphetamine on vertical activity (a) and total distance (b) in the
RSAA model. Amphetamine, administered i.p., did not increase vertical rearing above
CFA/vehicle levels. Total distance was significantly increased over CFA/vehicle levels.
Bars, mean rears or total distance traveled ± S.E.M. (n = 10/group). ***, p < 0.001 com-
pared with the noninjected group. +, p < 0.05 or ++, p < 0.01 compared with the
CFA/vehicle group. Methods and full results are presented in Matson et al., 2007 (17).
Figure reproduced from Matson et  al., 2007 (17) with kind permission from the
publisher

References

1. Le Bars D, Gozariu M, Cadden SW (2001) assessment of candidate analgesic drugs: recent

Animal models of nociception. Pharmacol
Rev 53:597–642
2. Chaplan SR, Bach FW, Pogrel JW, Chung JM,
Yaksh TL (1994) Quantitative assessment of
tactile allodynia in the rat paw. J Neurosci
Methods 53:55–63
3. Randal L, Selitto J (1957) A method for
measurement of analgesic activity on inflamed
tissue. Arch Int Pharmacodyn Ther 111:
409–419
4. Hargreaves K, Dubner R, Brown F, Flores C,
Joris J (1988) A new and sensitive method for
measuring thermal nociception in cutaneous
hyperalgesia. Pain 32:77–88
5. Negus SS, Vanderah TW, Brandt MR, Bilsky
MR, Becerra L, Borsook D (2006) Preclinical
advances and future challenges. J Pharmacol
Exp Ther 319:507–514
6. Stevenson GW, Bilsky EJ, Negus SS (2006)
Targeting pain-suppressed behaviors in pre-
clinical assays of pain and analgesia: effects of
morphine on acetic acid-suppressed feeding in
C57BL/6J mice. J Pain 7:408–416
7. Cortright DN, Matson DJ, Broom DC (2008)
Assessing pain and analgesia in laboratory
animals – new methods in drug discovery.
Expert Opin Drug Discov 3(9):1099–1108
8. Whiteside GT, Harrison J, Boulet J, Mark L,
Pearson M, Gottshall S, Walker K (2004)
Pharmacological characterization of a rat
model of incisional pain. Br J Pharmacol
141:85–91
78 Matson, Broom, and Cortright
9. Pomonis JD, Boulet JM, Gottshall SL, Phillips
S, Sellers R, Bunton T, Walker K (2005)
Development and pharmacological character-
ization of a rat model of osteoarthritis pain.
Pain 114:339–346
10. Wilson AW, Medhurst SJ, Dixon CI, Bontoft
NC, Winyard LA, Brackenborough KT, De
Alba J, Clarke CJ, Gunthorpe MJ, Hicks GA,
Bountra C, McQueen DS, Chessel IP (2006)
An animal model of chronic inflammatory
pain: pharmacological and temporal differen-
tiation from acute models. Eur J Pain 10(6):
1099–1108
11. Clarke KA, Heitmeyer SA, Smith AG, Taiwo
YO (1997) Gait analysis in a rat model of
osteoarthrosis. Physiol Behav 62:951–954
12. Hamers FP, Lankhorst AJ, van Laar TJ,
Veldhuis WB, Gispen WH (2001) Automated
quantitative gait analysis during overground
locomotion in the rat: its application to
spinal cord contusion and transection injuries.
J Neurotrauma 18:187–201
13. Coulthard P, Pleuvry BJ, Brewster M, Wilson
KL, Macfarlane TV (2002) Gait analysis as an
objective measure in a chronic pain model.
J Neurosci Methods 116:197–213
14. Coulthard P, Simjee SU, Pleuvry BJ (2003)
Gait analysis as a correlate of pain induced by
carrageenan intraplantar injection. J Neurosci
Methods 128:95–102
15. Vrinten DH, Hamers FF (2003) ‘Catwalk’
automated quantitative gait analysis as a novel
method to assess mechanical allodynia in the
rat; a comparison with von Frey testing. Pain
102:203–209
16. Simjee SU, Pleuvry BJ, Coulthard P (2004)
Modulation of the gait deficit in arthritic rats
by infusions of muscimol and bicuculline.
Pain 109:453–460
17. Matson DJ, Broom DC, Carson SR,
Baldassari J, Kehne J, Cortright DN (2007)
Inflammation-induced reduction of sponta-
neous activity by adjuvant: a novel model to
study the effect of analgesics in rats. J Pharmacol
Exp Ther 320:194–201
18. Chondrex, Inc. Protocol for the successful
induction of Collagen-Induced Arthritis in Rats,
V1.2 (2006) Chondrex, Inc., Redmond, WA