|Home | About | Journals | Submit | Contact Us | Français|
Chronic musculoskeletal pain is a major clinical problem and there is a general lack of animal models to study this condition. Carrageenan is commonly used to produce short-lasting acute inflammation and hyperalgesia in animal models. However, the potential of carrageenan to produce chronic, long-lasting hyperalgesia has not been evaluated. In the present study, we investigated the long-term effects of carrageenan injected into joint or muscle in rats. Rats were injected with 0.3, 1 or 3% carrageenan in one knee joint or gastrocnemius muscle and hyperalgesia to mechanical (measured as decreased withdrawal threshold) and heat (measured as decreased withdrawal latency) stimuli of both paws assessed before and at varying times after injection, through 8 weeks. Histological changes were examined only after injection of 3% carrageenan. Three percent carrageenan injected in the muscle or knee produced hyperalgesia to mechanical and heat stimuli ipsilaterally, which lasted 7–8 weeks and spread to the contralateral side 1–2 weeks after injection. One percent carrageenan injected to the knee joint or gastrocnemius muscle, produced hyperalgesia that was shorter-lasting and remained ipsilateral; 0.3% carrageenan injected into the knee joint or gastrocnemius muscle had no effect. Three percent carrageenan injected into the skin surrounding the knee joint did not produce hyperalgesia. A similar pattern of inflammatory changes was observed histologically for both the joint and muscle tissues. Acute inflammation was observed for the first 24 h with edema and neutrophilic infiltration evident as early as 4 h. At 1 week, the inflammation converted to primarily a macrophage response with scattered mast cells. The data suggest that animals injected with 1 or 3% carrageenan in the knee joint or gastrocnemius muscle could be used as models of acute inflammation through 24 h and chronic inflammation after 1 week. Furthermore, 3% carrageenan injected into deep tissues produces hyperalgesia that spreads to the contralateral side, at the same time period as the inflammation transforms from acute to chronic.
Chronic pain due to arthritis or neuropathy is a major clinical problem globally. To study the underlying pathological mechanisms of chronic pain, a number of good experimental animal models have been developed over the years (De Castro Costa et al., 1981; Bendele et al., 1999; Kehl et al., 2000; Sluka et al., 2001). However, there is a general lack of experimental models for chronic musculoskeletal pain. Most of the animal models available to study musculoskeletal pain are of acute nature and the chronic studies carried out to date do not extend for more than 3 or 4 weeks. Acute, short-lasting hyperalgesia models may not share the same mechanisms of chronic pain and may not be suitable to study the chronic pain mechanisms.
Complete Freund’s adjuvant (CFA), carrageenan (alone or in combination with kaolin), zymosan or urate crystals are used to induce experimental inflammation/arthritis and chronic pain in animal models (Pearson and Wood, 1959; Winter et al., 1962; Di Rosa et al., 1971; Keystone et al., 1977; Coderre and Wall, 1987; Decaris et al., 1999). CFA normally consists of heat-killed bacteria, Mycobacterium butyricum or Mycobacterium tuberculosis, as an emulsion in sterile mineral oil (Pearson and Wood, 1958; Swingle and Grant, 1977). CFA causes systemic effects like febrile response and inflammation at distant locations due to their systemic spread (Philippe et al., 1997; Deacris et al., 1999). Therefore, their use may not be appropriate to study-specific mechanisms, especially the neuronal mechanisms, of pain induction and/or maintenance. CFA (Haak et al., 1996) and zymosan, a glycan derived from yeast cell wall (Keystone et al., 1977), produce an immune-mediated chronic inflammation whereas agents like carrageenan, probably, produce a non-immune-mediated inflammation (Di Rosa et al., 1971; Moreno, 1993; Guthrie et al., 1996). Also, the severity of CFA-induced arthritis is strain specific (Swingle et al., 1969; Muir and Dumonde, 1982; Crowe et al., 1985). The potential of carrageenan in inducing chronic hyperalgesia has not been evaluated so far. Lambda carrageenan (type IV) is a water-extractable polysaccharide derived from marine plants, Gigartina aciculaire and Gigartina pistillata (Sigma Chemical Co. Catalogue). It causes inflammation and hyperalgesia when injected into tissues but not on topical application or ingestion in reasonable quantities (Nicklin and Miller, 1984). The mechanisms of carrageenan-induced acute inflammation and the subsequent hyperalgesia have been studied extensively (for review, see Schaible and Grubb, 1993; Mense, 1993). The acute inflammatory process is characterized by the accumulation of neutrophils in the perivascular space (Diehl et al., 1988), which is accompanied by the local release of noxious chemicals such as glutamate, prostaglandins, histamine and serotonin (Neil et al., 1987; Guilbaud et al., 1989; Nantel et al., 1999; Lawand et al., 2000; Hong et al., 2002). These noxious chemicals sensitize primary afferents resulting in primary hyperalgesia (Berberich et al., 1988; Diehl et al., 1988; Schaible and Schmidt, 1985, 1988; Hargreaves et al., 1988; Kehl et al., 2000).
Injection of carrageenan into deep tissues activates dorsal horn neurons causing central sensitization either spinally or supraspinally (Schaible et al., 1987; Neugebauer and Schaible, 1990; Dougherty and Willis, 1992; Hoheisel et al., 1993, 1995; Urban et al., 1999). The central sensitization, together with the increased sensitivity of the peripheral nociceptors, is manifested as secondary hyperalgesia (Sluka and Westlund, 1993). Secondary hyperalgesia is usually observed in the areas adjacent to the injury and sometimes in distal locations, which could be due to neuronal changes occurring spinally or supraspinally (Coderre and Melzack, 1985; Urban et al., 1999; Chen et al., 2000, Sluka et al., 2001; Sluka, 2002).
The mechanisms of pain and hyperalgesia induced by injecting inflammatory agents into different tissues like muscle, joint and skin are thought to be different (Sluka, 2002; reviewed by Mense, 1993 and Schaible and Grubb, 1993). The afferent innervation of the spinal cord from skin and deep tissues like muscles and joints are also different (Mense and Craig, 1988; Craig et al., 1988; Willis and Coggeshall, 1991). It has been reported that stimulation of C-fibers from a muscle nerve causes a long-lasting enhancement of the ventral root reflex compared to C-fibers from a cutaneous nerve (Wall and Woolf, 1984). Stimulation of deep tissues like muscle or knee joint with capsaicin produces a longer-lasting hyperalgesia compared to the stimulation of cutaneous tissue (Sluka, 2002).
With the above observations in mind, the potential of lambda carrageenan to induce chronic hyperalgesia, after injection to different tissues viz. muscle, joint and skin, was investigated in the present study using behavioral testing. Since histological changes are known to occur during chronic inflammation, a histopathological study of the tissues, parallel to the behavioral studies were also carried out.
Sprague—Dawley rats (Harlan, USA), weighing 225–300 g, kept at 12 h dark—light cycle with free access to standard rat chow and water, were used for the experiments. Animals were brought to the behavioral testing room the day before, to acclimatize them to the testing environment. Behavioral tests were usually done between 9 a.m. and 2 p.m. except for the 8 h testing after inflammation, which was done before 5 p.m. All experiments were approved by University of Iowa Animal Care and Use Committee and were carried out according to the guidelines of the International Association for the Study of Pain and National Institute of Health (Zimmermann, 1983).
Animals were kept in plexi-glass restrainers on an elevated platform with a clear glass top for about 30 min for acclimatization. A high-intensity radiant heat source was used as the stimulus. The heat source was positioned on the plantar skin of the hind limb and the beam was switched on, simultaneously starting a built-in timer. When the animal withdrew the paw abruptly to heat stimulus, the heat source and timer were stopped. The duration in seconds from the start of heat application to paw withdrawal was taken as the paw withdrawal latency (PWL). PWLs were determined five times bilaterally, with an interval of 5 min between each test and the mean of five readings was taken as the PWL for a particular time point. The intensity of the heat source was kept constant in all experiments with the aid of a constant voltage-power supply. This method of testing results in an increase in the paw temperature of the animal until it withdraws the paw and the voltage was set to produce an intensity required to obtain a baseline response time between 12 and 16 s. Cut-off time was set to 30 s to minimize heat damage to the skin. The validity and reliability of this testing method was previously established (Hargreaves et al., 1988; Sluka et al., 1999). A decrease in withdrawal latency is interpreted as heat hyperalgesia for the purpose of this study.
Animals were kept in a plexi-glass restrainer on an elevated platform with a mesh wire top. Threshold to mechanical stimuli was tested using von Frey filaments with increasing bending force as described elsewhere (Sluka et al., 2001). Briefly, the filament with the lowest threshold was applied to the plantar surface of the hind limb two times and observed for a withdrawal. If there was no response, the next higher force filament was tested. The value of the lowest force filament causing a withdrawal of the paw was taken as the mechanical threshold. One hundred and sixty-two millinewton was set as the cut-off. The following bending forces were assessed: 8, 12, 16, 32, 44, 56, 75, 104, 162 mN. The reliability of this testing method was previously established (Gopalkrishnan and Sluka, 2000). A decrease in withdrawal threshold is interpreted as mechanical hyperalgesia in this study.
Seven groups of animals were used for the behavioral testing. Three groups were given a single injection of 0.3 (n = 6), 1 (n = 6) or 3% (n = 12) lambda carrageenan (Type IV, 100 μl, dissolved in sterile saline, Sigma Chemical Company, St. Louis, USA) in the left knee joint anteriorly. The other three groups were given a single injection of 0.3 (n = 6), 1 (n = 6) or 3% (n = 12) carrageenan (100 μl) in the left gastrocnemius muscle belly. Another group (n = 6) was given a single injection of 3% carrageenan in the skin surrounding the anterior aspect of the knee. All carrageenan injections were given under halothane anesthesia (2–4% v/v in oxygen). PWLs to heat and mechanical withdrawal thresholds were recorded in all groups, except the skin-injected group, before and 4, 8, 24 h, 1, 2, 3, 4, 5, 6, 7 and 8 weeks after injection of carrageenan. In the skin group, the PWLs and mechanical withdrawal threshold were measured before and 2 and 4 h after injection.
Seven groups of two animals each were injected with 3% carrageenan into the knee joint under light halothane anesthesia. Groups 1–7 were sacrificed at 4, 8, 24 h, 1, 2, 4 and 8 weeks, respectively, after the injection. Both ipsi- and contralateral knee joints were dissected and fixed in 10% formalin. In another seven groups of two animals each, 3% carrageenan was injected in the gastrocnemius muscle and the muscle was dissected and fixed according to the same protocol and time schedule as the knee joint. Knee joints were subjected to decalcification prior to embedment in paraffin. Paraffin sections of all tissues were stained with hematoxylin and eosin (H and E) and examined by light microscopy. Analysis of histological findings was descriptive and performed in a blinded fashion by a pathologist.
Statistical analyses were performed using SPSS 10.1 statistical software. The level of significance was set at P < 0.05. Differences in PWL in heat testing were determined across time (baseline to 8 weeks) and among different carrageenan doses (0.3, 1 and 3%) using repeated measures of analysis of variance (ANOVA) followed by Tukey’s post hoc test for differences between groups and paired t-test for differences from baseline. For mechanical testing, non-parametric tests were used. Friedman’s ANOVA assessed differences across time, followed by sign test to determine differences from baseline. Kruskal—Wallis one-way ANOVA was used to test differences between groups followed by sign test to determine differences between individual groups. The area under the curve (AUC) for time—response curves for PWL and mechanical thresholds were calculated for individual animals using Sigmaplot® software which takes into account time on X-axis and response on the Y-axis. Student’s t-test was used to compare differences between group means. All data are presented as mean ± SEM.
Intraarticular injection of carrageenan produced an ipsilateral decrease in the PWL to radiant heat that was long-lasting depending on the dose. Baseline withdrawal latencies were 17.7 ± 0.57 s and decreased to 12.0 ± 0.38 s within 4 h after injection of 3% carrageenan into the knee joint. Injection of 3% carrageenan produced a decrease in PWL ipsilaterally by 4 h that lasted through 6 weeks (F1,11 = 3.7, P = 0.02) (Fig. 1A). Contralaterally, there was a significant decrease in PWL to heat (F1,11 = 12.9, P = 0.001) after injection of 3% carrageenan into the knee (Fig. 1B). Lower doses of carrageenan produced only ipsilateral effects that were shorter-lasting, 24 h for 0.3% and up to 3 weeks for 1% carrageenan (Fig. 1B). Analysis of the AUC for the PWL to heat was greater ipsilaterally for the group that received 3% carrageenan when compared to the 0.3% and contralaterally compared to 0.3 or 1% dose (Fig. 2A).
Intramuscular injection of 3% carrageenan decreased PWL to heat ipsilaterally that was long-lasting, i.e. decreased through 8 weeks (F1,11 = 16.7, P = 0.006) (Fig. 1D). Baseline withdrawal latencies were 17.2 ± 0.94 s and decreased to 11.8 ± 0.47 s within 4 h after injection of 3% carrageenan into the gastrocnemius muscle. There was a contralateral decrease in PWL to heat with 3% i.m. carrageenan by 1 week that also lasted through 8 weeks (F1,11 = 21.4, P = 0.04) (Fig. 1C). There were no significant differences in PWL to heat for the group injected with either 0.3 or 1% carrageenan into the gastrocnemius muscle, ipsi- or contralaterally. Analysis of the AUC for the PWL to heat from the group that received 3% carrageenan was significantly greater both for ipsi- and contralateral sides than the group that received 1% carrageenan (Fig. 2C).
The mechanical withdrawal threshold decreased ipsilaterally following intraarticular injection of carrageenan (χ2 = 54.2, P = 0.0001). This decrease was dependent on dose and differences between doses occurred 4 (P = 0.03), 8 (P = 0.001), 24 h (P = 0.05), 3 (P = 0.02), 5 (P = 0.004), 6 (P = 0.02), 7 (P = 0.001) and 8 weeks (P = 0.02) after carrageenan. The group that received 3% carrageenan showed significantly decreased withdrawal thresholds compared to 0.3% group at 4, 8, 24 h, 3, 4, 5 and 7 weeks and compared to 1% group at 24 h, 3, 4, 5, 6 and 7 weeks. Ipsilaterally, 0.3% carrageenan did not produce a significant decrease in mechanical withdrawal threshold at any time after injection. Injection of 1% carrageenan into the knee joint showed significant decreases compared to baseline 4, 8, 24 h and 1 week, ipsilaterally. Injection of 3% carrageenan produced significant decreases in withdrawal threshold to mechanical stimuli 4, 8, 24, 1, 3 and 5 weeks compared to baseline values (Fig. 3A). Analysis of the AUC of time—response curves from the group that received 3% carrageenan was significantly greater ipsilaterally, than the group that received 0.3 and 1% carrageenan.
A contralateral decrease in mechanical threshold also occurred with 3% carrageenan by 3 weeks that remained through 6 weeks. Significant differences between doses occurred 3 (P = 0.02), 4 (P = 0.005), 5 (P = 0.005), 6 (P = 0.03) and 7 weeks (P = 0.03) after intraarticular injection of carrageenan with the group that received 3% carrageenan showing greater reductions in withdrawal thresholds compared to 0.3 (3, 4 and 5 weeks) and 1% carrageenan (3, 4, 5, 6 and 7 weeks) (Fig. 3B). Analysis of the AUC for time—response curves from the group with 3% carrageenan was significantly greater than the group with 1% carrageenan (Fig. 2B).
The mechanical withdrawal threshold decreased ipsilaterally following intramuscular injection of carrageenan (χ2 = 35.8, P = 0.0001). This decrease was dependent on dose and differences between doses occurred 4 (P = 0.001), 8 (P = 0.001), 24 h (P = 0.004), 1 (P = 0.003), 2 (P = 0.006), 3 (P = 0.04), 4 (P = 0.01), 5 (P = 0.005), 6 (P = 0.01) and 8 weeks (P = 0.05) after carrageenan injection. The group received 3% carrageenan showed significantly decreased withdrawal thresholds ipsilaterally, compared to 0.3% group at 4, 8, 24 h, 3, 4, 5, 6 and 8 weeks and compared to 1% group at 4, 8 h, 1, 2, 3, 4, 5 and 6 weeks. Ipsilaterally, 0.3% carrageenan did not produce a significant decrease in mechanical withdrawal threshold at any time after injection. Injection of 1% carrageenan into the gastrocnemius muscle showed significant decreases in mechanical withdrawal threshold compared to baseline, 4 and 8 h ipsilaterally. Injection of 3% carrageenan produced significant decreases in withdrawal threshold to mechanical stimuli 4, 8, 24 h, 1, 2, 4 and 5 weeks compared to baseline values (Fig. 3C). Analysis of the AUC of time—response curves from the group that received 3% carrageenan was significantly greater ipsilaterally than the group that received 0.3 and 1% carrageenan (Fig. 2D).
A contralateral decrease in mechanical threshold also occurred following intramuscular injection of 3% carrageenan by 3 weeks that remained through 6 weeks (Fig. 3D). Significant differences between doses occurred 1 (P = 0.008), 2 (P = 0.03), 3 (P = 0.03), 4 (P = 0.004), 5 (P = 0.002), 6 (P = 0.007), 7 (P = 0.02) and 8 weeks (P = 0.03) after intramuscular injection of carrageenan with the group that received 3% carrageenan showing greater reductions in withdrawal thresholds compared to 0.3 (1–8 weeks) and 1% carrageenan (1–8 weeks). Analysis of the AUC for the time—response curves from the contralateral paw showed that 3% carrageenan was significantly greater than the group with 1% (Fig. 2D).
Injection of 3% carrageenan into the skin surrounding the anterior aspect of the knee did not produce any significant changes in withdrawal latency to heat or withdrawal threshold to mechanical stimuli (data not shown). The withdrawal latency to heat ipsilaterally was 17.6 ± 1.1 s before injection and 15.7 ± 0.75 s and 16.9 ± 0.87 s 2 and 4 h after injection of 3% carrageenan into the skin. Mechanical withdrawal threshold remained at a median value of 162 mN before and after injection of 3% carrageenan into the skin.
Animals showed spontaneous pain behaviors such as guarding the injected paw and weight-bearing on the contralateral paw during the first 24–48 h. After 48 h, there was no sign of spontaneous pain except that there was curling of the paw ipsliaterally for 1–2 weeks.
Both knee and muscle developed similar patterns of acute inflammation during the first week following carrageenan injection (Fig. 4). Mild hemorrhage, edema and minimal inflammatory cell infiltrates (mostly neutrophils) were seen 4 h after injection in both tissues. More neutrophils were found at 8 h. At 24 h, acute inflammation was severe and accompanied by myonecrosis and fibrinous exudate into the joint space. By 1 week, the inflammation had converted to primarily a macrophage response with a few scattered mast cells. Muscle inflammation was mostly epimysial and perimysial. Changes at week 2 were very similar to that at week 1 for muscle and joint tissue. By the fourth week, macrophages were still present, but less in number and by the eighth week, only mild chronic inflammation (primarily macrophages) was seen. There was no evidence of inflammatory cell infiltrates in the contralateral knee joint or muscle tissues at any time points (Fig. 4).
The present study examined dose effects, time course and laterality of carrageenan-induced hyperalgesia, when injected unilaterally into knee joint or gastrocnemius muscle in rats. Three percent carrageenan injected unilaterally in either muscle or knee joint produces acute unilateral hyperalgesia to heat and mechanical stimuli that spreads contralaterally within 1–2 weeks and is long-lasting (weeks). Lower doses of carrageenan produce acute unilateral hyperalgesia that is transient. Injection of carrageenan into the skin in a comparable dermatome to the muscle and joint did not produce hyperalgesia to mechanical or heat stimuli. Histologically, there is an acute inflammatory response with neutrophilic infiltration for the first week. This converts to a macrophage-dominated chronic inflammation by 1 week that lasts through 8 weeks. Thus, carrageenan can be used as a model of acute inflammation of muscle or joint for the first 24 h and as a model of chronic inflammatory hyperalgesia after 1–2 weeks.
Pain originating from muscle or joint is uniquely different from that of pain originating from skin. Muscle pain is diffuse, longer lasting and more unpleasant (Torebjork et al., 1984; Marchettini et al., 1996; Svensson et al., 1997; Witting et al., 2000). This may relate to differences in central anatomical pathways or different biochemical mediators. Dorsal root ganglion (DRG) cells that innervate muscle and joint have more calcitonin gene-related peptide and substance P and less isolectin B4 and somatostatin when compare to DRG cells innervating skin (O’Brien et al., 1989; Plenderleith and Snow, 1993). The central projections from primary afferents innervating muscle and joint are predominately to laminae I and deeper dorsal horn, while those from cutaneous tissue also project to laminae II (Mense and Craig, 1988; Craig et al., 1988; Willis and Coggeshall, 1991; Mense, 1993; Schaible and Grubb, 1993). Thus, pain from muscle or joint would be expected to produce a different patterned response than pain from skin.
It has been previously shown that most of the injected carrageenan is removed within 48 h from the knee joint (Santer et al., 1983) and possibly faster from other tissues. However, secondary hyperalgesia produced by carrageenan persists even after 48 h and transforms into chronic hyperalgesia that is long-lasting (weeks). The induction, maintenance and spread of chronic hyperalgesia could result from a series of peripheral and central changes occurring at the site of insult and at spinal or supraspinal sites.
Peripherally, carrageenan injected into the muscle causes the symptoms of myositis viz. hyperemia, edema and infiltration by neutrophils (Berberich et al., 1988). When injected into the knee joint, carrageenan leads to synthesis and release of inflammatory mediators, which cause edema and rapid infiltration of neutrophils within the first few hours (Schaible and Grubb, 1993). Injection of carrageenan/kaolin mixture into the knee joint causes an immediate increase in glutamate and nitric oxide metabolites in the knee joint, which persists for hours (Lawand et al., 2000). The increase is prevented by intraarticular administration of lidocaine suggesting that the glutamate is released from neuronal endings in the joint (Lawand et al., 2000). In the present study, when carrageenan was injected into the knee joint or muscle, an acute phase of inflammation followed by a chronic phase, lasting up to 8 weeks, was observed. The local changes occurring after carrageenan insult are likely responsible for the sensitization of the peripheral nociceptors and primary afferents, which contribute to the development of secondary hyperalgesia, along with central changes (Mense, 1993; Schaible et al., 2002).
Injection of carrageenan into the paw of rats induces cyclooxygenase-2 (COX-2) and produces prostaglandin E2 locally (Di Rosa et al., 1971; Nantel et al., 1999), induces COX-2 mRNA in the lumbar spinal cord (Hay and de Belleroche, 1997) and releases glutamate, aspartate, substance P, nitric oxide and prostaglandin E2 in the dorsal horn (Schaible et al., 1990; Sluka and Westlund, 1992; Sorkin et al., 1992; Yang et al., 1996; Rivot et al., 2002). Carrageenan has also been found to cause an acute increase in immunoreactive calcitonin gene-related peptide and substance P in the spinal cord that lasts through at least 1 week (Garry and Hargreaves, 1992; Sluka and Westlund, 1993). Some or all of these spinal changes could be responsible for the central sensitization occurring spinally, following carrageenan-induced inflammation of a joint or muscle (Schaible et al., 1987; Dougherty and Willis, 1992; Hoheisel et al., 1995). Supraspinal involvement in secondary hyperalgesia produced by peripheral tissue injury, including knee joint carrageenan, has been reported (Pertovaara, 1998; Urban et al., 1996, 1999). The chronic phase of hyperalgesia probably is maintained by the central sensitization.
Our data also indicate that the duration and contralateral spread of chronic hyperalgesia depends on the severity of initial inflammation produced by carrageenan and the site of injection, i.e. deep tissues. In a different study from our laboratory (Sluka, 2002), capsaicin injected into cutaneous tissue produces only short-lasting secondary mechanical hyperalgesia whereas injection of capsaicin into deep tissues, muscle or joint produces a long-lasting bilateral mechanical hyperalgesia in rats. Zymosan injected around the sciatic nerve at a low dose produces ipsilateral mechanical hyperalgesia whereas higher doses produce a bilateral hyperalgesia that is not a result of systemic spread of Zymosan (Chacur et al., 2001). A dose-dependent increase in hyperalgesia induced by carrageenan paw inflammation also occurs but remains ipsilateral and is short-lasting (Hargreaves et al., 1988). Further, CFA or calcitonin gene-related peptide injected ipsilaterally into the joint dose-dependently produces contralateral inflammation and hyperalgesia that depends on central neural mechanisms (Donaldson et al., 1993, 1995; Rees et al., 1996; Bileviciute et al., 1993, 1998). It has to be re-emphasized in this context that, in the current study, there was a contralateral spread of mechanical and heat hyperalgesia with 3% carrageenan, both in muscle and joint-injected animals, whereas, there was no contralateral spread with the lower concentrations (0.3 and 1%) and with skin injections. It may, therefore, be hypothesized that, greater the initial tissue insult and deeper the injected tissue, hyperalgesia will be more chronic and bilateral.
Bilateral effects of unilateral injury have been reported and reviewed by many investigators (Fitzgerald, 1982; Woolf, 1983; Coderre and Melzack, 1985; Kayer and Guilbaud, 1987; Bileviciute et al., 1993; Donaldson et al., 1993; Mapp et al., 1993; Kissin et al., 1998; Chen et al., 1999, Decaris et al., 1999; Donaldson, 1999, Koltzenburg et al., 1999; Lowrie, 1999; Chacur et al., 2001; Sluka et al., 2001; Sluka, 2002). C-fiber stimulation of the nerve to the gastrocnemius—soleus muscle or application of mustard oil intraarticularly produces bilateral increases in the flexion reflex (Woolf, 1983; Woolf and Wall, 1986). The contralateral increase in the flexion reflex is unaffected by blockade of afferent input from the site of injury (Woolf, 1983). Similarly, Sluka et al. (2001) showed that unilateral injection of acidic saline produces a long-lasting bilateral hyperalgesia in rats, which is not abolished by lidocaine injection into the gastrocnemius muscle or unilateral dorsal rhizotomy. Also, capsaicin injected into the muscle or knee joint produces a long-lasting bilateral mechanical, but not heat hyperalgesia (Sluka, 2002). This mechanical hyperalgesia is dependent on the early activation of the cAMP pathway spinally during the first 24 h after capsaicin injection (Sluka, 2002). Chen et al. (2000) showed that a unilateral subcutaneous injection of bee venom into the plantar surface of the hindpaw produces bilateral heat and mechanical hyperalgesia in rats, after 4 h. The contralateral hyperalgesia is not abolished after ipsilateral sciatic nerve axotomy, but the development of contralateral effect is prevented by prior administration of N-methyl-D-aspartate (NMDA) or non-NMDA receptor antagonists intrathecally, indicating a central sensitization mediated by spinal excitatory amino acids (Chen et al., 2000). Thus, contralateral spread of hyperalgesia likely depends on plastic changes in the central nervous system.
Following carrageenan-induced hindpaw inflammation in rats, phosphorylated-cAMP-responsive element binding protein (P-CREB) increases bilaterally in the spinal cord (Messersmith et al., 1998). Although an increase in P-CREB is not a sufficient parameter, it is a necessary factor for nociceptive-specific increases in c-fos expression (Messersmith et al., 1998), which could increase neuronal excitability and hence, hyperalgesia. Bilateral increases in substance P and calcitonin gene-related peptide were also observed in the spinal dorsal horn during the acute phase of experimentally induced monoarthritis in rat knee (Mapp et al., 1993) and for up to 1 week following carrageenan knee joint inflammation (Sluka and Westlund, 1993). Thus, bilateral changes in transcription factors and neurotransmitters may contribute to the long-lasting, bilateral hyperalgesia associated with carrageenan-induced muscle or joint inflammation.
Although most of the above mentioned mechanisms and other reviews (Koltzenburg et al., 1999) on the bilateral effects of unilateral injury points towards a spinal role, a supraspinal mechanism cannot be ruled out, since supraspinal centers are involved in the development and maintenance of secondary hyperalgesia (Herrero and Cervero, 1996; Pertovaara, 1998; Urban et al., 1996, 1999). For instance, A- and C-fiber-mediated wind-up of flexor motoneurons following knee joint injection of carrageenan is prevented by spinal transection (Herrero and Cervero, 1996). Further, inactivation of the rostral ventral medulla (RVM) by lidocaine reverses and lesion of RVM by pre-treatment with ibotenic acid completely blocks secondary heat hyperalgesia produced by knee joint injection of carrageenan (Urban et al., 1999). These manipulations in the RVM do not affect the primary hyperalgesia produced by carrageenan injected into the plantar paw. Thus, supraspinal centers play a major role in the production and maintenance of secondary hyperalgesia. It is quite possible that the contralateral hyperalgesia observed in the current study could be mediated by RVM or other supraspinal centers through descending facilitatory pathways, although we do not have data to support these suggestions.
Interestingly, the contralateral hyperalgesia did not occur until 1–2 weeks after induction of inflammation. The reasons for this are unclear, but may relate to conversion of the inflammatory response from acute to chronic, since the contralateral spread follows the same time course. Alternatively, this could result from initiation of gene transcription mediating plastic changes in the central nervous system as a consequence of the tissue injury.
Histopathological examination of the tissues in the current study shows inflammatory changes that parallel the long-lasting hyperalgesia observed. The chronic hyperalgesia, therefore, could be maintained by chronic inflammation observed in the tissues. However, there are no contralateral signs of inflammation observed in our studies, supporting a neuronal role, either spinal or supraspinal, for the contralateral spread of hyperalgesia. Several previous studies show a bilateral development of articular inflammation and degeneration after a unilateral injection of CFA (Donaldson et al., 1995; Decaris et al., 1999). The absence of the contralateral histological changes in our studies could be due to the difference in the inflammatory agent used, i.e. carrageenan vs. CFA. The histopathological changes observed in the present study correlate well with the hyperalgesic time periods observed with 3% carrageenan injection. Both muscle and joint tissues on the ipsilateral side show the earliest signs of acute inflammation around 4–8 h, changing to chronic inflammation by 1–2 weeks that persists for up to 8 weeks. Both muscle and joint groups show hyperalgesia starting by 4 h, with the 3% group showing continued hyperalgesia through 7–8 weeks.
Findings from this study clearly show a dose-dependent development of chronic hyperalgesia following injection of carrageenan into muscle or joint in the rat and contralateral spread of hyperalgesia at the higher dose. The data also support the earlier findings that injury to deeper tissues result in a robust and long-lasting contralateral hyperalgesia compared to cutaneous insult. The study provides animal models for acute as well as chronic hyperalgesia, one induced by muscle inflammation and the other by joint inflammation with carrageenan, both of which are possibly maintained by spinal or supraspinal neuronal mechanisms.
The authors wish to thank Tammy Lisi, Chris Bromley and Jan Rogers for excellent technical assistance and Ms Carol Leigh for assistance with manuscript preparation. This work was supported by National Institutes of Health grants R01 NS39734 and K02 AR02201 (KAS).