- Top of page
- Conflict of interest
Chemokines are small molecular weight proteins (8–10 kDa) thought to play an important role in the recruitment of leukocytes from the blood into tissues during inflammatory processes. Among the four classes of chemokines, CXC chemokines acting on their CXCR1 and CXCR2 receptors have been shown to coordinate neutrophil migration and activation in several models of inflammation (Bizzarri et al., 2006; Busch-Petersen, 2006; Reutershan, 2006). For example, pharmacological blockade of CXCR2 or administration of anti-CXC chemokines prevented neutrophil influx and tissue injury following intestinal, hepatic and cerebral ischaemia and reperfusion in rats (Souza et al., 2004; Cavalieri et al., 2005; Garau et al., 2005). More recently, we also showed that blockade of CXCR1/2 receptors prevents neutrophil influx and joint damage following adjuvant-arthritis in rats (Barsante et al., 2007).
In addition to mediating inflammatory tissue injury, neutrophil migration into tissue appears to be involved in the genesis of inflammatory pain (Levine et al., 1984). The sensitization of primary afferent nociceptors is a common denominator of all kinds of inflammatory pain that leads to a state of hyperalgesia and/or allodynia, better described as hypernociception in animal models (Millan, 1999). Hypernociception is induced by the direct action of the final mediators prostaglandins and sympathetic amines (for example, dopamine, adrenaline) on peripheral nociceptors (Ferreira et al., 1978; Nakamura and Ferreira, 1987; Khasar et al., 1999). These direct-acting hyperalgesic mediators are ultimately released in the inflamed tissue in response to a range of inflammatory stimuli that trigger the release of a cascade of cytokines (tumour necrosis factor (TNF)-α, interleukin-1β and CXC chemokines) by resident and incoming cells (Cunha et al., 2005; Verri et al., 2006). In mice, the chemokine CXCL1 (also known as KC, keratinocyte-derived chemokine) is released in response to several inflammatory stimuli, including carrageenan and lipopolysaccharide (LPS), and induces neutrophil influx and hypernociception when given intraplantarly (i.pl.). CXCL1-induced hypernociception is partially dependent on the production of sympathetic amines (Cunha et al., 2005). CXCL1 and TNF-α may also trigger the interleukin-1β/prostaglandin pathway, which has a synergistic effect with sympathetic amines to induce inflammatory hypernociception (Cunha et al., 2005).
In the present paper, we addressed the hypothesis that the effect of CXC chemokines on CXCR1/2 is important to induce neutrophil migration and inflammatory hypernociception induced by several inflammatory stimuli in mice. To address this hypothesis, mice were treated with DF 2162 (4-[(1R)-2-amino-1-methyl-2-oxoethyl]phenyl trifluoromethane sulphonate), a compound belonging to the class of CXCR1/2 allosteric modulators (Bertini et al., 2004; Bizzarri et al., 2006; Barsante et al., 2007). Initial experiments confirmed the selectivity and mode of action of DF 2162. The compound was then tested in models in which neutrophil influx and inflammatory hypernociception were induced by i.pl. injection of the following stimuli: the chemokine CXCL1, carrageenan and LPS. We also determined whether DF 2162 could have a synergistic effect with blockade of prostaglandin synthesis (indomethacin) or with TNF-α blockade (using TNF receptor type-1 (TNFR1)-deficient mice) to prevent inflammatory hypernociception in mice. In addition, the effects of oral treatment with DF 2162 were investigated in a model of zymosan-induced arthritis and in a model of collagen-induced arthritis (CIA) in DBA/1J mice.
- Top of page
- Conflict of interest
DF 2162 is an orally active compound that belongs to the class of 2-arylpropionamide drugs, a novel class of allosteric modulators of CXCR1 and CXCR2 (Bertini et al., 2004; Barsante et al., 2007). In the present study, we confirmed that DF 2162 blocks the action of CXCR1- and CXCR2-active chemokines. Moreover, evidence is provided to show that, akin to reparixin (Bertini et al., 2004), DF 2162 does not displace the ligand but appears to act via binding to an allosteric site at the chemokine receptor. In our in vivo experiments, treatment with DF 2162 greatly prevented the influx of neutrophils induced by CXCL1, a chemokine active on murine CXCR1/2 receptors, carrageenan, LPS and zymosan. More importantly, the compound diminished by a similar extent inflammatory hypernociception induced by the same stimuli, suggesting that CXCR1/2-mediated neutrophil influx is an important element in the cascade of events leading to inflammatory hypernociception. Further experiments in a model of CIA confirmed the ability of the compound to diminish neutrophil influx, oedema, tissue damage (arthritic index) and hypernociception.
Initial experiments were carried out to confirm the ability of DF 2162 to block the activity of the relevant chemokine in mice. Our results clearly show that DF 2162 dose-dependently inhibited the ability of the CXCR2-active chemokine CXCL1 to induce neutrophil recruitment after i.pl. injection. DF 2162 also blocked the neutrophil recruitment induced by carrageenan and LPS. These results are consistent with the known neutrophil-recruiting activity of CXCR2 in mice and the importance of this receptor for neutrophil influx in several models of inflammation (Bozic et al., 1994; Lee et al., 1995; McColl and Clark-Lewis, 1999; Bertini et al., 2004; Souza et al., 2004; Bizzarri et al., 2006; Reutershan, 2006; Barsante et al., 2007). More recently, Fan et al. (2007) described murine CXCR1, a functional receptor for CXCL6 and human IL-8. DF 2162 blocked both CXCR1 and CXCR2 in human cells (Barsante et al., 2007), but the tools are not yet widely available to test whether DF 2162 will block murine CXCR1 with a similar efficacy and potency as it blocks the function of murine CXCR2 and human CXCR1 and CXCR2.
Previous studies from our and other groups (Levine et al., 1984; Bezerra et al., 2007; Cunha et al., 2008), especially in rat models of inflammation, have demonstrated that neutrophils may play a role in mediating inflammatory hypernociception induced by several stimuli, including carrageenan and LPS. The relevance of CXCR2 in the neutrophil-mediated effects was, to the best of our knowledge, not known. In the present study, we showed that blockade of neutrophil influx with DF 2162 was associated with inhibition of inflammatory hypernociception induced by CXCL1, carrageenin and LPS. Indeed, there was a good association between the extent of inhibition of neutrophil influx and the extent of inhibition of inflammatory hypernociception. The inhibition of neutrophil migration and consequently of hypernociception by DF 2162 in a model of zymosan-induced arthritis also support a neutrophil-dependent hypernociceptive mechanism. In the latter model, inhibition of neutrophil migration, by fucoidin or neutralizing antibody against neutrophils, blocked mechanical hypernociception (Guerrero et al., 2008). The ability of CXCL1 to induce neutrophil recruitment by a mechanism that can be inhibited by DF 2162 is consistent with the role of neutrophils in mediating inflammatory hypernociception. Thus, our results clearly indicate that neutrophils recruited in response to CXCR1/2 stimulation are essential in the cascade of events leading to inflammatory hypernociception induced by several inflammatory stimuli.
Blockade of prostaglandins and sympathetic amines is known to diminish hypernociception induced by known mediators of inflammation (Cunha et al., 2005; Verri et al., 2006). Indeed, prostaglandins and sympathetic amines are considered to be direct-acting hypernociceptive mediators (Ferreira et al., 1978; Khasar et al., 1999). They do not induce neutrophil recruitment but act directly on their receptors, present in the nociceptor, to induce hypernociception. DF 2162 greatly reduced the hypernociception induced by carrageenan and LPS but had no effect on the hypernociception induced by PGE2 and dopamine. Two important conclusions may be derived from these latter results. First, the effects of DF 2162 on hypernociception appear to be specific, as the drug does not block hypernociception induced by all stimuli. Second, the action of DF 2162 and, by inference, the role of CXCR1/2, appear to be upstream of the release of prostaglandins and sympathetic amines and, therefore, this compound does not directly affect nociceptor sensitization. Thus, our studies clearly favour the hypothesis that CXCR1/2 is essential in the cascade of non-neuronal events leading to the release of the direct-acting mediators of inflammatory pain, which ultimately cause sensitization of nociceptors and consequently mechanical hypernociception. This is consistent with our findings in mice (Cunha et al., 2005) and studies showing that inhibition of neutrophil influx prevents the local release of PGE2 in rats (Cunha et al., 2008). Thus, the antihypernociceptive actions of DF 2162 appear to be secondary to its ability to prevent neutrophil influx or activation and consequent neutrophil-induced release of other mediators. However, we cannot discard the possibility that DF 2162 also prevents the action of CXCL1 directly on sensory nerves. Indeed, CXCR2 is present on sensory nerves and the activation of this receptor may trigger nociceptor sensitization directly (Qin et al., 2005).
The experiments showing a synergism between DF 2162 and indomethacin or with TNF-α blockade favour our previous hypothesis that two main hypernociceptive pathways mediate carrageenan-induced mechanical inflammatory hypernociception in mice. One pathway, initiated by TNF-α, triggers interleukin-1β and prostaglandin release, and the other pathway, initiated by CXCL1, triggers the release of sympathetic amines and may also stimulate the interleukin-1β/prostaglandin pathway (Cunha et al., 2005).
Although there are a reasonable number of effective drugs used for symptomatic relief of inflammatory disease (for example, non-steroidal anti-inflammatory drugs), there are few safe drugs that modify the pathological processes responsible for chronic inflammation (Olsen and Stein, 2004). Cytokine-based therapies, especially strategies that prevent the effects of TNF-α, have been used for the treatment of rheumatoid arthritis and found to be useful in preventing progression of this disease in several groups of patients. However, the latter therapies are based on the use of exogenous proteins (such as antibodies) that are costly, are given by injection and have the inherent possibility of eliciting an immune response against the administered protein (Olsen and Stein, 2004). As DF 2162 blocked two important components of inflammation, neutrophil influx and hypernociception, thought to be central in human arthritis, we tested the effects of the compound in a model of CIA. We chose to give the compound in a therapeutic regime, that is, 1 day after the onset of the disease. Our experiments showed that DF 2162 decreased paw oedema, arthritic index and neutrophil influx into affected paws. This was accompanied by a small but significant decrease of inflammatory hypernociception, which appeared to be greater towards the end of the treatment. These results expand our recent findings showing that DF 2162 decreased adjuvant-induced arthritis in rats. However, the findings are novel as they indicate that, in addition to decreasing tissue inflammation and damage, CXCR2 blockade may also have the additional benefit of providing relief to the unpleasant symptoms of pain.
In conclusion, this study shows that treatment with DF 2162 greatly prevented neutrophil influx induced by a range of stimuli injected i.pl. or in the knee joint in mice. More importantly, this study is the first to show that treatment with DF 2162, a non-competitive allosteric inhibitor of CXCR2, greatly diminished inflammatory hypernociception. Taken together with the protective effects of the drug in the model of CIA and with our previous studies (Barsante et al., 2007), we suggest that blockade of CXCR2 may be a useful strategy for the treatment of chronic arthritis in humans. In addition to modifying fundamental pathological processes, these compounds may have the additional benefit of providing partial relief for pain. These possibilities need to be tested in the clinical situation.