Berenice B. Lorenzetti

The pivotal role of tumour necrosis factor α in the development of inflammatory hyperalgesia

1 The hyperalgesic activities in rats of interleukin‐1β (IL‐1β), IL‐6, IL‐8, tumour necrosis factor α (TNFα) and carrageenin were investigated. 2 IL‐6 activated the previously delineated IL‐1/prostaglandin hyperalgesic pathway but not the IL‐8/sympathetic mediated hyperalgesic pathway. 3 TNFα and carrageenin activated both pathways. 4 Antiserum neutralizing endogenous TNFα abolished the response to carrageenin whereas antisera neutralizing endogenous IL‐1β, IL‐6 and IL‐8 each partially inhibited the response. 5 The combination of antisera neutralizing endogenous IL‐1β + IL‐8 or IL‐6 + IL‐8 abolished the response to carrageenin. 6 These results show that TNFα has an early and crucial role in the development of inflammatory hyperaglesia. 7 The delineation of the roles of TNFα, IL‐1β, IL‐6 and IL‐8 in the development of inflammatory hyperalgesia taken together with the finding that the production of these cytokines is inhibited by steroidal anti‐inflammatory drugs provides a mechanism of action for these drugs in the treatment of inflammatory hyperalgesia.

Central and peripheral antialgesic action of aspirin-like drugs

The peripheral and central effects of some non-steroid anti-inflammatory drugs, aspirin, indomethacin, paracetamol and phenacetin were studied by comparing their intraplantar and intracerebroventricular effects on hyperalgesia induced by carrageenin injected into the rat paw. Hyperalgesia was measured by a modification of the Randall-Selitto test. The agents tested had antialgesic effects when given by any route. Their intraventricular administration enhanced the antialgesic effect of anti-inflammatory drugs administered into the paw. Previous treatment of one paw with carrageenin reduced the oedema caused by a second injection of carrageenin in the contralateral paw. In contrast, it had no effect on the intensity of hyperalgesia but shortened the time necessary for it to reach a plateau. Administration of a prostaglandin antagonist (SC-19220) in the cerebral ventricles, in the rat paw or in both sites, significantly inhibited the hyperalgesia evoked by carrageenin. The maximal hyperalgesic effect of intraplantar injections of prostaglandin E2 could be further enhanced by its cerebroventricular administration. It was suggested that carrageenin hyperalgesia has a peripheral and a central component and that the cyclo-oxygenase inhibitors used may exert an antialgesic effect by preventing the hyperalgesia induced by a peripheral and/or central release of prostaglandins.

Peripheral analgesia and activation of the nitric oxide-cyclic GMP pathway

We have previously described the analgesic effect of dibutyryl cyclic GMP or acetylcholine (ACh) injected into rat paws. Since ACh induces nitric oxide (NO) release from endothelial cells, we investigated the possible involvement of the NO-cyclic GMP pathway in ACh-induced analgesia, using a modification of the Randall-Selitto rat paw test. We found that sodium nitroprusside, which releases NO non-enzymatically, caused antinociception in the rat paw made hyperalgesic with prostaglandin E2. The analgesic effect of sodium nitroprusside and ACh was enhanced by intraplantar injection of an inhibitor of cyclic GMP phosphodiesterase (MY 5445) and was blocked by a guanylate cyclase inhibitor, methylene blue (MB). The analgesia induced by ACh, but not by sodium nitroprusside, was blocked by NG-monomethyl-L-arginine (L-NMMA), an inhibitor of the formation of NO from L-arginine. L-arginine itself had little or no effect upon prostaglandin-induced hyperalgesia but caused significant analgesia in paws inflamed with carrageenin. This analgesia was blocked by MB, as well as by L-NMMA, and was potentiated by MY 5445. These results suggest that ACh-induced analgesia was mediated via the release of NO. The results also indicate that the guanylate cyclase system is stimulated in the inflammatory reaction. The analgesia resulting from activation of this system is possibly overshadowed by substances that concomitantly stimulate nociceptor hyperalgesic mechanisms.

Interleukin-8 as a mediator of sympathetic pain.

1 The hyperalgesic effects of interleukin‐8 (IL‐8), interleukin‐1β (IL‐1β) and carrageenin were measured in a rat paw pressure test. 2 IL‐8 evoked a dose‐dependent hyperalgesia which was attenuated by a specific antiserum, the β‐adrenoceptor antagonists atenolol and propranolol, the dopamine1 receptor antagonist SCH 23390 and the adrenergic neurone‐blocking agent guanethidine. The hyperalgesia was not attenuated by the cyclo‐oxygenase inhibitor indomethacin or the IL‐1β analogue Lys‐d‐Pro‐Thr. 3 IL‐1β‐evoked hyperalgesia was attenuated by indomethacin and Lys‐d‐Pro‐Thr but not by atenolol or SCH 23390. 4 Carrageenin‐evoked hyperalgesia was attenuated by atenolol, indomethacin and anti‐IL‐8 serum. The effects of atenolol and anti‐IL‐8 serum were not additive. The effects of indomethacin and anti‐IL‐8 serum were additive: this combination abolished carrageenin‐evoked hyperalgesia. 5 A new biological activity of IL‐8 is described, namely the capacity to evoke hyperalgesia by a prostaglandin‐independent mechanism. IL‐8 is the first endogenous mediator to be identified as evoking hyperalgesia involving the sympathetic nervous system. Since IL‐8 is released by activated macrophages and endothelial cells it may be a humoral link between tissue injury and sympathetic hyperalgesia.

The molecular mechanism of action of peripheral morphine analgesia: stimulation of the cGMP system via nitric oxide release

In addition to the central and spinal sites of action of morphine both our laboratory and others have demonstrated that opiates can also cause analgesia through a peripheral mechanism (Ferreira, 1983). However, the molecular basis of the mechanism of the analgesic actions of opiates remains unknown. In vitro studies show that morphine is able to inhibit activation of adenylate cyclase (Collier and Roy, I974) as well as to stimulate formation of cGMP (Minneman and Iversen, I976). Recently, using the rat paw hyperalgesia test, we showed that in the periphery, acetylcholine-induced analgesia was mediated via the activation of the nitric oxide/cGMP pathwny (Duarte et al., I990). This conclusion was based upon the fact that intraplantar injection of sodium nitroprusside, a substance which nonenzymatically releases nitric oxide (NO), caused analgesia. Furthermore, the analgesic effects of acetylcholine (ACh) and sodium nitroprusside were enhanced by intraplantar injection of an inhibitor of cyclic GMP phosphodiesterase, My5445. In addition, methylene bk:e (MB), an inhibitor of guanylate cyclase, blo1.:ked the analgesia induced by acetylcholine and sodium nitroprusside. On the other hand, the analgesia induced by acetylcholine, but not by sodium nitroprusside, was blocked by N°-monomethyl-L-arginine (LNMMA), an inhibitor of the formation of NO from L-arginine. Due to the similarity of the local action of ACh and opiates we investigated whether agents which affect the arginine/NO-cGMP pathway also interfere with morphine-induced peripheral analgesia as tested with our modification of the Randall-Selitto rat paw pressure test (see Duarte et al., I990). In this test a constant pressure of 20 mmHg is applied to the hind paw of rats

Bradykinin initiates cytokine-mediated inflammatory hyperalgesia

1 The hyperalgesic activities in rats of bradykinin, carrageenin and lipopolysaccharide (LPS) were investigated in a model of mechanical hyperalgesia. 2 Bradykinin and carrageenin evoked dose‐dependent hyperalgesia with maximum responses of similar magnitude to responses to LPS (1 and 5 μg). 3 Hoe 140, an antagonist of BK2 receptors, inhibited in a dose‐dependent manner hyperalgesic responses to bradykinin, carrageenin and LPS (1 μg) but not responses to LPS (5 μg), prostaglandin E2, dopamine, tumour necrosis factor α (TNFα), IL‐1, IL‐6 and IL‐8. 4 Responses to bradykinin and LPS (1 and 5 μg) were inhibited by the cyclo‐oxygenase inhibitor, indomethacin and by the β‐adrenoceptor antagonist, atenolol. The effects of indomethacin and atenolol were additive: their combination abolished responses to bradykinin and LPS (1 μg) and markedly attenuated the response to LPS (5 μg). 5 Antiserum neutralizing endogenous TNFα abolished the response to bradykinin whereas antisera neutralizing endogenous IL‐1β, IL‐6 and IL‐8 each partially inhibited the response. The combination of antisera neutralizing endogenous IL‐1β + IL‐8 or IL‐6 + IL‐8 abolished the response to bradykinin. 6 Antisera neutralizing endogenous TNFα, IL‐1β, IL‐6 and IL‐8 each partially inhibited responses to LPS (1 and 5 μg). Increasing the dose of antiserum to TNFα or giving a combination of antisera to IL‐1β + IL‐8 or IL‐6 + IL‐8 further inhibited responses to LPS (1 and 5 μg). 7 These data show that bradykinin can initiate the cascade of cytokine release that mediates hyperalgesic responses to carrageenin and endotoxin (1 μg). The lack of effect of Hoe 140 on hyperalgesic responses to LPS (5 μg) suggests that the release of hyperalgesic cytokines can be initiated independently of bradykinin BK2 receptors.

Analgesia by direct antagonism of nociceptor sensitization involves the arginine-nitric oxide-cGMP pathway

We tested the hypothesis that activation of the nitric oxide (NO)-cGMP pathway is involved in the mechanism of two directly acting non-opiate peripheral analgesics, myrcene and dipyrone, using our modification of the Randall-Selitto test. The NO inhibitor, NG-monomethyl-L-arginine (50 micrograms/paw) and methylene blue (500 micrograms/paw) abolished the analgesic effect of dipyrone and myrcene. Dibutyryl cyclic adenosine monophosphate (DbcAMP) caused a dose-dependent hyperalgesia (20, 50 and 100 micrograms/paw). Only responses to low doses of DbcAMP were inhibited by the two analgesics. Pretreatment with MY5445 (50 micrograms/paw) resulted in potentiation of the effects of both analgesics. These results support our hypothesis that the sensitivity of nociceptors may be controlled by the balance between the levels of cAMP and cGMP. Stimulation of the NO-cGMP pathway is probably the common denominator for the mode of action of peripheral analgesics which block hyperalgesia directly.

Mode of analgesic action of dipyrone: direct antagonism of inflammatory hyperalgesia.

Dipyrone blocked carrageenin-induced oedema and hyperalgesia in a dose-dependent manner. In contrast with indomethacin, paracetamol and acetyl salicylic acid, much lower doses of dipyrone were necessary for blocking hyperalgesia (ED50 = 19 mg/kg, i.p.) than oedema (180 mg/kg, i.p.) Dipyrone administered intraperitonially or intraplantarly was able to antagonise PGE2-, isoprenaline- and calcium chloride-induced hyperalgesia, effects which are not observed with non-steroid anti-inflammatory drugs. Systemic or local administration of dipyrone had no effect upon Db-cAMP-induced hyperalgesia while a centrally acting analgesic, morphine, given systemically, was highly effective. These results support our suggestion that the mechanism of action of dipyrone is different from that of classical non-steroidal anti-inflammatory drugs. Although the site of action is peripheral its analgesic effect does not derive from inhibition of the synthesis of prostaglandins but is exerted via direct blockade of the inflammatory hyperalgesia.

Role of lipocortin-1 in the anti-hyperalgesic actions of dexamethasone

The effect of dexamethasone, lipocorton‐12–26 and an antiserum to lipocortin‐12–26 (LCPS1) upon the hyperalgesic activities in rats of carrageenin, bradykinin, tumour necrosis factor α (TNFα), interleukin‐12, interleukin‐6 (IL‐6), interleukin‐8 (IL‐8), prostaglandin Eβ (PGE2) and dopamine were investigated in a model of mechanical hyperalgesia. Hyperalgesic responses to intraplantar (i.pl.) injections of carrageenin (100 μg), bradykinin (500 ng), TNFα (2.5 pg), IL‐1β (0.5 pg), and IL‐6 (1.0 ng), but not responses to IL‐8 (0.1 ng), PGE2 (100 ng) and dopamine (10 μg), were inhibited by pretreatment with dexamethasone (0.5 mg kg−1, subcutaneously, s.c., or 0.04–5.0 μg/paw). Inhibition of hyperalgesic responses to injections (i.pl.) of bradykinin (500 ng) and IL‐1β (0.5 pg) by dexamethasone (0.5 mg kg−1, s.c.) was reversed by LCPS1 (0.5 ml kg−1, injected s.c., 24 h and 1 h before hyperalgesic substances) and hyperalgesic responses to injections (i.pl.) of bradykinin (500 ng), TNFα (2.5 pg) and IL‐1β (0.5 pg), but not responses to PGE2 (100 ng), were inhibited by pretreatment with lipocortin‐12–26 (100 μg/paw). Also, lipocortin‐12–26 (30 and 100 μg ml−1) and dexamethasone (10 μg ml−1) inhibited TNFα release by cells of the J774 (murine macrophage‐like) cell‐line stimulated with LPS (3 μg ml−1), and LCPS1 partially reversed the inhibition by dexamethasone. These data are consistent with an important role for endogenous lipocortin‐12–26 in mediating the anti‐hyperalgesic effect of dexamethasone, with inhibiton of TNFα production by lipocortin‐12–26 contributing, in part, to this role. Although arachidonic acid by itself was not hyperalgesic, the hyperalgesic response to IL‐1β (0.25 pg, i.pl.) was potentiated by arachidonic acid (50 μg) and the potentiated response was inhibited by dexamethasone (50 μg, i.pl.) and lipocortin‐12–26 (100 μg, i.pl.). Also, lipocortin‐12–26 (30 and 100 μg ml−1) inhibited/abolished PGE2 release by J774 cells stimulated with LPS (3 μg ml−1). These data suggest that, in inflammatory hyperalgesia, inhibition of the induction of cyclo‐oxygenase 2 (COX‐2), rather than phospholipase A2, by dexamethasone and lipocortin‐12–26 accounts for the anti‐hyperalgesic effects of these agents. The above data support the notion that induction of lipocortin by dexamethasone plays a major role in the inhibition by dexamethasone of inflammatory hyperalgesia evoked by carrageenin, bradykinin and the cytokines TNFα, IL‐1β and IL‐6, and provides additional evidence that the biological activity of lipocortin resides within the peptide lipocortin‐12–26. Further, the data suggest that inhibition of lipocortin‐12–26 of eicosanoid production by COX‐2 also contributes to the anti‐hyperalgesic effect of lipocortin‐1.

Bradykinin B1 and B2 receptors, tumour necrosis factor α and inflammatory hyperalgesia

The effects of BK agonists and antagonists, and other hyperalgesic/antihyperalgesic drugs were measured (3 h after injection of hyperalgesic drugs) in a model of mechanical hyperalgesia (the end‐point of which was indicated by a brief apnoea, the retraction of the head and forepaws, and muscular tremor). DALBK inhibited responses to carrageenin, bradykinin, DABK, and kallidin. Responses to kallidin and DABK were inhibited by indomethacin or atenolol and abolished by the combination of indomethacin+atenolol. DALBK or HOE 140, given 30 min before, but not 2 h after, carrageenin, BK, DABK and kallidin reduced hyperalgesic responses to these agents. A small dose of DABK+a small dose of BK evoked a response similar to the response to a much larger dose of DABK or BK, given alone. Responses to BK were antagonized by HOE 140 whereas DALBK antagonized only responses to larger doses of BK. The combination of a small dose of DALBK with a small dose of HOE 140 abolished the response to BK. The hyperalgesic response to LPS (1 μg) was inhibited by DALBK or HOE 140 and abolished by DALBK+HOE 140. The hyperalgesic response to LPS (5 μg) was not antagonized by DALBK+HOE 140. These data suggest: (a) a predominant role for B2 receptors in mediating hyperalgesic responses to BK and to drugs that stimulate BK release, and (b) activation of the hyperalgesic cytokine cascade independently of both B1 and B2 receptors if the hyperalgesic stimulus is of sufficient magnitude.