Abstract
Carbon monoxide-releasing molecules are emerging as a new class of pharmacological agents that regulate important cellular function by liberating CO in biological systems. Here, we examined the role of carbon monoxide-releasing molecule 3 (CORM-3) in modulating neuroinflammatory responses in BV-2 microglial cells, considering its practical application as a novel therapeutic alternative in the treatment of stroke. BV-2 microglia cells were incubated for 24 h in normoxic conditions with thrombin alone or in combination with interferon-γ to simulate the inflammatory response. Cells were also subjected to 12 h of hypoxia and reoxygenated for 24 h in the presence of thrombin and interferon-γ. In both set of experiments, the anti-inflammatory action of CORM-3 was evaluated by assessing its effect on nitric oxide production (nitrite levels) and tumor necrosis factor (TNF)-α release. CORM-3 (75 μM) did not show any cytotoxicity and markedly attenuated the inflammatory response to thrombin and interferon-γ in normoxia and to a lesser extent in hypoxia as evidenced by a reduction in nitrite levels and TNF-α production. Inactive CORM-3, which does not liberate CO and is used as a negative control, failed to prevent the increase in inflammatory mediators. Blockade of endogenous CO production by tin protoporphyrin-IX did not change the anti-inflammatory activity of CORM-3, suggesting that CO liberated from the compound is responsible for the observed effects. In addition, inhibition of the mitogen-activated protein kinases phosphatidyl inositol 3 kinase and extracellular signal-regulated kinase amplified the anti-inflammatory effect of CORM-3. These results suggest that the anti-inflammatory activity of CORM-3 could be exploited to mitigate microglia activity in stroke and other neuroinflammatory diseases.
Stroke is the major cause of disabilities in adults, leaving more than half of the survivors dependent on others for everyday activities (Wolfe et al., 2000). It is also a major cause of dementia, depression, epilepsy, and falls (Rothwell et al., 2004). This pathological event is characterized by blockade of blood supply to the brain and consequently oxygen and glucose deprivation, which lead to necrotic cell death and tissue infarct. Rescuing the surrounding partially ischemic penumbra depends on the severity of brain edema, subsequent neuroinflammation, and production of free radicals.
Microglia are the main inflammatory-reacting cells in the brain after ischemia (Suk, 2004). They act through redoxsensitive inflammatory enzymes such as inducible nitric oxide synthase (iNOS) and NAD(P)H oxidase, which produce NO, superoxide, peroxynitrite, and other reactive oxygen species (ROS) that mediate their phagocytic ability. Free radicals generated during the inflammatory process can directly damage neurons by interacting with their lipid-rich membranes or by increasing α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate receptor susceptibility to the toxic effects of glutamate (Zhao et al., 2004).
Another important element in the cascade of events leading to the progression and exacerbation of stroke is thrombin. Apart from its basic role in forming and propagating the obstructing clot, thrombin contributes to stroke-related brain injury at more than one level. It is now well recognized that thrombin contributes to the formation of brain edema (Guan et al., 2004), neuroinflammation (Hanisch et al., 2004), and neuronal damage (Smirnova et al., 1998). These effects are further augmented by decreased levels of oxygen (hypoxia) and the consequent formation of ROS and proinflammatory cytokines.
Since the late 1990s, carbon monoxide gas (CO) emerged from the shadows of being a catabolic byproduct with no physiological significance into full recognition as a ubiquitous key modulator of cellular response, particularly in its role as signaling mediator and contributor to cell survival during conditions of oxidative stress (Motterlini et al., 2005a). CO is produced by the enzyme heme oxygenase (HO) through heme degradation. HO exists as constitutive (HO-2) and inducible (HO-1) isoforms, the latter being of particular importance in stressful conditions because of its ability to attenuate oxidative stress-mediated injury in response to a wide range of stimuli including ischemia reperfusion injury (Clark et al., 2000). In fact, HO-1 induction has been demonstrated to protect against such injuries in almost all cell types including cardiomyocytes (Clark et al., 2003), hepatocytes (Amersi et al., 1999), lung tissue (Minamino et al., 2001), and neuronal tissue (Panahian et al., 1999). Many of the protective effects of HO are now believed to be mediated by CO (Otterbein et al., 2000), which has been shown to reduce the inflammatory response in a similar fashion to compounds that are known to potently induce HO-1 (Sawle et al., 2005).
In recent years, more emphasis on the possible therapeutic role of CO incited researchers toward the development of new methods to facilitate its use in biological systems; this approach led eventually to the development of carbon monoxide-releasing molecules (CO-RMs) (Clark et al., 2003; Motterlini et al., 2002, 2005a, b). These compounds are biologically compatible CO carriers that can liberate CO in physiological solutions. Different chemical classes of CORMs are currently being developed, but the most extensively studied so far are transition metal carbonyls, which contain ruthenium or iron as core metal and can release CO with different rates in aqueous solutions (Motterlini et al., 2002; Clark et al., 2003; Fairlamb et al., 2005; Sandouka et al., 2006). Ruthenium-based CO-RMs (CORM-2 and CORM-3) have been tested in different models of disease showing promising beneficial effects and are ideal to use because they are easy to synthesize, provide a predictable pattern of CO release, and have no toxic effects at low micromolar concentrations (Motterlini et al., 2002, 2005a; Clark et al., 2003; Sawle et al., 2005; Sandouka et al., 2006). In this study, we examined the effect of CORM-3 on thrombin-induced neuroinflammation in BV-2 microglia in normoxic and hypoxic conditions to establish a potential biological role of CO in stroke and neuroinflammation.
Materials and Methods
Reagents. Tricarbonylchloro(glycinato)ruthenium (II) ([Ru(CO)-3Cl(glycinate)] or CORM-3) was synthesized as previously described (Clark et al., 2003; Foresti et al., 2004). CORM-3 was freshly prepared as a 10 mM stock solution in pure distilled water. An inactive form, which is incapable of releasing CO, was used as a negative control. CORM-3 was “inactivated” (iCORM-3) by adding the compound to cell culture medium and leaving it for 18 h at 37°C in a 5% CO2 humidified atmosphere to liberate CO. The iCORM-3 solution was finally bubbled with nitrogen to remove the residual CO present in the solution. Since CORM-3 contains ruthenium as transition metal, RuCl2(DMSO)4 was also used as an additional negative control. Thrombin, interferon-γ, tin protoporphyrin-IX, SB203580, SP600125, PD98059, LY294002, and all other reagents were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise specified.
Cell Culture. The mouse BV-2 microglia cell line was a kind gift from Professor Rosario Donato (University of Perugia, Italy). The cell line was obtained and characterized as previously described (Adami et al., 2001). Cells were initially grown to confluence in 75-cm2 flasks using RPMI containing 10% fetal bovine serum, 4 mM glutamine, 100 U/ml penicillin, and 10 μg/ml streptomycin in H2O-saturated 5% CO2 atmosphere at 37°C. Cells were subsequently subcultured (1:3 ratio) and grown to 90% confluence in 24-well plates prior to testing.
Experimental Protocol. BV-2 microglia were treated with thrombin (10 U/ml) alone or in combination with interferon-γ (10 ng/ml) in the presence or absence of CO-RM-3 (75 μM) or iCORM-3 (75 μM). Cells were then incubated under normoxic conditions for 24 h. To examine the potential anti-inflammatory action of CORM-3, nitrite levels and tumor necrosis factor (TNF)-α production were determined at the end of the incubation period.
In a similar set of experiments, cells were treated with thrombin (10 U/ml) alone or in combination with interferon-γ (10 ng/ml) in the presence or absence of CO-RMs under hypoxic conditions (95% N2/5% CO2) using an established protocol previously described by our group (Motterlini et al., 2000). Cells were exposed to hypoxia for 12 h and then reoxygenated in a normoxic environment for an additional 24 h. CORM-3 or iCORM-3 was applied to cells prior to placing them in the hypoxic chamber, and no addition was made during the reoxygenation period. Nitrite production and TNF-α levels were also determined from medium samples at the end of the incubation period. In addition, cell viability was assessed by different methods at the end of the experiments.
Determination of Nitrite Levels. Nitrite levels were determined using the Griess method as previously described by our group (Sawle et al., 2005). Briefly, the medium from treated cells cultured in 24-well plates was removed and placed into a 96-well plate (50 μl per well). The Griess reagent was added to each well to begin the reaction, the plate was shaken for 10 min, and the absorbance was read at 550 nm on a Molecular Devices VERSAmax plate reader (Molecular Devices, Sunnyvale, CA). The nitrite level in each sample was calculated from a standard curve generated with sodium nitrite (0-300 μM in cell culture medium).
Measurement of TNF-α Production. The level of TNF-α present in each sample was determined using a commercially available enzyme-linked immunosorbent assay kit (Mouse TNF-α Quantikine immunoassay kit; R&D Systems Europe, Oxford, UK) (Sawle et al., 2005). The assay was performed according to the manufacturer's instructions. Briefly, cell culture supernatants (n = 5 per each group) were collected immediately after the treatment and spun at 13,000g for 2 min to remove any particulates. The medium was added to a 96-well plate precoated with affinity-purified polyclonal antibodies specific for the mouse TNF-α. An enzyme-linked polyclonal antibody specific for mouse TNF-α was added to the wells and left to react for 2 h followed by a final wash to remove any unbound antibody enzyme reagent. The intensity of the color detected at 450 nm (correction wavelength, 570 nm) was measured after addition of a substrate solution and was proportional to the amount of TNF-α produced.
Cell Viability. Cell viability was determined using an Alamar Blue assay kit (Serotec, Oxford, UK) and carried out according to the manufacturer's instructions as previously reported by us (Sawle et al., 2005). The assay is based on the detection of metabolic activity of living cells using a redox indicator, which changes from an oxidized (blue) form to a reduced (red) form. The intensity of the red color is proportional to the metabolism of the cells, which is calculated as the difference in absorbance between 570 and 600 nm and expressed as a percentage of control. An assay for the release of lactate dehydrogenase (LDH) activity was also performed using a cytotoxicity detection kit according to manufacturer's instructions (Roche Diagnostics, Indianapolis, IN). Briefly, at the end of the incubation period, cell supernatants (n = 6 per each group) were collected, and any cell residue was removed by centrifugation at 250g. The reaction mixture (which is composed of the catalyst and the dye solution) was then added to the cell-free supernatant and incubated at room temperature for 15 min, after which the absorbance was measured at 490 and 690 nm. LDH activity was expressed as percentage of maximal LDH release, which was obtained by lysis of cells with Triton X-100 (1% in Dulbecco's modified Eagle's medium at 25°C).
Statistical Analysis. Differences among the groups were analyzed using one-way analysis of variance and Bonferroni tests. Values were expressed as mean ± S.E.M., and differences between groups were considered to be significant at p < 0.05.
Results
CORM-3 Attenuates Thrombin-Induced Nitrite Release in BV-2 Microglia under Normoxic and Hypoxic Conditions. Thrombin is a well known inducer of NO release from microglia under normoxic conditions (Ryu et al., 2000), a phenomena responsible for its devastating effect on neurons. We examined the effects of ruthenium-based CORMs on this response. Preliminary experiments were conducted to determine the most effective concentration of CORM-3 to reduce thrombin [or interferon-γ (IFN-γ)]-mediated increase in nitrite production. We found that CORM-3 (10-100 μM) significantly reduced nitrite increase in a concentration-dependent manner (data not shown). Since the inhibition of nitrite production by 75 μM CORM-3 was not markedly different from the one elicited by 100 μM (82.6 ± 0.7% versus 90.2 ± 0.8%, respectively) and because at 75 μM we did not observe any change in cell viability on microglia, neurons, and astrocytes, this ideal concentration was used in all future experiments. As shown in Fig. 1A, exposure to thrombin for 24 h markedly induced nitrite release in BV-2 cells. This response was significantly reduced by CORM-3 but not by the inactive compound (iCORM-3), indicating that such effect can be attributed to the liberated CO. It is important to note that cell viability was preserved in all the treatments performed (Fig. 1B), suggesting that the effect of CORM-3 and the consequent release of CO are not toxic to the cells. The increase in NO production (nitrite levels) induced by thrombin was further augmented following exposure of BV-2 cells to hypoxia/reoxygenation (Fig. 2A). Hypoxia/reoxygenation also caused a significant cellular damage that was exacerbated by the presence of thrombin (Fig. 2B). Interestingly, CORM-3 (but not iCORM-3) promoted a significant reduction in nitrite production induced by thrombin under hypoxic conditions and significantly reduced the consequent reoxygenation-mediated cellular damage (Fig. 2, A and B).
CORM-3 Attenuates Thrombin- and Interferon-γ-Induced Nitrite Release in BV-2 Microglia under Normoxic and Hypoxic Conditions. IFN-γ is a strong microglial stimulant produced mainly by the lymphocytes and increases the microglial ability to induce and propagate inflammation (Blasko et al., 2001). It is commonly used alone or in combination with other stimuli (e.g., LPS) to simulate and investigate neuroinflammation. We examined the effect of IFN-γ on BV-2 cells in the presence of thrombin. Predictably, the combination of thrombin with IFN-γ amplified nitrite production (Fig. 3A), and CORM-3 once again significantly attenuated this effect. Although iCORM-3 partially affected nitrite levels, the effect of CORM-3 was far more pronounced (p < 0.01, CORM-3 versus iCORM-3), and no significant cellular damage was detected by all the treatments (Fig. 3B). Similarly, but to a different extent, CORM-3 was able to partially reduce nitrite increase by the combination of thrombin and IFN-γ following hypoxia/reoxygenation (Fig. 4A). Under these conditions, a marked increase in cell injury (LDH release) was observed, and both CORM-3 and iCORM-3 partially reduced the toxic effect (Fig. 4B).
CORM-3 Attenuates Thrombin- and Interferon-γ-Induced TNF-α Production in BV-2 Microglia under Normoxic and Hypoxic Conditions. TNF-α is a key immunomodulatory and proinflammatory cytokine that is upregulated after stroke. It is believed to promote cellular injury during and after the stressful events (Wang et al., 2004a) and contributes to cellular apoptosis. In our setting, incubation of BV-2 microglia with thrombin plus IFN-γ for 24 h resulted in a significant increase in TNF-α production (Fig. 5A). This response was markedly reduced by CORM-3 and not affected by the inactive form. Treatment of cells for 12 h under hypoxic conditions followed by 24 h of reoxygenation in the presence of thrombin and IFN-γ amplified TNF-α production (Fig. 5B). Addition of CORM-3 prior to hypoxia reoxygenation caused a significant decrease in TNF-α production mediated by the combination of thrombin and IFN-γ, an effect that was not reproduced by iCORM-3.
The Anti-Inflammatory Effect of CORM-3 Is Independent of the Endogenous Activity of Heme Oxygenase. To test whether endogenous CO contributes to the anti-inflammatory action of CORM-3, we used the heme oxygenase inhibitor tin protoporphyrin-IX (SnPPIX; 10 μM). Blockade of heme oxygenase activity significantly increased the baseline production of nitrite in response to the combination of thrombin and IFN-γ (Fig. 6A), an effect also seen in the presence of CORM-3 alone (data not shown). However, SnPPIX did not seem to affect the anti-inflammatory activity of CORM-3 and did not change the effect of the CO-depleted compound (iCORM-3), further supporting the hypothesis that CO liberated by CORM-3 is responsible for the observed effects. Notably, there was no cytotoxicity related to the use of CORM-3, iCORM-3, or tin protoporphyrin-IX (Fig. 6B). It has to be noted that the presence of SnPPIX reduced LDH release promoted by the combination of thrombin and IFN-γ.
Contribution of Mitogen-Activated Protein Kinases (MAPKs) on the Anti-Inflammatory Activity Mediated by CORM-3. In our model of thrombin-interferon-induced inflammation, we found that the anti-inflammatory effect of CORM-3 was potentiated by blockade of phosphatidyl inositol 3 kinase (PI3K) and partially by extracellular signalregulated kinase (ERK). Interestingly, ERK inhibitor PD98059 seemed to selectively augment CO-RM's inhibitory effect on TNF-α release (Fig. 7B) without affecting NO production (Fig. 7A). On the other hand, PI3K inhibitor LY294002 potentiated the anti-inflammatory activity of CORM-3 in reducing both NO production and TNF-α release (Fig. 7, A and B). We did not observe any contribution by other classical MAPK pathways because P38 (SB203580) and JNK (SP600125) inhibitors did not alter the CORM-3-mediated attenuation of NO production and TNF-α release in response to thrombin and INF-γ (Fig. 7, A and B).
Discussion
CO is a pivotal signaling mediator that regulates a variety of cellular functions in mammals. Due to its versatility and pleiotropic properties, it is becoming apparent that this gas produced endogenously by the enzyme heme oxygenase could be exploited for treating different disorders and pathological states. The importance of CO as a potential therapeutic agent is exemplified by the recent development of CO-RMs, a new class of chemicals that have been specifically designed to carry and deliver small amounts of CO into biological environments (Motterlini et al., 2003, 2005a, b). Previous studies using water-soluble CO-RMs have demonstrated the ability of these compounds to exert vasodilatory and hypotensive effects (Motterlini et al., 2005a), mitigate ischemia-reperfusion injury (Motterlini et al., 2002; Clark et al., 2003), prevent organ graft rejection (Clark et al., 2003), and attenuate the inflammatory response in stimulated macrophages (Sawle et al., 2005). Here, we extend our previous investigations by showing that CORM-3, a transition metal carbonyl that is fully soluble in aqueous solutions and rapidly liberates CO, modulates neuroinflammation in microglia cells.
It is known that cerebral ischemia triggers a strong inflammatory response that involves early recruitment of microglia, which proceeds and predominates over macrophage infiltration in brain tissue (Schilling et al., 2003). The brain is particularly sensitive to oxidative stress due to its high rate of oxygen consumption, elevated lipid content, and increased rate of ROS generation. Although the initial insult to neurons may not be caused directly by ROS, the resulting events mediated by increased free radicals production can cause secondary damage that is far greater and is pathologically relevant to neurodegeneration (Crack and Taylor, 2005). As a result of breaching the blood brain barrier, serum factors such as thrombin leak into the brain to come into direct contact with neurons and astroglia. The presence of thrombin in direct contact with brain tissue induces inflammatory infiltration, brain edema, and reactive gliosis (Nishino et al., 1993). Thrombin has a direct toxic effect on neurons by inducing apoptosis (Smirnova et al., 1998) and an indirect damaging effect through activating microglia (Carreno-Muller et al., 2003), leading to stimulation of NO and ROS production (Ryu et al., 2000; Kim et al., 2002; Hanisch et al., 2004). As a consequence of the change in the redox state of these cells, cytokines such as TNF-α and IFN-γ are released, and this further contributes to the exacerbation of tissue damage (Rothwell, 1997; Kim et al., 2002).
In the present study, we found that stimulating BV-2 microglia cells with thrombin results in increased NO production (nitrite levels), and this effect is markedly attenuated by addition of CORM-3. The fact that the inactive control (iCORM-3), which does not release CO, was far less effective in reducing nitrite levels after stimulation with thrombin emphasizes the direct involvement of CO in the observed effect. It needs to be pointed out that the partial beneficial effect of iCORM-3 observed in some experiments, particularly on cell viability, might be due either to the compound itself and/or to the residual carbonyl groups still present in the molecule (Clark et al., 2003; Motterlini et al., 2005a). These data are in agreement with a previous report showing that CORM-3 significantly decreased NO production as a consequence of induced iNOS expression following exposure to endotoxin (Sawle et al., 2005). Interestingly, we also found that exposure of BV-2 cells to hypoxia/reoxygenation in the presence of thrombin caused a much higher production of nitrite compared with thrombin alone, and CO liberated from CORM-3 was still very effective in reducing this inflammatory response. It is known that NO generated from iNOS exacerbates ischemia-reperfusion injury to the brain and that NO and other cytokines are highly increased in microglia in response to hypoxia (Park et al., 2002). The fact that CORM-3 was very effective in reducing nitrite production, and attenuated LDH release in microglia in the presence of thrombin under hypoxic conditions is indicative of the potential therapeutic effect of CO in the context of brain injury mediated by ischemic events. The beneficial effects of CORM-3 were also manifested when INF-γ, one of the most potent stimulants of microglia, was introduced in our experimental protocol to simulate the synergistic effect of proinflammatory cytokines in maximizing cellular damage and driving multiple destructive cascades. Indeed, we found that the combination of thrombin and INF-γ synergized NO production in BV-2 cells under normoxic conditions, an effect that was markedly attenuated by CORM-3 but not the inactive compound. The protective effect of CORM-3 in reducing nitrite levels by the combination of thrombin and INF-γ was only partially observed under hypoxia; this might be explained by the fact that the combination of cytokines and low oxygen tension is too extreme to BV-2 microglia because it caused severe cellular damage (>90% LDH release), which is likely to be irreversible. However, the salutary effects of CORM-3 were clearly evident when TNF-α levels were measured in microglia exposed to thrombin plus INF-γ either under normoxia or hypoxia reoxygenation. In both conditions, we found that CORM-3 significantly reduced TNF-α in response to the proinflammatory cytokines; this effect can be attributed to CO because iCORM-3 was far less effective. These data are in agreement with and supported by a previous study from our group showing that CO-RMs can effectively mitigate the TNF-α release from murine macrophages stimulated in vitro with lipopolysaccharide (Sawle et al., 2005). Furthermore, blockade of endogenous heme oxygenase-derived CO, bilirubin, and biliverdin by tin protoporphyrin-IX (10 μM) did not affect CORM-3-induced anti-inflammatory or cytoprotective activity. The bioactive properties of CO emerging from the present investigation are in line with the ability of microglia to express the CO-producing enzyme HO-1 (Kitamura et al., 1998) and are sustained by the neuroprotective role of HO-1 against cerebral ischemia, an effect that is likely mediated by CO (Panahian et al., 1999). It has to be noted that the presence of SnPPIX reduced LDH release promoted by the combination of thrombin and IFN-γ. Although this could be counterintuitive, we have previously shown that SnPPIX, apart from acting as a competitive inhibitor of heme oxygenase activity, also possesses some scavenging properties against free radicals, and this could explain SnPPIX's favorable effect on cell viability obtained in this study (Foresti et al., 1999).
MAPKs are believed to play a significant role in the regulation of inflammatory response in cerebral ischemia (Wang et al., 2004b). However, the contribution of each individual pathway is rather difficult to define because of the complex interactions between all the involved pathways that can directly or indirectly affect the final outcome. MAPK families can be classified as stress activated (P38/JNK) or ERK. In addition, PI3K is now proven to play a crucial role in cell survival signals and modulation of iNOS expression (Hwang et al., 2004). Recent studies have shown the involvement of P38, JNK, ERK, and PI3K pathways in the brain tissue response to ischemia (Takagi et al., 2000; Barone et al., 2001; Ferrer et al., 2003; Crack and Taylor, 2005).
The importance of each kinase varies with the type of stimulus. In ischemia reperfusion-induced inflammation, P38 seems to play a major role, whereas JNK's role is rather permissive or modulatory related more to cytokine regulation. In addition, although ERK is not stress-induced, it appears to participate, at least partly, in the production of inflammatory cytokines (Toledo-Pereyra et al., 2004), whereas thrombin-induced NO release is inhibited by p38 (SB203580) and ERK (PD98059) inhibitors (Ryu et al., 2000). However, SB203580 has no effect by itself on IFN-γ-mediated NO release and needed the addition of PD98059 to reach a significant inhibitory effect in microglia (Han et al., 2002; Shen et al., 2005). On the other hand, PI3K has a very potent effect on IFN-γ-induced NO production and iNOS gene in microglia, an effect that is likely mediated by signal transducer and activator of transcription 1 and partly by nuclear factor-κB (Hwang et al., 2004).
In our settings, P38 inhibition had a minimal effect on NO production and TNF-α release (the apparent increase in TNF-α level is due to decreased cell viability in the presence of SB203580; data not shown). JNK inhibition reduced both inflammatory markers (the increase in TNF-α was also associated with an increased cell injury; data not shown). However, neither P38 nor JNK inhibition seemed to affect CORM-3-mediated anti-inflammatory action. Although ERK inhibition had a relatively small inhibitory effect on both NO and TNF-α release, its effect on CORM-3 activity was rather selective, facilitating CORM-3 inhibitory effect on TNF-α release but not NO production, which can be attributed to the partial permissive role for ERK in cytokine production. On the other hand, PI3K inhibition had a significant inhibitory effect by itself and potentiated the effect of CORM-3 in reducing both NO production and TNF-α release. More mechanistic studies are required to better understand the interaction between the signal transduction pathways involved; interestingly, a previous report in line with our findings showed that the expression of the main inflammatory mediators, iNOS and cyclooxygenase-2, involves activation of multiple signaling proteins where PI3K appears to play a major role (Jang et al., 2005).
In conclusion, our results provide strong evidence that CO liberated from CO-RMs significantly reduces the inflammatory response in BV-2 microglia stimulated by thrombin, cytokines, and hypoxia reoxygenation, an effect that is particularly relevant to stroke-related neuroinflammatory response and cellular damage. CORM-3 seems to affect a spectrum of pathways in the cellular inflammatory response that extends from direct enzymatic inhibition to interaction with MAPK pathways. Further research needs to be conducted to explore the exact mechanism(s) by which CO transduces the signal into an anti-inflammatory response and how CO-RMs can be best exploited for maximizing the neuromodulatory activity of CO in the context of stroke and protection of brain tissue against ischemia-reperfusion injury.
Acknowledgments
We thank Rosario Donato (University of Perugia, Italy) for providing the BV-2 microglia cell line and Tony R. Johnson for the synthesis of CORM-3. This research was supported by a grant from the Henry Smith Charity (to R.M.).
Footnotes
-
doi:10.1124/jpet.106.104729.
-
ABBREVIATIONS: iNOS, inducible nitric oxide synthase; ROS, reactive oxygen species; HO, heme oxygenase; CO-RM, carbon monoxide-releasing molecule; CORM-3, tricarbonylchloro(lglycinato)ruthenium (II); SP600125, 1,9-pyrazoloanthrone; iCORM-3, inactivated CORM-3; SB203580, 4-(4-fluorophenyl)-2-(4-methylsulfinylphenyl)-5-(4-pyridyl)1H-imidazole; PD98059, 2′-amino-3′-methoxyflavone; LY294002, 2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride; TNF, tumor necrosis factor; LDH, lactate dehydrogenase; IFN-γ, interferon-γ; SnPPIX, tin protoporphyrin-IX; MAPK, mitogen-activated protein kinase; PI3K, phosphatidyl inositol 3 kinase; ERK, extracellular signal-regulated kinase; JNK, c-Jun NH2-terminal kinase; Thr, thrombin.
- Received March 16, 2006.
- Accepted June 12, 2006.
- The American Society for Pharmacology and Experimental Therapeutics