| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University, Baltimore, Maryland 21217
Abstract
Prostaglandins (PGs) originate from the degradation of membranar arachidonic acid by cyclooxygenases (COX-1 and COX-2). The prostaglandin actions in the nervous system are multiple and have been suggested to play a significant role in neurodegenerative disorders. Some PGs have been reported to be toxic and, interestingly, the cyclopentenone PGs have been reported to be cytoprotective at low concentration and could play a significant role in neuronal plasticity. They have been shown to be protective against oxidative stress injury; however, the cellular mechanisms of protection afforded by these PGs are still unclear. It is postulated that the cascade leading to neuronal cell death in acute and chronic neurodegenerative conditions, such as cerebral ischemia and Alzheimer’s disease, would be mediated by free radical damage. We tested the hypothesis that the neuroprotective action of cyclopentanone could be caused partially by an induction of heme oxygenase 1 (HO-1). We and others have previously reported that modulation of HO total activity may well have direct physiological implications in stroke and in Alzheimer’s disease. HO acts as an antioxidant enzyme by degrading heme into iron, carbon monoxide, and biliverdin that is rapidly converted into bilirubin. Using mouse primary neuronal cultures, we demonstrated that PGs of the J series induce HO-1 in a dose-dependent manner (0, 0.5, 5, 10, 20, and 50 µg/ml) and that PGJ2 and dPGJ2 were more potent than PGA2, dPGA2, PGD2, and PGE2. No significant effects were observed for HO-2 and actin expression. In regard to HO-3 expression found in rat, with its protein deducted sequence highly homologous to HO-2, no detection was observed in HO-2-/- mice, suggesting that HO-3 protein would not be present in mouse brain. We are proposing that several of the protective effects of PGJ2 could be mediated through beneficial actions of heme degradation and its metabolites. The design of new mimetics based on the cyclopentenone structure could be very useful as neuroprotective agents and be tested in animal models of stroke and Alzheimer’s disease.
Key Words: Alzheimer’s disease • antioxidant • cerebral ischemia • free radical • iron • oxidative stress • prostaglandin • neurodegeneration • stroke
Prostaglandins (PGs) were first discovered in the 1930s and it is now known that they have various intrinsic biological actions. PGs are generated from the enzymatic degradation of 20-carbon arachidonic acid by cyclooxygenases. There are two isoforms of cyclooxygenase: COX-1 is expressed constitutively in most tissues and is present under normal conditions at very low levels in the brain. COX-2, the inducible isoform, is found acutely expressed in several cell types after head injury, in cerebral ischemia, and in Alzheimer’s disease. Ischemic infarct, especially the reperfusion phase, is associated with significant formation and release of arachidonic acid and its metabolites. Within the PG family, the PGs of A and J series contain a cyclopentenone ring structure that is characterized by 
,ß-unsaturated carbonyl group (1). Evidence indicates that this reactive unsaturated group is required for many of the biological actions (1, 2). The cyclopentenone PGs PGA2 and PGJ2 are made from dehydration within the cyclopentane ring of PGE2 and PGD2, respectively. Note that PGD2 is the major prostanoid in the murine central nervous system (3, 4). Some of the effects of these eicosanoids have been described as an increase in blood flow, inhibition of platelet function, and inhibition of the activation/extravasation of granulocytes (5). Many of these effects are receptor-mediated through either stimulation of phospholipase C to produce inositol 1, 4, 5-trisphosphate (IP3) and diacylglycerol or through modulation of adenylyl cyclase via the guanine nucleotide-binding regulator proteins (G-proteins) (6–14). In contrast with other PGs, most of the cyclopentenones interact with other specific cellular targets. These cyclopentenone PG members have been shown to be "cytoprotective" and have antineoplastic, anti-inflammatory and antiviral properties (see review in ref. 1), although the mechanisms of action are still unclear.
In this study we investigated whether some of the biological actions of these cyclopentenone PGs could be partially mediated through modulation of HO. HO catalyzes the cleavage of the heme to form iron, carbon monoxide, and biliverdin/bilirubin. We and others have previously shown that modulation of HO activity could afford neuroprotection (see Refs. 15–18 for reviews). There are abundant heme-containing enzymes in mitochondria and endoplasmic reticulum, which presumably undergo turnover during oxidative stress; as such, HO would play an important role in assuring that pro-oxidant heme does not increase to toxic levels. Furthermore, in cultured neurons, HO could make biliverdin/bilirubin at sufficient concentrations to act as antioxidants (19). A specific significant increase of HO levels is observed after treatment of primary neuronal cultures with PGA2 and PGJ2, or their derivatives, while no induction is detectable with PGE2 and PGD2.
This family of compounds offers unique mechanisms of action and their roles and potential therapeutic applications in acute and/or chronic neurological conditions are still under investigation. We are proposing that several of the protective effects of PGA2 and PGJ2 could be mediated through beneficial actions of heme degradation and its metabolites.
Materials and Methods
Primary Cultures of Neuronal Cells.
Cultures of cortical neuronal cells were obtained from 17-day-old embryos of timed pregnant mice. Animal care was administered according to protocols and guidelines of the Johns Hopkins University Animal Care Committee. Cultures were prepared in serum-free conditions. Neurons were plated onto poly-D-lysine coated 24-well dishes at a density of 1 x 106 cells/well in the B27 supplement HEPES-buffered high glucose Neurobasal medium, as previously described (19). Half of the initial medium was removed at day 4 and replaced with the prewarmed medium. Cells were maintained in growth medium at 37°C in a 95% air/5% CO2 humidified atmosphere until the day of experiment. Materials used for cell cultures were obtained from InvitroGen (Carlsbad, CA). Unless stated otherwise, all other chemicals were purchased from Sigma Co. (St. Louis, MO).
Experimental Treatments.
After 8 days in culture, cells were incubated in fresh medium containing PGs obtained from Cayman Chemical (Ann Arbor, MI). Experimental treatments with PGs, or with vehicle-control, were conducted in the B27 minus antioxidant supplement (InvitroGen) HEPES-buffered high-glucose Neurobasal medium. All experiments were conducted under a dim light to avoid heme pigment photodegradation.
Assessment of Cell Survival.
After treatments, neurons were maintained for an additional period of 23 hr and their survival was assessed by phase-contrast microscopy with Trypan Blue exclusion assay and quantified using MTT [(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide)] (Sigma) colorimetric assay. MTT is an indicator of the mitochondrial activity of living cells and is widely used as an index of cell survival (19). After a 2-hr incubation at 37°C, living cells containing MTT formazan crystals were solubilized in a solution of anhydrous isopropanol, HCl 0.1 N, and 0.1% Triton X100. The optical density was determined at 570 nm. Experiments were performed in triplicates and repeated with at least three separate batches of cultures.
Preparation of Mouse Brain Homogenates.
Cortical brain region from mouse brains was dissected on ice, rinsed with cold PBS (phosphate buffered saline), add 600 µl of lysis buffer [phosphate-buffered saline, containing protease inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN) and 0.1 mM phenylmethylsulfonyl fluoride (PMSF; Sigma)]. The mixtures were sonicated for 7 sec twice, and then centrifuged at 7000g at 4°C for 30 min. The supernatants collected and centrifuged at 100,000g at 4°C for 40 min. The supernatant (cytosol-enriched fraction) and pellet (microsome-enriched fraction) were collected. Microsomal fractions were resuspended into 600 µl of lysis buffer, and protein concentration was determined and 25 µg was loaded onto the gel.
Western Blot Analysis.
Neuronal cultures were solubilized with 250 µl of lysis buffer (50 mM Tris-HCl, pH 7.4; 150 mM NaCl, 0.5% Triton X-100) including protease inhibitor cocktail (Roche) on ice for 30 min and then centrifuged for 10 min at 12,000g then collected the supernatant. Protein quantification was accomplished using BCA assay (Pierce, Rockford, IL). Western blots (sodium dodecyl sulfate polyacrylamide gel electrophoresis) were performed using 12% gels (Novex, San Diego, CA) and proteins were transferred to nitrocellulose membranes (Novex) (19). Membranes were blocked for 1h at RT with 5% milk in PBS with 0.1% Tween 20 before incubation at 4°C overnight with primary antibodies. Blots were washed and incubated with second antibodies for 1 hr at room temperature and then developed by ECL (Amersham Biosciences, Piscataway, NJ). Gels were stained with Ponceau S Solution to verify that equal amounts of proteins were loaded in each lane. Primary polyclonal antibodies to HO-1 and HO-2 were obtained from StressGen Inc. (Victoria, BC) and anti-actin was obtained from Sigma, and used at a dilution of 1:3,500, 1:2,500, and 1:5,000 respectively.
Results and Discussion
Our data demonstrate that the cyclopentenones (Fig. 1
), and especially the ones of the J series can induce HO-1 protein expression in mouse primary cortical cells (Figs. 2 and 3). Our results indicate that the HO-1 induction effect of dPGJ2 was dose dependent. It was observed as low as 0.5 µM with a maximal effect at 20 µM after a 6-hr incubation (Fig. 2). Previous studies have shown that induction of HO-1 could act as an antioxidant system. Moreover, no significant effect was observed on the HO-2 expression levels, which was also confirmed by similar expression levels of actin (Fig. 1). We also observed that at these concentrations no toxicity effect would be measured, and this up to 24 hr of treatment (data not shown). It has been previously reported that a cDNA coding for a transcript, with a predicted amino acid sequence close to 90% similar to HO-2 and a similar pattern of distribution, could be present in rat tissues (20). This transcript was referred as HO-3, although its mRNA expression is present in several tissues, its protein expression has not been detected. Using different polyclonal antibodies against HO-2 (raised against peptides which shares approximately 75% homology and against a rat purified native HO-2), we could not detect the presence of other cross-reacting proteins in mouse brain tissues obtained from HO-2-/- mice (Fig. 4). This would suggest that, at least in mouse, HO3 protein would not be detectable.
|
|
|
|
). We questioned whether HO activity could play a role in dPGJ2’s neuroprotective function.
|
, HO degrades the pro-oxidant heme. Free heme in the cell can be rapidly generated from heme-containing proteins/enzymes (such as myoglobin, catalase, glutathione peroxidase, cytochrome, soluble guanylate cyclase, superoxide dismutase, nitric oxide synthase, etc.). For example, hypoxia during ischemic injury may trigger micromolar concentrations of heme to be released in the intracellular pool. HO, through rapid degradation of pro-oxidant heme, can limit its capacity to enter into a generation of free radical cycle. HO, by directly metabolizing heme, acts by itself as a potent antioxidant enzyme. The observation reported here that PGJ2 can increase HO-1 levels in neuronal cultures is likely to be an important factor in degradation of the pro-oxidant heme. In addition, this HO enzymatic reaction generates several catalytic metabolites with intrinsic actions (i.e., iron, carbon monoxide and biliverdin). The degradation of hemin/heme by HO generates iron. Then, modulation of HO activity could be a limiting factor in controlling the rate by which iron is being eliminated from the cells (21). Controlling the iron homeostasis within the cell is essential. For example, free iron is a key element in the traditional Fenton reaction, which, by reaction with H2O2, generates free radicals. The iron cellular homeostasis is a complex system and is regulated by abundant proteins, some of which are being identified and characterized. The strict control of the iron level within a cell is critical, and we believe that our previous work suggests that HO is an important enzyme in this regard. The direct increase of HO-1 by dPGJ2 is likely to directly affect the intracellular iron levels. In the opposite scenario, one would expect that a decrease in HO activity would be sufficient to change the iron levels within cells. In fact, this is what was revealed by the use of the HO-1 knockout mice generated by Poss and Tonegawa, in which they observed accumulation of iron in several organs (22). PGJ2, and its analogues, could then be protective by acting indirectly through the modulation of iron free levels via other iron binding proteins. Numerous iron-binding proteins have been found and their expression can modulate the iron intracellular free pool (23). To cite only one example, ferritin in the cell can sequester free iron, and its intracellular levels can be very rapidly induced by free iron (24), which could represent by itself another pathway by which dPGJ2 can be protective. The therapeutic implications of controlling iron levels are numerous. For instance, one study revealed that the administration of desferroxiamine, a trivalent ion chelator, over a two-year period slowed the clinical progression of symptoms in AD (25). By demonstrating here that dPGJ2 significantly alters HO-1 levels, it could subsequently protect several cells and organs against iron-mediated toxicity.
The opening of the heme porphyrin ring by HO generates biliverdin. We and others have demonstrated that biliverdin by itself could be a potent antioxidant (26–28). It is generally believed that biliverdin is almost immediately converted into bilirubin by biliverdin reductase (29). Moreover, it has been recently proposed that the production of bilirubin (BR) is dependent on the autophosphorylation of biliverdin reductase (30). BR is well known as a toxic agent in infants because an accumulation of micromolar concentrations in the brain tends to aggregate. Yellowing of several brain regions is a landmark of kinecterus. Similarly, we have observed in neuronal cultures at high micromolar concentrations that BR can aggregate and stick to cellular membranes (28), lending a yellow appearance to the cells. Aggregates of BR incorporated to the cell membrane are likely to affect the normal cellular functions. However, when BR is used at physiological levels, it is protective against oxidative stress injury (19, 31–33). Interestingly, BR, by reducing free radicals, would partially be converted into biliverdin (Fig. 5). A reducing loop can then be an effective way to act as a radical scavenger; in addition, its has been suggested that BR metabolites can have intrinsic vasodilating properties (34).
Carbon monoxide (CO) is a gas that in cells is mostly generated by the degradation of heme from HO (35, 36). CO can travel freely throughout intracellular and extracellular compartments. The literature regarding the role of carbon monoxide is complex, with many controversies that have yet to be resolved (18, 37–41). CO is known to be toxic and the inhalation of high concentrations can cause death. Some of this toxicity has been attributed to the fact that it saturates hemoglobin and decreases the ability to transport oxygen. The affinity of CO to several proteins is generally lower than nitric oxide, though its longer half-life could be a key factor in modifying several key enzymes containing a heme moiety (42). At a cellular level, physiological levels of CO generated from degradation of intracellular heme are then likely to have biological actions on several heme-containing proteins. For example, CO may act as a vasodilator by binding to soluble guanylate cyclase and modulate its activity (18, 38). CO can also act by opening calcium-activated potassium channels (KCa channels; ref. 43). CO has also been reported to have specific anti-inflammatory and antiapoptotic effects (44, 45). Further investigations are necessary to clarify the cellular effect of physiologic concentrations of CO. dPGJ2, by increasing HO activity, would generate more CO, within physiological levels, allowing cells and tissues to benefit from many of its biological actions especially in a scenario where blood flow is reduced and cell survival is triggered.
Puzzling effects of the PGs have been reported (46), sometimes good and sometimes bad. We are proposing here that significant induction of HO-1 in neuronal cultures would participate in the protective actions of dPGJ2. At this point, we have not yet identified the specific cascade that triggers HO-1 upregulation: influence of HO-1 mRNA levels, longer mRNA half-life or longer HO-1 half-life (47). An increase of HO activity in cells provides a way to degrade free heme generated from heme-containing proteins, and generation of the different metabolites (i.e., iron, biliverdin/bilirubin, carbon monoxide). These are likely to effect cell survival and may contribute to the beneficial effects of dPGJ2 in therapeutic situations. Design of new mimetics based on the cyclopentenone structure could be very useful as neuroprotective agents (48). Further studies will be necessary to determine whether the cyclopentenone PGs or analogues can be of use as preventive agents against acute neurodegenerative conditions, especially in limiting damage following a stroke, or reducing the progression of diseases with chronic neurodegeneration like Alzheimer’s disease and possibly even retarding signs of "normal" aging.
Acknowledgments
The authors wish to thank James M. Carlisle for his assistance.
Footnotes
This work was supported by a Grant-in-Aid from the American Heart Association, the Alzheimer’s Association and the American Heath Assistance Foundation.
1 To whom requests for reprints should be addressed at Johns Hopkins University, School of Medicine, Department of Anesthesiology and Critical Care Medicine, 600 N. Wolfe Street, (Blalock 1404A), Baltimore, MD 21217. E-mail: sdore{at}jhmi.edu 
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
