Calcium in the Mechanism of Ammonia-Induced Astrocyte Swelling

Abstract

Brain edema, due largely to astrocyte swelling, is an important clinical problem in patients with acute liver failure. While mechanisms underlying astrocyte swelling in this condition are not fully understood, ammonia and associated oxidative/nitrosative stress (ONS) appear to be involved. Mechanisms responsible for the increase in reactive oxygen/nitrogen species (RONS) and their role in ammonia-induced astrocyte swelling, however, are poorly understood. Recent studies have demonstrated a transient increase in intracellular Ca²⁺ in cultured astrocytes exposed to ammonia. As Ca²⁺ is a known inducer of RONS, we investigated potential mechanisms by which Ca²⁺ may be responsible for the production of RONS and cell swelling in cultured astrocytes after treatment with ammonia. Exposure of cultured astrocytes to ammonia (5 mM) increased the formation of free radicals, including nitric oxide, and such increase was significantly diminished by treatment with the Ca²⁺ chelator BAPTA-AM. We then examined the activity of Ca²⁺-dependent enzymes that are known to generate RONS and found that ammonia significantly increased the activities of NADPH oxidase (NOX), constitutive nitric oxide synthase (cNOS) and phospholipase A2 (PLA2) and such increases in activity were significantly diminished by BAPTA. Pretreatment of cultures with 7-nitroindazole, apocyanin and quinacrine, respective inhibitors of cNOS, NOX and PLA2, all significantly diminished RONS production. Additionally, treatment of cultures with BAPTA or with inhibitors of cNOS, NOX and PLA2 reduced ammonia-induced astrocyte swelling. These studies suggest that the ammonia-induced rise in intracellular Ca²⁺ activates free radical producing enzymes that ultimately contribute to the mechanism of astrocyte swelling.

INTRODUCTION

Hepatic encephalopathy (HE) is the neurological disorder resulting from severe liver failure. In its acute form (acute liver failure, ALF) it occurs following massive liver necrosis due to viral hepatitis, acetaminophen toxicity, or exposure to other hepatotoxins. It is characterized by an abrupt onset of delirium, seizures and coma and it has a high mortality rate (Capocaccia and Angelico 1991). Elevated levels of ammonia are consistently found in patients with HE, and such rise in ammonia levels has been strongly implicated in disease pathogenesis (Hazell and Butterworth 1999). Astrocytes appear to represent the principal neural cells affected by ammonia toxicity (Norenberg 1996)

Brain edema and associated increase in intracranial pressure leading to brain herniation are important complications of ALF and are the leading cause of death in these patients (Blei 1991). Although the basis for the edema remains to be determined, swelling of astrocytes represents a major component of the edema (Martinez 1968, Traber et al. 1989, Kato et al. 1992). Swelling of cerebral cortical astrocytes has also been observed after ammonia infusion in animals (Takahashi et al. 1991), and following the application of ammonia to brain slices (Ganz et al. 1989) and to cultured astrocytes (Norenberg et al. 1991).

Oxidative/nitrosative stress (ONS) has been suggested to play a major role in ammonia neurotoxicity. Pathophysiological levels of ammonia were shown to increase reactive oxygen and nitrogen species (RONS) production in experimental hyperammonemia in vivo, as well as in cultured astrocytes (for review, see Norenberg 2003). Reduced levels of antioxidant enzymes and lipid peroxidation in brain were reported in experimental models of hyperammonemia (Kosenko et al. 1997). Additionally, protein carbonylation and protein tyrosine nitration have been documented in cultured astrocytes exposed to ammonia (Schliess et al. 2002, Norenberg et al. 2007, Widmer et al. 2007).

Oxidative stress is also known to result in astrocyte swelling. Free radicals have been shown to cause cell swelling in cultured astrocytes (Chan et al. 1989), as well as in brain slices (Chan et al. 1982). Similarly, nitrosative stress was shown to cause astrocyte swelling in brain slices and in cultured astrocytes (Zielinska et al. 2003, Jayakumar et al. 2006). Although ONS appears to be major factor in astrocyte swelling (Norenberg et al. 2005), the precise means by which ammonia generates RONS and the mechanisms by which RONS contribute to the ammonia-induced astrocyte swelling remain to be defined.

One of the earliest events known to occur in cultured astrocytes following ammonia exposure is a rise in intracellular Ca²⁺ ([Ca²⁺]_i) (Schliess et al. 2002, Norenberg et al. 2003, Rose et al. 2006). Since Ca²⁺ is known to promote RONS production (Chinopoulos and Adam-Vizi 2006), we hypothesized that the ammonia-induced rise in [Ca²⁺]_i might contribute to RONS production and thereby initiate a process leading to astrocyte swelling. One possible means by which Ca²⁺ may generate RONS is through the activation of various Ca²⁺-dependent enzymes. Among these include NADPH oxidase (NOX) (Suzuki et al. 1985), which forms superoxide (Nakamura et al. 1989); constitutive nitric oxide synthase (cNOS) (Mayer et al. 1990), which generates nitric oxide; and the cytosolic form of phospholipase A2 (cPLA2) (Kramer and Sharp 1997) whose product, arachidonic acid, is known to produce free radicals (Cocco et al. 1999).

The present study examined the role of [Ca²⁺]_i and Ca²⁺-dependent enzymes (cNOS, NOX and cPLA2) in the mechanisms of RONS production and cell swelling following ammonia treatment of cultured astrocytes.

MATERIALS AND METHODS

Astrocyte cultures

Primary cultures of rat cortical astrocytes were prepared as described previously Ducis et al. (1990). For details, see supplemental materials.

Measurement of reactive oxygen/nitrogen species (RONS) production

RONS production was measured by the method of Thorburne and Juurlink (1996) using 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate (DCFDA) with minor modifications (Murthy et al. 2001).

Nitric oxide synthase (NOS) activity

Activity of NOS was determined by the conversion of [³H]arginine to [³H]citrulline in the presence (total NOS activity) or absence of Ca²⁺ (iNOS activity) in the reaction mixture, using the method of Bredt and Snyder (1989). For details, see supplemental materials.

NADPH oxidase (NOX) activity

NOX activity was determined by measuring superoxide production employing a chemiluminiscence method (Jaimes et al. 2004). For details, see supplemental materials.

Phospholipase A2 (PLA2) activity

Activity of PLA2 was determined by the method of Poulsen et al. (2007). For details, see supplemental materials.

Cell volume measurement

Astrocyte cell volume was determined using 3-O-methyl-[³H]glucose (OMG, NEN Life sciences, Boston, MA) by the method of Kletzien et al. (1975), as modified for astrocyte cultures by Kimelberg (1987) and Norenberg et al. (1991). For details, see supplemental materials.

Statistical analysis

All experiments performed were repeated 3–5 times using cells derived from different batches of astrocyte cultures. The number of individual culture plates in each group was 5–7 for all experiments. Data of all experiments were subjected to ANOVA followed by Tukey’s pos hoc comparison. A value of p<0.05 was considered significant. Values are mean ± SEM.

RESULTS

Effect of BAPTA on ammonia-induced reactive oxygen/nitrogen species (RONS) generation

Treatment of cultures with ammonia (5 mM NH₄Cl) increased RONS production (Figure 1). The concentration of ammonia used is relevant as similar concentrations are found in brains of animals with acute liver failure (Swain et al. 1992). This treatment results in a rapid decline in ammonia concentration to undetectable levels by 30–60 min (unpublished observations). At 3 min, ammonia increased RONS by 111.4% (p<0.05) and remained elevated for at least 2 h (84.07%) (p<0.05). Pretreatment (10 min) of cultures with BAPTA-AM (25 μM) significantly blocked ammonia-induced RONS production at 3 min (64.2%, p<0.05), 2 h (51.6%, p<0.05) (Figure 1), and >80% at 4 h (data not shown).

Figure 1.

Effect of the Ca²⁺ chelator BAPTA on ammonia-induced reactive oxygen/nitrogen species (RONS) production. Cultures were pretreated (15 min) with BAPTA/AM (25 μM), and then exposed to ammonia (5 mM NH₄Cl). RONS production was determined by DCFDA fluorescence at 3 min and 2 h after ammonia treatment. Values are expressed as arbitrary fluorescence intensity units (AFIU) of DCFDA per mg cell protein. *p<0.01 vs. control; **p<0.05 vs. NH₄+. C, control; BAP, BAPTA/AM.

Effect of ammonia on cNOS, NOX and PLA2 activities

Ammonia increased cNOS activity by 40 and 45% (p<0.05) at 30 min and 2 h, respectively. Pretreatment (15 min) with BAPTA (25 μM) completely blocked the ammonia-induced increase in cNOS activity (Figure 2). Ammonia also significantly increased NOX activity (172.5%, p<0.01) as early as 5 min and such increase progressed to 268.9% (p<0.01) at 2 h. Pretreatment (15 min) of cultures with BAPTA partially diminished NOX activity at 5 min, whereas it was completely blocked at 2 h (Figure 3). Likewise, ammonia resulted in a 73% (p<0.01) increase in PLA2 activity by 30 min, which further increased at 2 h (103%; p<0.01). Pretreatment of cultures with BAPTA completely blocked PLA2 activity at both time points (Figure 4).

Figure 2.

Effect of BAPTA on ammonia-induced constitutive nitric oxide synthase (cNOS) activity. Ammonia (5 mM NH₄Cl) increased cNOS activity at 30 min and 2 h. Pretreatment of cultures with BAPTA/AM (25 μM) completely blocked cNOS activity by ammonia. *p<0.01 vs. control; **p<0.01 vs. NH₄.

Figure 3.

Effect of BAPTA on ammonia-induced NADPH oxidase (NOX) activity. Cultures were exposed to ammonia (5 mM NH₄Cl), and NOX activity was determined. Ammonia significantly increased NOX activity at 5 min and 2 h. Pretreatment (15 min) of cultures with BAPTA significantly diminished NOX activity. *p<0.05 vs. control; **p<0.01 vs. NH₄.

Figure 4.

Effect of BAPTA on ammonia-induced phospholipase A2 (PLA2) activity. Ammonia (5 mM NH₄Cl) increased the activity of PLA2 at 30 min and 2h respectively. Pretreatment of cultures with BAPTA/AM (25 μM) completely blocked PLA2 activity. *p<0.01 vs. control; **p<0.01 vs. NH₄ (NH₄Cl).

Effect of inhibitors of cNOS, NOX and cPLA2 on ammonia-induced RONS generation

Treatment of cultures with ammonia significantly increased RONS production by 112% as early as 3 min, and remained elevated for at least 4 h. Pretreatment of cultures with 7-nitroindazole (7-NI, 50 μM), an inhibitor of cNOS, partially (63.9 %) blocked ammonia-induced RONS production at 3 min, whereas the activity was completely blocked at 2 and 4 h. Apocyanin, an inhibitor of NOX, did not block RONS production at 3 min, but completely blocked it at 2 and 4 h. Likewise, quinacrine, an inhibitor of PLA2, had no inhibitory effect on RONS production at 3 min; however, it partially blocked ammonia-induced RONS generation at 2 and 4 h (46%) (see Figure. S1, supplemental materials).

Effect of BAPTA on ammonia-induced astrocyte swelling

Astrocyte cultures were pretreated (15 min) with BAPTA (25 μM) and cell volume was measured 24 h after exposure to ammonia. Ammonia treatment resulted in a 55% (p<0.01) increase in cell volume which was almost completely blocked by BAPTA (90%, p<0.01) (see Figure. S2, supplemental materials).

Effect of inhibitors of cNOS, NOX and cPLA2 on ammonia-induced astrocyte swelling

Astrocyte cultures were pretreated (15 min) with 7-nitroindazole (7-NI), apocyanin (100–500 μM) or quinacrine (10 μM). Ammonia was added and cell volume was determined at 24 h. Treatment with 7-NI showed a 40% (p<0.05) blockade of astrocyte swelling, whereas apocyanin and quinacrine completely blocked the cell swelling (see Figure. S3, supplemental materials).

DISCUSSION

Brain edema is an important complication of acute ammonia neurotoxicity and ALF (for review, see Blei 1991). Astrocyte swelling (cytotoxic edema) appears to be the major component of brain edema in ALF (Martinez, 1968; Traber et al. 1989, Kato et al. 1992). While mechanisms mediating astrocyte swelling in ALF are not clear, we have recently shown that oxidative/nitrosative stress is involved in such swelling (Jayakumar et al. 2006). Factors responsible for the increase in RONS after ammonia treatment, however, remain poorly understood.

Several studies have shown that a transient rise in [Ca²⁺]_i is one of the earliest events in cultured astrocytes exposed to ammonia (Schliess et al. 2002, Norenberg et al. 2003, Rose 2006). One consequence of elevated levels of [Ca²⁺]_i is the formation of RONS (Chinopoulos and Adam-Vizi 2006). The present study demonstrates that astrocyte cultures exposed to ammonia increased RONS formation, and that such increase was significantly diminished by BAPTA, suggesting that a rise in [Ca²⁺]_i was a major factor in the generation of RONS by ammonia.

While the mechanisms by which elevated levels of [Ca²⁺]_i leads to RONS formation is not well known, several studies have shown that increased [Ca²⁺]_i can activate NOX (Suzuki et al. 1985), and that such activation results in the production of superoxide and H₂O₂ (Nakamura et al. 1989). It has been previously shown that exposure of astrocyte cultures to ammonia stimulates the production of RONS (Murthy et al. 2001), and that RONS production was diminished by superoxide dismutase and catalase (Murthy et al. 2001), suggesting that among the free radicals generated by ammonia include superoxide and H₂O₂. The present study demonstrates that exposure of astrocyte cultures to ammonia significantly increased NOX activity and that treatment of cultures with the NOX inhibitor apocyanin significantly diminished ammonia-induced RONS production. This is in agreement with the findings of Reinehr et al. (2007) who reported increased levels of the NOX subunit p47(Phox) in astrocytes exposed to ammonia, and that the NOX inhibitor apocynin blocked RONS formation.

Another enzyme that can be activated by increased [Ca²⁺]_i is cNOS (Mayer et al. 1990). Our study demonstrating that BAPTA significantly blocked ammonia-induced cNOS activity, indicate that the increase in [Ca²⁺]_i also contributes to NO production. Additionally, astrocyte cultures exposed to 7-nitroindazole (7-NI), an inhibitor of cNOS completely blocked the ammonia-induced RONS formation. Consistent with these findings, various studies have shown that NOS activity, gene expression and NO levels are increased in experimental models of chronic HE (Rao et al. 1995; Master et al. 1999).

Increased [Ca²⁺]_i is also known to activate PLA2 (Kramer and Sharp 1997). The mechanism by which PLA2 leads to the formation of RONS is likely through the increased production of arachidonic acid (AA). AA is known to produce various free radicals, including superoxide and hydrogen peroxide, presumably though AA’s inhibition of Complex I of the electron transport chain (Cocco et al. 1999). In the present study, we found that astrocyte cultures exposed to ammonia increased PLA2 activity and that BAPTA significantly blocked this activity. Additionally, astrocytes exposed to quinacrine, an inhibitor of PLA2, significantly diminished the ammonia-induced RONS generation.

The present study demonstrates that an early rise in [Ca²⁺]_i appears to be responsible for the activation of cNOS, NOX and PLA2 following ammonia treatment, as this activation was blocked by BAPTA. Whether such effects of Ca²⁺ are direct or indirect, however, are not known. It is possible that a transient rise in [Ca²⁺]_i could activate intracellular signaling pathways, which in turn, may stimulate the activities of these Ca²⁺-dependent enzymes. For example, cPLA2 has a binding domain for Ca²⁺ and its catalytic activity is significantly increased when this enzyme is phosphorylated by MAPKs (Hirabayashi and Shimizu 2000).

While our current data strongly supports the view that activation of Ca²⁺-dependent enzymes leading to ONS represents one mechanism for the ammonia-induced astrocyte swelling, the means by which Ca²⁺ and ONS results in such swelling is not completely understood. Excess Ca²⁺ is known to inhibit the mitochondrial electron transport chain leading to increased free radical production and associated ONS (Chinopoulos and Adam-Vizi 2006). It is also well known that ONS and arachidonic acid are important factors for the induction of the mPT (Scorrano et al. 2001), a phenomenon characterized by a sudden increase in the permeability of the inner mitochondrial membrane due to the opening of the permeability transition pore. This results in the collapse of inner mitochondrial membrane potential, leading to defective oxidative phosphorylation and free radical production (for reviews on the mPT, see (Zoratti and Szabo 1995). Induction of the mPT can lead to mitochondrial dysfunction and bioenergetic failure (Zoratti and Szabo 1995). Cell volume regulation is an energy-dependent process due to the operation of various ionic transport systems and extrusion of the osmotically active amino acids, all of which require energy (Kimelberg et al. 1993, Olson et al. 1992). Such mitochondrial dysfunction and associated disturbance in energy metabolism by ammonia, conceivably resulting from Ca²⁺, ONS and the mPT, may perturb astrocyte cell volume regulation and lead to cell swelling.

Consistent with the involvement of ionic transport systems and RONS in ammonia-induced astrocyte swelling, we have recently found that astrocyte cultures exposed to ammonia increased the activation of the Na-K-Cl cotransporter-1 (NKCC1) (Jayakumar et al. 2008), an ion transporter involved in cell volume regulation, Additionally, we found that oxidation and nitration of NKCC1 increased its activity. One may speculate that membrane proteins that are critically involved in the regulation of cell volume may become modified by oxidation/nitration, and that such protein modification may lead to their dysfunction and to cell swelling.

In summary, our study shows that an early increase in [Ca²⁺]_i is an important factor in the mechanism of RONS production and cell swelling in astrocytes after exposure to ammonia. The increase in RONS, in part, appears to be a consequence of an ammonia-induced activation of certain Ca²⁺-dependent enzymes (cNOS, NOX and cPLA2). Inhibition of these enzymes, as well as treatment with BAPTA diminished RONS formation and cell swelling in cultured astrocytes. These studies highlight a critical role of Ca²⁺ in the generation of RONS, and in the development of the astrocyte swelling associated with ammonia neurotoxicity.

Supplementary Material

Supp Fig 1. Figure S1.

Effect of Ca²⁺-dependent enzyme blockers on ammonia-induced RONS production. All inhibitors were added 15 min before exposure to ammonia (5 mM NH₄Cl) and RONS production was determined at 3 min and 2h after ammonia exposure. Values are expressed as arbitrary fluorescence intensity units (AFIU) of DCFDA per mg cell protein. *p<0.01 vs. control; **p<0.01 vs. NH₄+ (NH₄Cl). C, control; 7-NI, 7-nitroindazole, inhibitor of cNOS; APO, apocyanin, inhibitor of NADPH oxidase; QNC, quinacrine, inhibitor of PLA2.

Supp Fig 2. Figure S2.

Effect of BAPTA on ammonia-induced astrocyte cell volume. Cultures were pretreated with BAPTA/AM (25 μM) prior to exposure to ammonia (5 mM NH₄Cl) and cell volume was determined 24 h after ammonia treatment. *p<0.01 vs. control; **p<0.01 vs. NH₄+.

Supp Fig 3. Figure S3.

Effect of cNOS, NOX and PLA2 inhibitors on ammonia-induced astrocyte cell volume. All inhibitors were added 15 min before exposure to ammonia (5 mM NH₄ Cl) and cell volume was determined 24 h later. *p<0.01 vs. control; **p<0.01 vs. NH₄ (5 mM NH₄Cl). C, control; 7-NI, 7-nitroindazole, inhibitor of cNOS; APO, apocyanin, inhibitor of NADPH oxidase; QNC (5 μM), quinacrine, inhibitor of PLA2.

Abbreviations

ALF

acute liver failure

AA

arachidonic acid

APO

apocyanin

BAPTA-AM

1,2-bis-(o-aminophenoxy)-ethane-N,N,-N′,N′-tetraacetic acid tetraacetoxy-methyl ester

BSA

bovine serum albumin

cNOS

constitutive nitric oxide synthase

cPLA2

cytosolic phopholipase A2

DMEM

Dulbecco’s modified Eagle’s medium

DTNB

5,5′-dinitrobis-(2-dinitrobenzoic acid

DTT

dithiothreitol

EDTA

ethylenediaminetetraacetic acid

EGTA

ethylene glycol tetraacetic acid

FAD

flavin adenine dinucleotide

HE

hepatic encephalopathy

HEPES

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

iNOS

inducible nitric oxide synthase

MAPKs

mitogen-activated protein kinases

mPT

mitochondrial permeability transition

NOX

NADPH oxidase

7-NI

7-nitroindazole

NO

nitric oxide

NOS

nitric oxide synthase

OMG

O-methyl-[³H]glucose

ONS

oxidative/nitrosative stress

QNC

quinacrine

RONS

reactive oxygen/nitrogen species