Calcium-sensitive regulation of monoamine oxidase-A contributes to the production of peroxyradicals in hippocampal cultures: implications for Alzheimer disease-related pathology

Background Calcium (Ca2+) has recently been shown to selectively increase the activity of monoamine oxidase-A (MAO-A), a mitochondria-bound enzyme that generates peroxyradicals as a natural by-product of the deamination of neurotransmitters such as serotonin. It has also been suggested that increased intracellular free Ca2+ levels as well as MAO-A may be contributing to the oxidative stress associated with Alzheimer disease (AD). Results Incubation with Ca2+ selectively increases MAO-A enzymatic activity in protein extracts from mouse hippocampal HT-22 cell cultures. Treatment of HT-22 cultures with the Ca2+ ionophore A23187 also increases MAO-A activity, whereas overexpression of calbindin-D28K (CB-28K), a Ca2+-binding protein in brain that is greatly reduced in AD, decreases MAO-A activity. The effects of A23187 and CB-28K are both independent of any change in MAO-A protein or gene expression. The toxicity (via production of peroxyradicals and/or chromatin condensation) associated with either A23187 or the AD-related β-amyloid peptide, which also increases free intracellular Ca2+, is attenuated by MAO-A inhibition in HT-22 cells as well as in primary hippocampal cultures. Conclusion These data suggest that increases in intracellular Ca2+ availability could contribute to a MAO-A-mediated mechanism with a role in AD-related oxidative stress.


Background
MAO-A and MAO-B, two isoforms of monoamine oxidase (MAO), are expressed on the mitochondrial outer membrane. MAO-mediated neurodegeneration can result from the formation of hydrogen peroxide (H 2 O 2 ) as a by-product of metabolism of aminergic neurotransmitters including serotonin and dopamine. If it is not detoxified by antioxidant systems such as glutathione peroxidase -one of the most abundant such systems in brain [1] -then H 2 O 2 can be converted by iron-mediated Fenton reactions to hydroxyl radicals that can initiate lipid peroxidation and cell death. This is exacerbated when antioxidant systems are compromised, such as during aging [2]. The reduction in the efficacy of these systems may simply be aggravated in chronic disease states such as Alzheimer disease (AD) [3].
It has been demonstrated that inhibitors of MAO-B, such as l-deprenyl and, more recently, rasagiline, are effective in the management of early symptoms of Parkinson's disease in the clinic and in animal models [4][5][6] as well as in patients with mild AD-type dementia [4,7]. Both MAO-Bpositive astrocytes [8] and reactive oxygen species (ROS) [9] have been found in the vicinity of β-amyloid (Aβ) plaques. L-Deprenyl and rasagiline, however, may exert some of their effects independently of MAO-B as the neuroprotection mediated by these drugs is often associated with concentrations of the drug that are well below those required for inhibition of the enzyme [6], and has been associated with activation of Bcl-2 family members, interactions with the mitochondrial pore complex, and modulation of amyloid precursor protein cleavage [4].
MAO-A also plays a role in neuropsychiatric and behavioral disorders. The importance of this isoform is suggested by the aggressive phenotype seen in male mice deficient in MAO-A [10] and in males in a Dutch kindred bearing a spontaneous mutation (resulting in a premature stop) in the mao-A gene [11]. Similarly, maltreated children, whose genotype confers low levels of MAO-A expression, more often develop conduct disorder, antisocial personality and adult violent crime than do children with a highactivity MAO-A genotype [12]. Relatively modest changes in MAO-A activity/function can have important neuropsychiatric consequences as demonstrated by the fact that [ 11 C]-harmine-labeled MAO-A is elevated by only 34% throughout the brain of untreated depressed patients compared to controls, yet it appears to be the major contributor to monoamine metabolism in these same patients [13]. Depression not only may promote cognitive impairment, but also may be a risk factor for AD [14]. Not surprisingly, MAO-A, which is often targeted for the treatment of depression, is also a potential risk factor for lateonset AD [15][16][17][18]. In contrast to irreversible inhibitors of MAO-A such as clorgyline, reversible inhibitors such as moclobemide are better tolerated and have been particularly efficacious in treating depression [19,20] and cognitive disorders [21]  These combined observations, in addition to the fact that cytoplasmic free Ca 2+ is elevated in aged neurons and even more so during neurodegeneration, such as that encountered during in AD [15,[34][35][36], certainly argue for examination of the relation between Ca 2+ and MAO.

Immortalized hippocampal cell cultures
The immortalized mouse hippocampal HT-22 cell line [20] was kindly provided by Dr. P. Maher (The Scripps Research Institute, La Jolla, CA, USA). Cells were cultured (5% CO 2 at 37°C) in DMEM/low glucose medium containing 10% fetal bovine serum, 100 IU/mL penicillin G sodium salt and 0.03% glutamine. extraction of labeled metabolites proceeded as described above. Data were analyzed for estimates of K m and V max using the Prism v3.01 software.

Transient overexpression of CB-28K
The pREP-CB-28K (calbindin-D28K) plasmid expression vector was kindly provided by Dr. A. Pollock (University of California, San Francisco, CA). HT-22 cells were seeded in log phase and transfected with plasmid DNA (1-2 µg/ well on a 24-well plate; seeded at 5 × 10 5 cells/well) using ExGen™500 (Fermentas) according to the manufacturer's directions. Expression of eGFP fluorescent protein revealed a transfection efficiency of approximately 50% using this technique. Cells were routinely harvested 24 h post-transfection.

Immunodetection of target proteins
Treated HT-22 cells were washed twice with ice-cold PBS and proteins were extracted in ice-cold lysis buffer (1% Triton X-100, 10% glycerol, 1 mM EDTA, 20 mM Tris, pH 7.5), containing protease inhibitor cocktail and 1 mM orthovanadate. Standard denaturing (SDS-PAGE) conditions were used to resolve proteins, which were then transferred to nitrocellulose. Protein expression in total cell lysates (20-30 µg/lane, precleared; 5000 × g, 10 min, 4°C) was visualized by enhanced chemiluminescence. Depicted immunoblots are representative of two-three independent experiments.

Determination of target mRNA by semi-quantitative reverse transcriptase-PCR (RT-PCR)
Total RNA from treated HT-22 cells was prepared with TRIZOL reagent according to the manufacturer's protocol and digested with RNase-free DNase to clear residual genomic DNA. First strand cDNA was reverse-transcribed from 2 µg of total RNA using oligo-(dT) (SuperScript™III First-Strand Synthesis System, Invitrogen). The cDNA was amplified using Taq  Visualization of cytoplasmic peroxide radicals 2',7'-Dihydrodichlorofluorescein diacetate (DCFH2-DA) [37] permeates the cell membrane and is hydrolyzed to DCFH2, a nonfluorescent compound that remains trapped within the cell, but which yields a fluorescent product upon oxidation by H2O2. The cells were seeded on cover-slips. After treatment, the cells were rinsed twice with PBS and incubated with DCFH 2 -DA (5 µM, 30 min at 37°C) or DMSO, vehicle). The cells were washed in prewarmed HEPES-buffered (20 mM) HBSS (pH 7.0) containing 5 mM glucose, and prepared for DCF fluorescence (excitation: 488 nm; emission: 530 ± 15 nm). Collected tissues were digested at 37°C with 0.25% trypsin-EDTA 15 min. The reaction was quenched with fetal bovine serum (FBS, 10%) and tissues were rinsed 3-4 times with HBSS to remove FBS. Following centrifugation at 800 × g for 10 min, the medium was removed and cells were resuspended in a chemically defined serum-free NeuroBasal medium supplemented with 1% N 2 , 2% B27, 50 µM L-glutamine, 15 mM HEPES, 10 U/ml penicillin and 10 µg/ml streptomycin. Neurons were then plated on coverslips (coated with 25 µg/ml poly-D-lysine), and grown at 37°C with 5% CO 2 -humidified atmosphere. The medium was replaced 24 h later with fresh NeuroBasal medium lacking L-glutamine and antibiotics. Medium was replaced after 4-5 days in vitro (DIV). Neurons were treated on DIV 7, following which they were fixed with 4% paraformaldehyde in 0.01 M PBS for 20 min at room temperature, washed several times with PBS, and stained with Hoechst 33258 (500 ng/ml, 10 min) for microscopic visualization of chromatin condensation (a characteristic of apoptotic cell death).

Statistical analyses
Significance (set at P < 0.05) was assessed by unpaired ttests or by one-way ANOVA with post hoc analyses relying on Bonferonni's Multiple Comparison Test (GraphPad Prism v3.01). Data are represented as mean ± standard deviation (SD).

The Ca 2+ ionophore A23187 increases MAO-A activity
HT-22 cells were treated with the Ca 2+ ionophore A23187 (5 µM; 30 min) so as to examine the effect of increasing Ca 2+ availability on MAO-A activity in living cells. Fluorescence microscopy (Fig. 3A) (Fig. 3B). MAO-A protein expression was unchanged compared to control levels (Fig. 3C).

Reduction of free intracellular Ca 2+ by overexpression of CB-28K reduces MAO-A activity
HT-22 cells were transfected with either the vector control (pREP) or the pREP-CB-28K expression plasmid (Fig. 4).
The Ca 2+ ionophore A23187 increases MAO-A activity   [13] have recently demonstrated that untreated depressed patients have levels of MAO-A density that are only 34% higher than that found in controls. Yet these same authors propose that this modest change could account for most of the change in biogenic amine levels observed in these patients. Obviously, small changes in MAO-A can significantly impact brain function.

Discussion
While low millimolar levels of Ca 2+ are necessary to reveal this effect, the fact that MAO-B activity remains unaffected clearly supports differences in MAO-A and MAO-B regulation. The need for millimolar concentrations also suggests pathological relevance. Interestingly, the mitochondria can modulate cytoplasmic Ca 2+ homeostasis by accumulating Ca 2+ into the very high micromolar range [43]. Furthermore, elevated Ca 2+ concentrations localized to microdomains may also represent unique means of modulating localized mitochondrial membrane and/or matrix function [44], including the activation of dehydrogenases [45]. In neurons and chromaffin cells, mitochondria rapidly and reversibly buffer Ca 2+ during cell stimulation to help clear large Ca 2+ loads [46-48]. The ensuing overloading of mitochondria with Ca 2+ may be involved in several pathological conditions, including ischemia-reperfusion lesions, neurotoxicity and neurodegenerative diseases, where ATP depletion, overproduction of ROS and release of apoptotic factors lead to cell damage [49].
Using the Ca 2+ ionophore A23187, we now confirm that increasing free intracellular Ca 2+ well above normal levels    and Aβ-induced toxicity in these cells is also sensitive to MAO-A inhibition (present study). Experimentation based on in vivo assessment of the effect of Ca 2+ on MAO-A as well as a closer examination of the relation between Ca 2+ /CB-28K and MAO-A in AD tissues is warranted.

Overexpression of the Ca 2+ -binding protein CB-28K decreases MAO-A activity
MAO-A is a risk factor in AD and changes in MAO-A activity parallel changes in the production of ROS, e.g. H 2 O 2 .
Given the neuroprotective role of CB-28K in human pathologies such as AD (and in models of AD), in addition to our demonstration that the toxicity of the ADrelated peptide, Aβ, is sensitive to MAO-A inhibition, we suggest that part of the oxidative stress associated with AD may rely on a Ca 2+ /CB-28K-sensitive, MAO-A-mediated mechanism.

Conclusion
The availability of free intracellular Ca 2+ is positively correlated with the activity of MAO-A, a mitochondrial H 2 O 2 -generating enzyme. The influence of Ca 2+ on MAO-A function could potentially impact AD-related pathology, which is often associated with altered Ca 2+ homeostasis, mitochondrial dysfunction and oxidative stress.