Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

In the nervous system, many proteins, ion channels, receptors, and transporters are regulated by phosphorylation and dephosphorylation (Mumby and Walter, 1993; Graves and Krebs, 1999). The proper balance between phosphorylation/dephosphorylation is needed for normal neuronal functioning. When this balance is shifted, damage can result. Studies have shown that overactivation of protein kinases can trigger neuronal cell death (Mattson et al., 1988; Favaron et al., 1990; Manev et al., 1990; Mattson, 1991). Prevention of dephosphorylation with phosphatase inhibitors is also detrimental. The best-studied phosphatase inhibitor is okadaic acid (OKA), a C38 fatty acid produced by marine dinoflagellates. Isolated from the marine sponge Hallichondria okadaii, this compound inhibits phosphatase 2A and 1, but has little effect on other phosphatases (Bialojan and Takai, 1988). In vitro, OKA induces death in cortical (Mattson, 1991; Arias et al., 1993), hippocampal (Arias et al., 1998; Runden et al., 1998), cerebellar granule (Favaron et al., 1990; Manev et al., 1990; Fernandez et al., 1993), and neuroblastoma cells (Boe et al., 1991). In vivo, microinjection of OKA into the hippocampus or basal nucleus of rats is neurotoxic (Arias et al., 1998). Nothing is known as to the effect of disruption of phosphorylation pathways in dopamine neurons, the major neurons lost in the neurodegenerative disease Parkinson's disease (PD).

Evidence indicates that phosphorylation pathways are a component of several neurodegenerative diseases, including Alzheimer's, amyotrophic lateral sclerosis, and PD (Julien and Mushynski, 1998). Autopsied brain from PD patients display Lewy bodies in neurons predominately from the substantia nigra, as well as in other areas. Neurofilament proteins are the major filament component of Lewy bodies. Neurofilaments found in Lewy bodies are abnormally phosphorylated and partially degraded (Forno et al., 1986; Bancher et al., 1989). Hyperphosphorylated tau, an integral component of neurofibrillary tangles was recently demonstrated to be present more often than previously thought in the brains of PD patients and patients with dementia with Lewy bodies (Arima et al., 1999). Mitochondrial defects have been identified in the sporadic PD population (for reviews, see Bowling and Beal, 1995; Cassarino and Bennett, 1999). In a recent study by Saporito et al. (1999), an analog of the general protein kinase inhibitor K252a was found to protect mouse nigrostriatal dopaminergic neurons from cell loss due to MPP⁺, an inhibitor of mitochondrial complex I. (Nicklas et al., 1985). Thus, phosphorylation events appear to be involved in cell damage due to inhibition of energy metabolism and may be a factor in dopamine cell loss in PD.

It was, therefore, of interest to study the effects of disruption of phosphorylation/dephosphorylation activity in dopaminergic neurons. In the current study, the effects of both protein kinase inhibition with staurosporine and phosphatase inhibition with OKA on cultured mesencephalic dopamine and GABA neurons were examined. In these studies, cultured mesencephalic dopamine neurons were found to be differentially vulnerable to phosphatase inhibition. Evidence is presented to show that inhibition of phosphatase 2A was responsible for the damage to dopamine neurons. In vivo studies of infusion of OKA into the left striatum of rats produced a differential loss of dopamine terminals consistent with the increased vulnerability of these neurons in vitro.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mesencephalic Culture and Toxin Treatment. The mesencephalon from embryonic day 15 Sprague-Dawley rat fetuses was dissected and plated as previously described (Zeevalk et al., 1995). Dissociated cells (2.5 × 10⁵ cells/cm²) were plated into 24-well plates previously coated with polyornithine and fetal bovine serum and incubated at 37°C, in a 5% CO₂ incubator. Culture medium consisted of 5.5 mM glucose containing Dulbecco's modified Eagle's medium supplemented with 2 mM glutamine, 10% fetal bovine serum, and 10% horse serum. Medium was replaced at 2 days. At 7 days in vitro, half of the medium was removed, supplemented with 5.5 mM glucose, and saved as conditioned medium (CM). CM was used for refeeding of the cultures following toxin treatment. Refeeding with other than CM can induce dopaminergic cell death. 5-Fluoro-2-deoxyuridine was added to the remaining medium on day 7 in vitro to reduce glial growth. Mesencephalic cultures grown in this manner are a mixed culture containing dopamine, GABA, and cholinergic neurons and glia. Dopamine neurons have been shown to represent approximately 3% of total cells, whereas the major neuronal population is GABAergic. OKA-sodium salt, calyculin-A, OKA-methyl ester, and staurosporine (Calbiochem-Novabiochem, La Jolla, CA) were added on day 8 in vitro at the concentrations indicated in the figures. Exposure times were 24 h. K252a (Calbiochem-Novobiochem) was added 1 h before OKA and was present during OKA exposure. On day 9 in vitro, the toxin-containing medium was replaced with CM and the cultures allowed to recover for 3 days prior to assaying toxicity with measurement of high-affinity uptake or immunostaining and counting of tyrosine hydroxylase (TH)-positive neurons or trypan blue-positive cells.

Dopamine and GABA High-Affinity Uptake. The high-affinity transport of dopamine and GABA in mesencephalic cultures was measured simultaneously as previously described (Zeevalk et al., 1995) with minor modifications. In brief, cultures were incubated with 20 nM [³H]dopamine plus 5 µM [¹⁴C]GABA in the presence of 1 mM ascorbate, 100 µM pargyline, 10 µM aminoxyacetic acid, and 1 mM -alanine for 15 min at 37°C. After washing, radioactivity in the cells was extracted with 95% ethanol and quantitated by scintillation counting. Uptake in samples incubated at 4°C was measured and subtracted from other determinations to obtain energy-dependent uptake.

TH Immunocytochemistry and Cell Counting. Cultures were washed, fixed, and immunostained as reported previously (Zeevalk et al., 1998). Anti-TH mouse monoclonal antibody (IncStar Corp., Stillwater, MN) was used at a 1:1500 dilution. The Elite Avidin Biotin Peroxidase Vectastain kit (Vector Laboratories, Burlingame, CA) was used for immunostaining. Diaminobenzidine was the chromogen substrate used for visualization. Cells were viewed under 100× magnification. For counting purposes, the bottom of each well was marked at the center and in each of four quadrants. Each mark was located and the cells to the right and left of the marked area were counted in a 1-mm² reticle. Each condition was run in duplicate in three separate experiments. Counts from 60, 1-mm² fields per condition were averaged, corrected for magnification, and reported as mean counts per square millimeter ± S.E.M.

Trypan Blue Staining. Cultures were rinsed with a balanced salt solution and stained with 0.4% trypan blue for 30 min. Cultures were rinsed twice and the number of trypan blue-positive neurons was counted as described for counts of TH-positive cells.

TH Activity. One hour following okadaic infusion, whole striata were homogenized in 10 mM sodium phosphate buffer, pH 7.0, at a concentration of 33.3 mg/ml, centrifuged, and the supernatant stored at 80°C until analyzed. An equal volume (50 µl) of sample and TH cocktail containing 25 µl of 0.1 M NSD 1015 (3-hydroxybenzyl-hydrazine) in potassium phosphate buffer, pH 6.0, 1 µl of catalase (43 mg/ml), 2 µl of dithiothreitol (0.5 M), 1 µl of L-tyrosine brought to 50 µl with potassium phosphate buffer were mixed and preincubated for 2 min at 37°C. The reaction was started with 50 µl of 1 mM tetrahydrobiopterin in 0.01 M HCl and allowed to continue for 30 min at 37°C. The reaction was stopped with 50 µl of 0.8 M perchloric acid, centrifuged 16,000g for 3 min, and L-dopa in the supernatants and in known standards was extracted with 50 mg of alumina in 500 µl of 3 M Tris plus 0.134 M EDTA, pH, 8.6, with shaking for 10 min. Following centrifugation (16,000g, 30 s), the alumina pellet was washed twice with water. The L-dopa was eluted from the alumina with 0.2 M perchloric acid and measured by high-performance liquid chromatography with electrochemical detection. The L-dopa in the samples was quantified by comparison to identically treated standards.

DNA Fragmentation. Analysis of DNA fragmentation was carried out immediately following a 24-h exposure to OKA. DNA extraction was done using the Amersham Genomic Prep DNA Isolation kit (27-5237-01) following procedures supplied with the kit. DNA purity was determined by 260/280 ratios. Samples containing 5 µg of DNA in ethidium bromide were loaded onto a 2% agarose gel and run at 40 V to separate the different sizes of DNA. A 100-base pair ladder was used as a standard.

Intrastriatal Infusion. Male Sprague-Dawley rats (300-350 g, approximately 12 weeks; Harlan Farms, Indianapolis, IN) were used in accordance with the National Institute of Health's Guide for the Care and Use of Laboratory Animals. Studies were approved by the local Institutional Animal Care Committee. Rats were grouped two per cage in a room maintained at 20-22°C on a 12-h light/dark cycle with food and water available ad libitum. Procedures were essentially as described previously (Zeevalk et al., 1997). Animals were anesthetized with Brevital (16.7 mg/kg) and the left striatum cannulated such that the tip of the cannula resided just above the region to be infused. For striatal infusions, coordinates for needle placement corresponded to those in the rat atlas of Konig and Klippel (1967): AP, 7.2; L, 2.6; and DV, 0.4. After a recovery of 3 days, OKA-sodium salt or saline at pH 7.6 was infused into the striatum of awake rats using an adapted microsyringe/needle system. Infusion volume was 2 µl, with a delivery rate of 0.5 µl/min. The needle was left in place for an additional 1 min prior to its removal. All animals for toxicity studies were sacrificed at 1 week. Two minutes prior to sacrifice, rats were given a tail vein injection of 3-mercaptopropionic acid to inhibit glutamic acid decarboxylase activity and prevent GABA synthesis (Korf and Venema, 1983). Striata were removed, laid flat, and the nucleus accumbens ventral to the anterior commissure was dissected away. The remaining striatum was subdivided into three regions of equal mass that were designated ventral, middle, and dorsal striatum. Each region was homogenized in 0.2 N perchloric acid and analyzed for dopamine and GABA. For acute studies of OKA, animals were cannulated and infused as described above. One hour after infusion, the animals were sacrificed and the striata rapidly removed. Whole striata were sonicated in 10 mM sodium phosphate buffer, pH 7.0, for TH activity. An aliquot of the homogenate was removed and acidified with perchloric acid (final concentration 0.2 N) for measurement of dopamine, dopamine metabolites, and GABA levels.

Dopamine and GABA Determination. Dopamine and its metabolites 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) were measured from 0.2 N perchloric acid extracts with electrochemical detection as routinely done in the laboratory and described elsewhere (Zeevalk et al., 1997). GABA in the same extract was quantified by reverse phase high-performance liquid chromatography of the -phthalaldehyde adduct using fluorescent detection as described elsewhere (Zeevalk et al., 1997). Comparison to known standards was used for quantitation.

Statistics. For most comparisons, results were analyzed by ANOVA using Tukey's multiple comparisons post test for significance. Student's t tests for comparing the infused side versus noninfused side from the same animal or saline versus OKA-infused striata were performed for some experiments as indicated in the legends to the figures. A P value of less than 0.05 was considered statistically significant. Curve fitting and EC₅₀ calculations were determined using GraphPad Prism software.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

OKA Differentially Effects Mesencephalic Dopamine Neurons in Vitro. Inhibition of phosphatase activity caused a dose-dependent loss of high-affinity dopamine and GABA uptake when measured 3 days following a 24-h exposure to various concentrations of OKA (Fig. 1a). Dopamine neurons were uniquely sensitive to phosphatase inhibition. The EC₅₀ for toxicity to dopamine neurons was 4-fold less than that for the mesencephalic GABA neurons: EC₅₀ = 1.5 versus 6.5 nM for OKA toxicity to dopamine and GABA neurons, respectively. Incubation for 24 h with 5 nM OKA methyl ester, an analog of OKA that lacks phosphatase inhibitory activity, was not toxic to the cultures (Fig. 1b), demonstrating that phosphatase inhibitory activity was responsible for the toxic response. Acute exposure to 5 nM OKA for 1 h immediately prior to uptake determination did not alter dopamine or GABA uptake (uptake for dopamine and GABA was 76,888 ± 6,856 and 21,593 ± 2,133 dpm/culture ± S.D. in controls versus 71,206 ± 5,320 and 21,308 ± 1,460 in OKA-treated cultures, n from two experiments run in duplicate), indicating that this concentration of OKA did not produce a pharmacological effect on uptake. In contrast to the sensitivity of mesencephalic neurons to phosphatase inhibition, inhibition of general protein kinase activity with staurosporine for 24 h followed by 3 days of recovery was not toxic at concentrations that inhibit a variety of kinase activities (IC₅₀ values of 0.7-20 nM) (Fig. 2). At 200 nM staurosporine, a modest loss of both dopamine and GABA high-affinity uptake was observed, 24 ± 1 and 17 ± 3.5% of control ± S.E.M. for dopamine and GABA, respectively. However, a higher concentration of staurosporine (500 nm) did not produce additional toxicity. Consistent with the lack of toxicity caused by staurosporine, another general protein kinase inhibitor, K252a (500 nM, 24-h exposure) was also found to be nontoxic to mesencephalic neurons (Fig. 6).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   a, dose response for OKA in mesencephalic cultures. Rat mesencephalic cultures were treated with various concentration of OKA, as indicated, for 24 h on day 8 in vitro. Cultures recovered for 3 days prior to assessment of toxicity with a functional assay to simultaneously measure [3H]dopamine and [14C]GABA high-affinity uptake as described under Materials and Methods. Uptakes for dopamine and GABA in control cultures were 80,490 ± 2,372 and 20,370 ± 1,104 dpm ± S.E.M. per well, respectively. Determinations are from four to five separate experiments run in duplicate. *, different from control; , different from GABA response at the corresponding OKA concentration. b, effect of okadaic acid-methyl-ester (OKA-ME). Mesencephalic cultures were treated with either 5 nM OKA or the methyl ester analog OKA-ME that lacks phosphatase inhibitory activity, for 24 h. Uptake was determined after 3 days of recovery. n is from four experiments run in duplicate. *, different from control; , different from OKA-methyl ester. ANOVA, Tukey's post test.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Staurosporine dose response in mesencephalic cultures. Cultures were treated with the indicated concentrations of staurosporine on day 8 in vitro for 24 h and toxicity determined as described in the legend to Fig. 1. Results are from five experiments run in duplicate. *, different from control. ANOVA, Tukey's post test.

View larger version (73K): [in this window] [in a new window]   Fig. 3.   Effect of OKA on tyrosine hydroxylase-positive cells in vitro. Light micrographs from control (a) and 5 nM OKA-treated cultures (b). Following a 24-h exposure on day 8 in vitro, cultures were allowed to recover for 3 days, fixed, and immunostained for tyrosine hydroxylase.

View larger version (83K): [in this window] [in a new window]   Fig. 4.   OKA induces DNA fragmentation. Agarose gel of DNA from mesencephalic cultures treated with 5 nM OKA for 24 h. DNA was prepared immediately following OKA exposure and loaded at 5 µg/well. Lane 1, 100-base pair DNA ladder; lane 2, control; lane 3, OKA.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Calyculin-A dose response in vitro. Mesencephalic cultures were exposed to the indicated concentrations of calyculin-A for 24 h on day 8 in vitro. High-affinity uptake was determined following 3 days of recovery. Determinations are from three separate experiments run in duplicate. *, different from control; , different from GABA response at the same concentration of calyculin-A. ANOVA, Tukey's post test.

View larger version (59K): [in this window] [in a new window]   Fig. 6.   Effect of protein kinase inhibition on OKA toxicity in vitro. On day 8 in vitro, mesencephalic cultures were treated with either 5 nM OKA, 500 nM K252a, or OKA plus K252a. The protein kinase inhibitor K252a was added 1 h before OKA and was present throughout the 24-h exposure to OKA. Following toxin treatment, cultures were returned to glucose supplemented CM (see Materials and Methods) and allowed to recover for 3 days. High-affinity dopamine transport was measured on day 12 in vitro. GABA transport (data not shown) was unaffected by 5 nM OKA treatment. n is from three experiments run in duplicate. *, different from control; , different from OKA plus K252a. ANOVA, Tukey's post test.

In Vivo Effects of Intrastriatal OKA. The response of the dopamine population to OKA was also examined in vivo. To examine long-term toxic effects on striatal dopamine, rats received an infusion of OKA (32-320 pmol in 2 µl) into the left striatum. The right striatum served as each animal's control. One week following infusion, each striatum was removed, dissected into three regions of equal mass (ventral, middle, and dorsal), and dopamine and GABA levels in each region were measured. The site of infusion was determined to be at the dorsal to middle interface. As shown previously (Zeevalk et al., 1997), subdividing the striatum and separate analysis of the different regions can help to distinguish nonspecific damage at the site of infusion from specific damage due to the infused substance. OKA caused a dose-dependent loss of striatal dopamine (Fig. 7a). No loss of dopamine was observed at 32 pmol, whereas 160 and 320 pmol of OKA produced significant loss that was found in the middle and dorsal regions with 160 pmol and in all regions with 320 pmol of OKA. Consistent with in vitro data, OKA caused no significant loss of striatal GABA (Fig. 7b).

View larger version (38K): [in this window] [in a new window]   Fig. 7.   Long-term in vivo effect of intrastriatal OKA. Male Sprague-Dawley rats were cannulated and infused with OKA (pmol in 2 µl) as described under Materials and Methods. One week following infusion, left and right striata were removed and subdivided into three regions of equal mass, i.e., ventral, middle, and dorsal thirds (V, M, D), for measurement of tissue levels of dopamine and GABA. Dopamine and GABA levels in the left-treated side for each animal were compared with the corresponding noninfused right side and expressed as a ratio of left to right (L/R). Mean levels of dopamine in saline-treated animals were as follows: D, left versus right, 10.63 ± 0.21 versus 11.17 ± 0.72; M, left versus right, 10.0 ± 0.88 versus 10.63 ± 0.59; and V, left versus right, 5.61 ± 0.80 versus 6.29 ± 0.57 µg/g of tissue ± S.E.M. Striatal GABA levels in saline-treated animals were as follows: D, left versus right, 2.68 ± 0.28 versus 2.37 ± 0.28; M, left versus right, 2.56 ± 0.53 versus 2.21 ± 0.31; and V, left versus right 2.13 ± 0.26 versus 1.96 ± 0.13 µmol/g of tissue ± S.E.M. n is from four to nine animals per group. *, different from saline; , different from 32 pmol of OKA. ANOVA, Tukey's post test. #, different left versus right side, paired Student's t test.

View larger version (34K): [in this window] [in a new window]   Fig. 8.   Acute effect of OKA on dopamine turnover. Rats were cannulated and infused with 320 pmol of OKA in 2 µl into the left striatum as described under Materials and Methods. One hour following infusion, the animals were euthanized, and the striata removed and homogenized in 0.2 N perchloric acid for measurement of dopamine and metabolites DOPAC and HVA. Dopamine values in controls are given under Results and were unchanged with OKA treatment. DOPAC/dopamine ratios and HVA were significantly elevated, indicating increased dopamine turnover. n is from three animals per group. *, different from left-infused saline; , different from right-infused OKA side. ANOVA, Tukey's post test.

View larger version (17K): [in this window] [in a new window]   Fig. 9.   Acute effect of OKA on tyrosine hydroxylase activity. One hour following infusion of 320 pmol of OKA in 2 µl into the left striatum of rats, striata were removed and analyzed for tyrosine hydroxylase activity as described under Materials and Methods. Seven animals in each group were analyzed. *, different for saline versus OKA-infused left striata. Unpaired Student's t test.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

OKA, responsible for diarrhetic shellfish poisoning, induces neuronal death (Boe et al., 1991; Fernandez-Sanchez et al., 1996) with apoptotic features (Mattson et al., 1988; Favaron et al., 1990; Manev et al., 1990; Mattson, 1991). Our report is the first to show that mesencephalic neurons are also killed by phosphatase inhibition. In contrast, protein kinase inhibition was not toxic. This indicates that, for short exposures, prevention of protein phosphorylation was not detrimental, whereas prevention of dephosphorylation leads to cell death. Damage was associated with inhibition of PP2A. Nearly complete destruction of dopamine and GABA neurons occurred at 7.5 nM OKA. OKA inhibits PP2A and PP1 with IC₅₀ values of 0.1 to 1 and 10 to 15 nM, respectively (Bialojan and Takai, 1988; Ishihara et al., 1989; Cohen et al., 1990). Thus, complete cell loss occurred at a concentration below the IC₅₀ for inhibition of PP1. Calyculin-A, another phosphatase inhibitor equipotent toward PP2A and PP1 (IC₅₀ of approximately 2 nM) (Cohen, 1989), also showed differential toxicity toward the dopamine population. Tautomycin, a phosphatase inhibitor with greater potency toward PP1 (IC₅₀ of 1 nM) than PP2A (IC₅₀ of 10 nM) did not cause loss of dopamine or GABA neurons when used at a concentration that was 20 times its IC₅₀ for PP1 and 2 times its IC₅₀ for PP2A. This argues that toxicity was via inhibition of PP2A. Additionally, the effects were mediated via phosphatase inhibition since okadaic-methyl-ester, an OKA analog lacking phosphatase inhibitory activity, was not toxic.

While several laboratories showed induction of neuronal death by OKA, this is the first study to separate out the effects of PP2A inhibition in a neuronal population. OKA is not an exclusive neurotoxin and other cell types are killed by OKA, including kidney epithelial (Davis et al., 1996), hepatocytes (Holen et al., 1992), human myeloid leukemia (Benito et al., 1997), breast cancer (Rossini et al., 1997), and lens epithelial cells (Li et al., 1998). In contrast to the sensitivity of mesencephalic neurons to PP2A inhibition, apoptosis induced in lens epithelial cells was associated with inhibition of PP1 and not PP2A (Li et al., 1998). Lens epithelial cells and mesencephalic neurons, although both sensitive to OKA, are clearly different in the phosphatase involved in mediating damage. It would be of interest to know how this distinction holds for cells of different origins as well as for different neuronal types.

The damage to dopamine neurons by OKA was demonstrated in vitro by loss of high-affinity uptake. Tyrosine hydroxylase and the dopamine transporter (DAT) are subject to phosphorylation control. Phosphorylation increases TH activity (Kumer and Vrana, 1996) but decreases DAT activity (Vaughan et al., 1997). The loss of uptake that we observed likely reflects toxicity rather than an acute effect of OKA on transporter activity for several reasons. Extremely high concentrations of OKA are needed to lower DAT activity. OKA at 500 nM decreased dopamine uptake into striatal synaptosomes by 13% (Copeland et al., 1996). Increased phosphorylation of DAT by OKA in rat striatal synaptosomes was observed in the range of 1 to 10 µM (Vaughan et al., 1997). In another study, 1 µM OKA did not alter DAT (Tian et al., 1994). This would suggest that phosphatases other than PP2A or PP1 were involved in DAT dephosphorylation. The concentrations used in the present study were an order of magnitude below those needed to affect DAT activity. In addition, the effect on DAT is rapid, occurring within 10 to 30 min of exposure (Vaughan et al., 1997). In the present studies, a 60-min preincubation with 5 nM OKA had no immediate effect on DAT activity. In toxicity studies, uptake was measured 3 days following OKA to allow for recovery. GABA transport was not affected even though the GABA transporter is also regulated by phosphorylation and OKA acutely decreases GABA uptake (Tian et al., 1994). Similar to DAT, the decrease in GABA transport requires micromolar concentrations of OKA (Tian et al., 1994). Furthermore, OKA also caused DNA fragmentation, a decrease in TH-positive cells, and an increase in trypan blue-stained neurons. In vitro findings were also supported by the selective in vivo loss of striatal dopamine with OKA. In total, the evidence indicates that the decrease in dopamine transport 3 days following OKA exposure was due to damage to the dopamine neurons. It should be noted that toxicity as measured by the loss of dopamine high-affinity uptake was greater than that observed by counts of TH-positive cells (70 versus 55% loss, respectively). Okadaic acid treatment was found to cause both a loss of TH-positive neurons as well as loss of processes on remaining neurons. This may have resulted in a decrease in the amount of dopamine transporter present per cell and, therefore, a greater loss of high-affinity uptake.

A hierarchy of sensitivity to OKA was observed with the dopamine neurons being more susceptible than the GABA cells. Differences in neuronal susceptibility to OKA are not unique for mesencephalic neurons. Selective neuronal degeneration was reported in rat hippocampal slice culture (Runden et al., 1998). Slices treated with OKA showed a dose- and time-dependent loss of viability with the following hierarchy of sensitivities: dentate granule cells > CA3 > CA1. Similar to our findings, K252a attenuated OKA-induced cell death. Microinjection of OKA into the hippocampus in adult rat also resulted in selective degeneration of hippocampal neurons (Arias et al., 1998). However, in contrast to hippocampal slice cultures, adult hippocampus showed a greater sensitivity of CA1 pyramidal neurons. The reason for the different findings in the hippocampal preparations is not known but may relate to the age of the neurons. In the current study, the response of midbrain dopamine neurons was similar in the immature mesencephalic cultures and in the dopamine terminals in the striatum of mature adult rats.

Mesencephalic neurons were insensitive to inhibition of protein kinase activity. Staurosporine at 200 nM had only modest effects while 500 nM did not cause further toxicity. K252a (500 nM) was also not toxic. The highest concentration of staurosporine used in this study was 25 to 700 times the IC₅₀ values reported for inhibition of the various kinases by staurosporine (IC₅₀ values of 0.7-20 nM). K252a was used at 20 to 300 times its reported IC₅₀ values (1.8-25 nM). Therefore, sufficient concentrations were used and the small amount of toxicity observed with staurosporine may have been the result of either the inhibition of an unidentified kinase that has a higher K_m for the inhibitors or to a nonspecific effect. Many studies however, have used staurosporine to induce death (Krohn et al., 1998; Kim et al., 1999; Ahlemeyer et al., 2000; Bursztajn et al., 2000; Deshmukh and Johnson, 2000). Generally, these studies use high concentrations of staurosporine (100 nM-1 µM) (Kim et al., 1999; Ahlemeyer et al., 2000; Bursztajn et al., 2000; Deshmukh and Johnson, 2000), although this is not always the case (Krohn et al., 1998). The effect of staurosporine on primary dopamine neurons has not been reported previously, although staurosporine at 1 µM was shown to be toxic to a dopaminergic cell line, MN9D (Kim et al., 1999). In contrast, K252a and staurosporine can provide neuroprotection in stroke/ischemia (Xu et al., 1997; for review, see Sweeney et al., 1995). An analog of K252a, CEP 1347, was recently found to protect dopamine neurons in vivo from MPP⁺ toxicity (Saporito et al., 1999, 2000). The protective effect of CEP 1347 was attributed to inhibition of c-Jun N-terminal kinase. This is of interest to dopamine loss in PD since mitochondrial defects have been identified in the sporadic PD population and MPP⁺ is an inhibitor of mitochondrial complex I (Nicklas et al., 1985). Our recent work found that K252a protected mesencephalic neurons from toxicity due to inhibition of succinate dehydrogenase with the reversible inhibitor malonate (Zeevalk and Nicklas, 2000). In mesencephalic cultures as well as rat striatum in vivo, malonate, like OKA, causes a relatively selective killing of dopamine neurons (Zeevalk et al., 1995, 1997). Low chronic exposure to the complex I inhibitor rotenone also selectively damages dopamine neurons (Betarbet et al., 2000). This raises the possibility that disturbances in phosphorylation may have greater effects on dopamine neurons and may contribute to their vulnerability during general metabolic inhibition.

Differential cell susceptibility to OKA in hippocampal slice cultures was suggested to be due to a different complement of protein kinases or phosphatases in cells (Runden et al., 1998). Whether this is true for dopamine neurons is not known. Another possibility is that hyperphosphorylation of certain proteins unique to a particular cell type may contribute to vulnerability. Tyrosine hydroxylase, the rate-limiting enzyme in dopamine synthesis is regulated by phosphorylation and may be a PP2A substrate (for review, see Kumer and Vrana, 1996). The current study found that OKA when infused into the striatum can acutely result in an increase in TH activity and increased dopamine turnover. Spina and Cohen (1989) reported that increased dopamine turnover results in an elevation in oxidized glutathione and an oxidative stress. Thus, the short-term effects of okadaic acid on TH activity and dopamine turnover (increased DOPAC with dopamine unchanged) could result in increased oxidative stress in the dopamine neurons and the selective damage to striatal dopamine terminals observed 1 week following exposure. This possibility is currently being investigated in the laboratory.

In summary, the findings show that inhibition of dephosphorylation of certain proteins rather than prevention of phosphorylation triggers damage to mesencephalic neurons. Dopamine neurons in vitro and striatal dopamine terminals in vivo show relatively greater vulnerability to disruption of phosphatase 2A. Disturbances in the normal phosphorylation/dephosphorylation events in dopamine neurons may contribute to the etiology of PD and targeting of phosphorylation pathways is being explored as an avenue for intervention. It is, therefore, important to develop an understanding of the effects of protein kinase and phosphatase inhibition on neurons in general and on dopamine neurons in particular.