Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Apoptotic mechanisms appear to contribute to excitotoxic neuronal injury in rat striatum (Portera-Cailliau et al., 1995; Qin et al., 1996). Although the role of the transcription factor nuclear factor-B (NF-B) in apoptosis is controversial, some studies indicate that it may participate in the excitotoxin-induced apoptotic process in postmitotic neurons in vivo (Clemens et al., 1997; Qin et al., 1998, 1999; Nakai et al., 1999a,b).

In cultured neurons and experimental animals, glutamate receptor stimulation activates neuronal NF-B by accelerating the degradation of its cytoplasmic binding protein IB- (Guerrini et al., 1995; Qin et al., 2000). Upon its nuclear translocation in striatal neurons, NF-B up-regulates the proapoptotic proteins c-Myc and p53 in response to excitotoxic insult; blockade of NF-B translocation with the recombinant peptide NF-B SN50 inhibits quinolinic acid- (QA) or kainic acid-induced apoptosis (Qin et al., 1999; Nakai et al., 1999a). NF-B activation also occurs in such human neurodegenerative disorders as Alzheimer's disease (AD) and Parkinson's disease as well as in animal models of ischemia (Clemens et al., 1997; Hunot et al., 1997; Kaltschmidt et al., 1997; Gabriel et al., 1999). NF-B activation could thus serve a crucial postsynaptic contributor to excitotoxin-induced neuronal destruction.

Recent pharmacological observations have begun to suggest that inhibition of the pathological activation of NF-B may confer protective benefit to mature central nervous system neurons. Early epidemiological studies indicated that nonsteroid anti-inflammatory drugs, such as aspirin and sodium salicylate, may have protective effects in AD (Breitner, 1996). More recently, aspirin and sodium salicylate were found to diminish glutamate neurotoxicity through NF-B inhibition (Grilli et al., 1996). Estrogen also has been reported to diminish neuronal loss in AD (Simpkins et al., 1997) and to protect neurons against excitotoxic neuronal injury (Goodman et al., 1996). Although its mechanisms of action remain to be determined, NF-B inhibition may be involved (Galien and Garcia, 1997). Immunodepressants such as cyclosporin A protect neurons against excitotoxin-, ischemia-, and oxidative stress-induced neuronal damage (Li et al., 1997; Matsuura et al., 1997). Interestingly, cyclosporin A also inhibits NF-B activation (Meyer et al., 1997). In addition, the neuroprotective activity of antioxidants and metabotropic glutamate receptor agonists against striatal injury is also associated with an inhibitory effect on NF-B activation (Nakai et al., 1999b; Wang et al., 1999).

Prostaglandin A₁ (PGA₁) is known to induce heat shock protein (HSP) synthesis and block the cell cycle and viral replication (Lacal et al., 1994; D'Onofrio et al., 1995). Recent in vitro studies have found that PGA₁ inhibits degradation of the NF-B inhibitory protein IB- (Rossi et al., 1997, 2000). The induction of HSPs is a highly conserved cellular defense mechanism against adverse environmental conditions. More specifically, HSP induction has been associated with the protection of neurons against the injurious effects of heat shock and ischemia, probably by a mechanism involving the inhibition of apoptosis (Lowenstein et al., 1991; Rordorf et al., 1991; Samaili and Cotter, 1996; Yenari et al., 1998). Taken together, these observations suggest that PGA₁ could act to attenuate excitotoxic neuronal damage. To evaluate this possibility, we studied the effects of PGA₁ on QA-induced striatal neuronal injury, NF-B activation, and HSP induction in an animal model of Huntington's disease. The results showed that PGA₁ inhibited NF-B activation, induced HSP expression, and protected striatal neurons against apoptotic death.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male Sprague-Dawley rats weighing 300 to 350 g were purchased from Taconic Farms (Germantown, NY). Rats were housed two per cage in an animal room with a 12-h light/dark cycle and had free access to food and water. All procedures were performed in accordance with National Institutes of Health Guidelines for the Care and Use of Laboratory Animals.

Drug Administration. Intrastriatal drug administration was performed as previously described (Qin et al., 1996). Synthetic PGA₁ was purchased from Sigma (St. Louis, MO), dissolved in absolute ethanol (EtOH), and then diluted with saline (final ethanol concentration of 40%). To study the effects of PGA₁ pretreatment on QA-induced internucleosomal DNA fragmentation, three experiments were performed. In the first, rats were pretreated with intrastriatal injection of PGA₁ (5-80 nmol) or vehicle (1 µl of 40% EtOH) 10 min before QA (60 nmol) and killed 24 h later for extraction of genomic DNA. In the second experiment, rats were pretreated with PGA₁ (80 nmol) or vehicle 10 min before QA (60 nmol) as described above and killed 12, 24, or 48 h later. In the third experiment, one group of animals was pretreated with intrastriatal injection of PGA₁ (80 nmol) 10 min before QA (60 nmol); other animals were first given QA and then PGA₁ (80 nmol) was injected intrastriatally 2, 4, or 6 h later. All were killed 24 h after QA treatment. The animals were then killed and striatal tissues used for DNA extraction. Three rats were used in each group. To study the effect of PGA₁ on the number of striatal GABAergic neurons, rats were treated with intrastriatal injection of PGA₁ (5-80 nmol) or vehicle before QA (60 nmol) as described above and killed 10 days later. Brains were sectioned for receptor autoradiography and in situ hybridization histochemistry. To confirm the neuroprotective effects of PGA₁, rats were pretreated with the intrastriatal injection of PGA₁ (5-80 nmol) or vehicle 10 min before QA (60 nmol) and were killed 10 days later. Brains were sectioned and processed for Nissl staining and microscopic examination. Four rats were used in each group. To study the effect of PGA₁ on QA-induced NF-B and AP-1 activation, rats were pretreated with intrastriatally infused PGA₁ (5-80 nmol) or vehicle 10 min before the intrastriatal administration of QA (60 nmol). Animals were killed 12 h later and their striata taken for nuclear protein extraction. To study the effects of PGA₁ on QA-induced IB- degradation, or on levels of HSPs, rats were treated with intrastriatal infusion of PGA₁ (80 nmol) or vehicle 10 min before QA (60 nmol) and killed 12 h later. Striatal proteins were extracted for Western blot analysis. To study the cellular localization of induced 70-kDa heat shock protein (HSP70), rats were treated with vehicle (40% EtOH, 1 µl), QA (60 nmol) plus vehicle, or PGA₁ plus QA and were killed 12 h later. Brains were perfused via the ascending aorta with 40 mM phosphate-buffered saline (PBS, pH 7.4) containing 4% paraformaldehyde. Brains were sectioned for immunohistochemistry.

Electrophoresis Mobility Shift Assay. Striatal nuclear proteins were prepared as previously described (Qin et al., 1998). Briefly, striatal tissues were gently homogenized and nuclear proteins were obtained with high salt extraction. Protein concentrations were determined with a bicinchoninic acid kit (Pierce, Rockford, IL). Double-stranded DNA oligonucleotides containing consensus sequences for NF-B and AP-1 (Promega, Madison, WI) were labeled with [³²P]ATP by T4 polynucleotide kinase (Promega). Nuclear proteins (8-12 µg) were incubated with radiolabeled DNA probes (approximately 40,000 cpm) for 15 min at room temperature in the binding buffer (Promega). Nonspecific binding was assessed by adding 60-fold nonradioactive NF-B probes to the reaction mixture. The sample was then electrophoresed on 4.5% nondenaturing polyacrylamide gel with 0.5× Tris borate-EDTA buffer. Autoradiograms were developed by exposing the vacuum-dried gels to X-ray film at 80°C with intensifying screens for 24 to 48 h. Results were quantitatively analyzed with an image analyzer (NIH Image 1.60). The specific binding of NF-B or AP-1 was obtained by subtracting nonspecific binding (with cold competitors) from total binding (without cold competitors).

Western Blot Analysis. Western blotting was performed as previously described (Qin et al., 1999). Striatal tissues were homogenized and protein concentrations determined using a bicinchoninic acid protein assay kit (Pierce). Samples were mixed with loading buffer and boiled for 5 min. An aliquot of 30 µg of protein from each sample was separated on 12% SDS-polyacrylamide gel electrophoresis gel using constant current. Proteins were subsequently transferred to Immobilon-P membranes (Millipore, Bedford, MA) with a semidry blotting system. After blocking for 1 h in 0.1 M PBS (pH 7.5) with 0.1% Tween 20 (PBST) and 5% nonfat dry milk, membranes were incubated for 3 h with primary antibodies in PBST containing 3% nonfat dry milk. Membranes were then washed and incubated with a horseradish peroxidase-conjugated secondary antibody in PBST containing 3% nonfat dry milk for 1 h. Immunoreactivity was detected by enhanced chemiluminescent autoradiography (ECL kit; Amersham Life Science, Arlington Heights, IL) in accordance with the manufacturer's instructions. A mouse monoclonal antibody against 72-kDa HSP (RPN 1197) was purchased from Amersham Life Science (Arlington Heights, IL). Antibodies against HSP70 and 70-kDa heat shock cognate protein (HSC70) were mouse monoclonal antibodies W27 and B-6. The antibody against B- [IB- (FL)] was a goat polyclonal antibody. All were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Genomic DNA Preparation and Electrophoresis. Striatal genomic DNA was prepared as previously described (Qin et al., 1996). Briefly, striatal tissues were homogenized in a buffer containing 100 mM NaCl, 25 mM EDTA-Na₂, 10 mM Tris-HCl (pH 8.0), 0.5% SDS, and 0.5 mg/ml RNase. Homogenates were incubated at 55°C for 2 h; 0.6 mg of protease K was added and the incubation continued overnight. The homogenates were then extracted with phenol/chloroform/isoamyl alcohol (25:24:1) three times and DNA was precipitated with 1 volume of isopropanol and 1/10 volume of 5 M ammonium acetate and centrifuged at 14,000 rpm for 20 min. DNA pellets were washed once with precooled 80% alcohol, vacuum-dried, and resuspended in 50 mM Tris-EDTA (TE) buffer. DNA fragments were separated on 2% agarose gel (NuSieve 3:1) and detected with an UV transilluminator after staining with ethidium bromide.

Receptor Autoradiography and in Situ Hybridization Histochemistry. To determine D₁ dopamine (DA) receptors, brain sections were rinsed twice in precooled 50 mM Tris-HCl buffer (pH 7.4) containing 120 mM NaCl, 5 mM KCl, 2 mM CaCl₂, and 1 mM MgCl₂. Sections were then incubated in 50 mM Tris-HCl buffer with 2 nM [³H]SCH-23390 and 80 nM ketanserin for 1 h at room temperature. Nonspecific binding was determined by incubating adjacent sections in the presence of 2 µM SCH-23390 added to the above-described solution. Sections were exposed to X-ray film with tritium standards (Amersham Life Science) for 10 days. Autoradiograms were quantitatively analyzed with an image analyzer (NIH Image 1.60). Antisense oligodeoxynucleotide probes complimentary to 67-kDa glutamic acid decarboxylase (GAD₆₇) mRNA were labeled with [³³P]dATP using terminal deoxynucleotidyl transferase and purified by filtration chromatography (Chroma Spin-10; Clontech, Palo Alto, CA). To determine GAD₆₇ mRNA, brain sections were fixed in 0.1 M PBS (pH 7.4) containing 4% paraformaldehyde for 10 min and rinsed in PBS. Sections were then incubated in tetraethylammonium-acetate buffer containing 0.9% NaCl, 0.25% acetate anhydride, and 0.2 M triethanolamine for 10 min, rinsed in PBS, and dehydrated. Sections were next hybridized with radioactively labeled oligodeoxynucleotide probes in a buffer containing 50% formamide, 4× standard saline citrate, 0.1% Denhardt's solution, 10% dextran sulfate, 0.25 mg/ml yeast transfer RNA, 0.5 mg/ml salmon sperm DNA, 10 mM dithiothreitol, and 5 × 10⁶ cpm ³³P-labeled oligodeoxynucleotide probes for 18 h at 40°C with coverslips. After hybridization, sections were washed in standard saline citrate solution with increasing stringency. The final wash was carried out at 58°C for 40 min. Autoradiograms were developed by exposing the sections to X-ray films (Hyperfilm -max; Amersham Life Science) for 10 days and quantitatively analyzed with an image analyzer (NIH Image 1.60). Determination of D₁ DA receptor density made use of a standard curve constructed using coexposed ³H Microscales (Amersham Life Science). Optical density of the GAD₆₇ mRNA autoradiograms resulting from in situ hybridization histochemistry made use of a standard curve constructed from a Kodak photographic table No 3 (Kodak, Rochester, NY). Both D₁ DA receptors and GAD₆₇ mRNA were measured in three brain sections from each animal (bregma 1.7-0.2). These three individual measurements were pooled to generate a mean density of D₁ DA receptors or optical density of GAD₆₇ mRNA, and the data were converted to percentage of control (contralateral striatum) after ANOVA analysis for presentation as bar graphs.

Immunohistochemistry and Nissl Staining. Sections were washed in 0.1 M Tris-buffered saline (pH 7.4) and then incubated in Avidin/Biotin Blocking solution (Vector Laboratories, Burlingame, CA) for 15 min at room temperature. After washing in 10 mM PBS (pH 7.4), sections were incubated free floating in 10 mM PBS containing primary antibodies (HSP70, W27), 0.3% Triton X-100, and 1% normal serum at 4°C for 72 h. Sections were finally washed and incubated with secondary antibodies using a Vectastain Elite kit (Vector Laboratories) according to the manufacturer's protocol. Some sections were also counterstained with thionin. The specificity of the antibodies was characterized by the manufacturer and tested in our studies by omitting the primary or the secondary antibodies. For Nissl staining, animals were perfused with 4% paraformaldehyde in PBS (pH 7.4) and then postfixed in the same perfusate for an additional 6 h. Brains were then frozen and sectioned.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of PGA₁ on QA-Induced Internucleosomal DNA Fragmentation. The effects of PGA₁ on QA-induced DNA fragmentation were first analyzed individually in each animal with agarose gels. Equal amount of DNA was then taken from three animals and pooled to run the gels again (Figs. 1 and 2). These studies showed that the intrastriatal administration of QA (60 nmol) produced intense internucleosomal DNA fragmentation. Pretreatment with PGA₁ (5-80 nmol) inhibited the QA-induced DNA fragmentation in a dose-dependent manner. Vehicle treatment appeared to slightly reduce the intensity of DNA fragmentation. PGA₁ alone did not produce appreciable DNA fragmentation (Fig. 1). Inhibition of QA-induced DNA fragmentation by a single dose of PGA₁ (80 nmol) was observed from 12 to 48 h after QA treatment (Fig. 2A). To determine whether PGA₁ can rescue striatal neurons shortly after excitotoxin exposure, PGA₁ (80 nmol) was administered 2, 4, or 6 h after the intrastriatal infusion of QA. Under these conditions, PGA₁ diminished QA-induced DNA fragmentation when administered up to 4 h after QA, but to a lesser degree than when given as a pretreatment (Fig. 2B).

View larger version (85K): [in this window] [in a new window]   Fig. 1.   Dose-dependence of effects of PGA1 on QA-induced internucleosomal DNA fragmentation. Rats were treated intrastriatally with PGA1 (5-80 nmol) or vehicle (40% EtOH, 1 µl) 10 min before intrastriatal administration of QA (60 nmol) and killed 24 h later. Genomic DNA was extracted from each animal (three animals in each group) and was first analyzed individually with agarose gels. Then an equal amount of pooled DNA was loaded onto each lane of the agarose gel for the presentation in Fig. 1. Lane 1, 100-bp DNA marker; 2, QA (60 nmol); 3, QA + EtOH (40%, 1 µl); 4, QA + PGA1 (5 nmol); 5, QA + PGA1 (20 nmol); 6, QA + PGA1 (80 nmol); 7, PGA1 (5 nmol); 8, PGA1 (20 nmol); 9, PGA1 (80 nmol).

View larger version (40K): [in this window] [in a new window]   Fig. 2.   Time course of effects of PGA1 on QA-induced internucleosomal DNA fragmentation. A, rats were treated intrastriatally with PGA1 (80 nmol) or vehicle (40% EtOH, 1 µl) 10 min before intrastriatal administration of QA (60 nmol) and killed 12, 24, or 48 h later. Genomic DNA was extracted from each animal (three animals in each group) and was first analyzed individually with 2% agarose gels. Then an equal amount of pooled DNA was loaded onto each lane of the agarose gel for the presentation in Fig. 2A. Lane 1, 100-bp DNA marker; 2, QA + Veh 12 h; 3, QA + PGA1 12 h; 4, QA + Veh 24 h; 5, QA + PGA1 24 h; 6, QA + Veh 48 h; 7, QA + PGA1 48 h. B, rats were treated intrastriatally with PGA1 (80 nmol) or vehicle (40% EtOH, 1 µl) 10 min before or 2, 4, and 6 h after intrastriatal administration of QA (60 nmol) and killed 24 h later. Genomic DNA was extracted from each animal (three animals in each group) and was first analyzed individually with 2% agarose gels. Then an equal amount of pooled DNA was loaded onto each lane of the agarose gel for the presentation in Fig. 2B. Lane 1, 100-bp DNA marker; 2, QA + Veh; 3, pretreatment with PGA1 15 min before QA; 4, post-treatment with PGA1 2 h after QA; 5, post-treatment with PGA1 4 h after QA; 6, post-treatment with PGA1 6 h after QA.

Effect of PGA₁ on Striatal Cell Death. Receptor autoradiography for D₁ DA receptors and in situ hybridization histochemistry for GAD₆₇ mRNA revealed that PGA₁ markedly reduced the loss of striatal neurons induced by QA. Quantitative analysis of these results showed that QA reduced the density of D₁ DA receptors in the ipsilateral striatum to 39 ± 7.3% of control (contralateral side) (p < 0.05, n = 6). PGA₁ (5-80 nmol) decreased the QA-induced loss of D₁ DA receptors in a dose-dependent manner: D₁ DA receptors increased from 39 ± 7.3% of control in the QA-treated group to up to 82 ± 2.7% of control in the QA plus PGA₁-treated (80 nmol) group (p < 0.05, n = 6, Fig. 3). Similarly, QA diminished striatal GAD₆₇ mRNA levels to 37 ± 6.7% of control (contralateral side) (p < 0.05, n = 6). Pretreatment with PGA₁ (5-80 nmol) also reduced the QA-induced decrease in GAD₆₇ mRNA levels: GAD₆₇ mRNA increased from 37 ± 6.7% of control in the QA-treated group to up to 91 ± 2.8% of control in the QA plus PGA₁-treated (80 nmol) group (p < 0.05, n = 6, Fig. 4). Pretreatment with vehicle also tended to reduce the QA-induced decrease in D₁ DA receptors (increased from 39 ± 7.3% of control in the QA-treated group to 50 ± 6.5% of control in the QA plus vehicle-treated group, n = 6) and in GAD ₆₇ mRNA levels (increased from 37 ± 6.7% of control in the QA-treated group to 49 ± 6.8% of control in the QA plus vehicle-treated group, n = 6), although these changes failed to attain statistical significance.

View larger version (45K): [in this window] [in a new window]   Fig. 3.   Effects of PGA1 on QA-induced loss of striatal D1 DA receptors. Rats were treated intrastriatally with PGA1 (5-80 nmol) or vehicle (40% EtOH, 1 µl) 10 min before intrastriatal administration of QA (60 nmol) and killed 10 days later. Brain sections were processed for D1 DA receptor autoradiography. Results from six animals in each group were quantitatively analyzed with an image analyzer and expressed as percentage of control (contralateral striatum, mean ± S.E.M.). Statistical comparisons of QA + Veh or QA + PGA1 with QA-treated group were performed using ANOVA followed by a Dunnett's t test. *p < 0.05.

View larger version (45K): [in this window] [in a new window]   Fig. 4.   Effects of PGA1 on QA-induced loss of striatal GAD67 mRNA. Rats were treated as described in the legend to Fig. 3. Brain sections were processed for in situ hybridization histochemistry for GAD67 mRNA. Results from six animals in each group were quantitatively analyzed with an image analyzer and expressed as percentage of control (contralateral striatum, mean ± S.E.M.). Statistical comparisons of QA + Veh or QA + PGA1 with QA-treated group were performed using ANOVA followed by a Dunnett's t test. *p < 0.05.

View larger version (129K): [in this window] [in a new window]   Fig. 5.   Effects of PGA1 on the QA-induced loss of striatal neurons. Rats were treated intrastriatally with PGA1 (5-80 nmol) or vehicle (40% EtOH, 1 µl) 10 min before intrastriatal administration of QA (60 nmol) and killed 10 days later. Brains were fixed with 4% paraformaldehyde and sectioned for Nissl staining. A and B, normal striatum. C and D, QA (60 nmol) treated. E and F, QA + PGA1 (80 nmol) treated. Note: arrows in B and F indicate Nissl-stained striatal neurons. Asterisks in A, C, and E indicate regions where high-power photographs (B, D, and F) were taken. Arrowheads in C and E indicate needle tracks. In the QA-injected striatum (B and C), the number of Nissl-stained neurons was greatly reduced but Nissl-stained glial nuclei increased. In the QA + PGA1 (80 nmol)-treated striatum (E and F), the loss of Nissl-stained neurons was reduced. Scale bar, 50 µm.

Effect of PGA₁ on QA-Induced Activation of NF-B. Under basal conditions, NF-B binding activity was relatively low. QA (60 nmol) induced a marked increase in NF-B (to 455 ± 54% of control, p < 0.05, n = 6) and AP-1 (to 270 ± 62% of control, p < 0.05, n = 6) binding activities in striatal nuclear extracts. Pretreatment with PGA₁ (5-80 nmol) diminished the QA-induced NF-B increment in a dose-dependent manner. The increased NF-B binding activity in nuclear extracts was reduced by about 50% at the highest dose of PGA₁ (80 nmol, p < 0.05, n = 6, Fig. 6A). In contrast, PGA₁ had no significant effect on QA-induced increases in AP-1 binding activity (Fig. 6B). QA treatment caused a robust reduction in IB- protein levels (to 19 ± 5.2% of control, p < 0.05, n = 5). Pretreatment with PGA₁ (80 nmol) reversed the QA-induced decline in IB- protein levels (increased from 19 ± 5.2% of control in the QA-treated group to 86 ± 11.6% of control in the QA plus PGA₁ group, p < 0.05, n = 5, Fig. 7).

View larger version (43K): [in this window] [in a new window]   Fig. 6.   Effects of PGA1 on QA-induced activation of NF-B and AP-1. Rats were treated intrastriatally with PGA1 (5-80 nmol), vehicle (40% EtOH, 1 µl), or surgical procedures 10 min before intrastriatal administration of QA (60 nmol) and killed 12 h later. Striatal nuclear proteins were extracted for electrophoresis mobility shift assay. Results from six animals in each group were quantitatively analyzed with an image analyzer. Statistical comparisons of QA + Veh or QA + PGA1 with control (animals received surgical procedures only) as well as QA + PGA1 or QA + Veh with QA-treated group were performed using ANOVA followed by a Dunnett's t test. Data were then converted to percentage of control (animals received surgical procedures only, mean ± S.E.M.) for presentation in the bar figures. *p < 0.05 (comparisons of QA + PGA1 with QA-treated group). A, NF-B; B, AP-1.

View larger version (43K): [in this window] [in a new window]   Fig. 7.   Effects of PGA1 on QA-induced degradation of IB-. Rats were treated intrastriatally with PGA1 (80 nmol), vehicle (40% EtOH, 1 µl) or surgical procedures 10 min before intrastriatal administration of QA (60 nmol) and killed 12 h later. Striatal total proteins were extracted for Western blot analysis. Results from five animals in each group were quantitatively analyzed with an image analyzer. Statistical comparisons of QA, QA + Veh, or QA + PGA1 with control group (animals received surgical procedures only) as well as QA + PGA1 or QA + Veh with QA-treated group were performed using ANOVA followed by a Dunnett's t test. The data were then converted to percentage of control (animals received surgical procedures only, mean ± S.E.M.) for presentation in the bar figures. *p < 0.05 (comparisons of QA, QA + PGA1, or QA + Veh with control group); #p < 0.05 (comparisons of QA + PGA1 with QA-treated group).

Effect of PGA₁ on Heat Shock Proteins. Basal striatal levels of HSP72 were undetectable. Striatally infused QA (60 nmol) produced only a modest rise in HSP72 expression. A more than 10-fold increase in HSP72 levels (to 1072 ± 110% of control, p < 0.05, n = 5, Fig. 8A) was observed in animals treated with PGA₁ (80 nmol) before QA administration. Basal levels of 70-kDa HSP were also very low. QA produced an additional band having a lower molecular mass (here named HSP70b as indicated in Fig. 8B). Animals given PGA₁ before QA injection had no statistically significant change in 70-kDa HSP (HSP70a). On the other hand, PGA₁ pretreatment markedly elevated HSP70b (to 1363 ± 219% of control, p < 0.05, n = 5, Fig. 8B). PGA₁ pretreatment had no significant effect on levels of 70-kDa HSC (HSC70, Fig. 8C). HSP70 immunoreactivity (HSP70-i) was very low in normal control animals (Fig. 9A). A few HSP70-i-positive cells near the injection site were observed in animals that received only vehicle treatment (Fig. 9B). Similarly, in QA plus vehicle-injected animals only scatter cells were intensely stained with HSP70 antibody (Fig. 9C). An increase in the number of cells expressing HSP70-i and in the intensity of HSP70-i in striatal neurons were observed near the injection site in QA plus PGA₁ injected striatum (Fig. 9D, filled arrows). High intensity of HSP70-i was also observed in the nerve fiber bundles (Fig. 9D, open arrow). The HSP70-i was totally eliminated by the omission of the primary or secondary antibody (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Effects of PGA1 on QA-induced alterations in HSPs. Rats were treated as described in the legend to Fig. 6. Striatal total proteins were extracted for Western blot analysis. The results from five animals in each group were quantitatively analyzed with an image analyzer. Statistical comparisons of QA, QA + Veh, or QA + PGA1 with control group (untreated animals) as well as QA + PGA1 or QA + Veh with QA-treated group were performed using ANOVA followed by a Dunnett's t test. Data were then converted to percentage of control (untreated animals, mean ± S.E.M.) for presentation in the bar figures. *p < 0.05 (comparisons of QA + PGA1 with control group); #p < 0.05 (comparisons of QA + PGA1 with QA-treated group). A, HSP72; B, HSP70; C, HSC70.

View larger version (119K): [in this window] [in a new window]   Fig. 9.   Immunohistochemical study of induction of HSP70 by PGA1 in striatal neurons. Rats were treated with intrastriatal injection of vehicle (40% EtOH, 1 µl), QA (60 nmol) + vehicle, or QA + PGA1 (80 nmol) and killed 12 h later. Brains were fixed and sectioned as described under Materials and Methods. Sections were processed for immunohistochemistry. A, normal control; B, vehicle; C, QA + vehicle; D, QA + PGA1. Filled arrows indicate striatal neurons with high intensity of HSP70-i after QA. Open arrow indicates nerve bundles with intense HSP70-i after QA. Scale bar, 50 µm.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Prostaglandins are a class of cyclic 20-carbon fatty acids synthesized from polyunsaturated fatty acid precursors derived from cell membranes. The type A and J prostaglandins are characterized by the presence of an , -unsaturated carbonyl group in the cyclopentane ring of the molecule. PGA₁ has been reported to inhibit viral replication, cause cell cycle arrest, increase HSP synthesis, and block NF-B activation (Lacal et al., 1994; D'Onofrio et al., 1995; Rossi et al., 1997, 2000). Based on the results of this study, it now appears that PGA₁ is also capable of attenuating the internucleosomal DNA fragmentation and neuronal loss induced by QA in rat striatum. The results suggest that PGA₁ exerts its neuroprotective effects by inhibiting apoptosis. Conceivably, this action might lead to a heightened degree of necrotic cell death. But this does not appear to have occurred, since there was no increased smearing on the agarose gels to indicate a rise in the random degradation of genomic DNA. Moreover, the decrease in internucleosomal DNA fragmentation was accompanied by a reduction in the loss of D₁ DA receptors and GAD₆₇ mRNA. Since there are no known direct effects of PGA₁ on striatal spiny GABAergic neuron markers, the reduced loss of these markers suggests neuroprotection by PGA₁. Furthermore, the protective effect of PGA₁ was confirmed by the reduced loss of striatal neurons revealed by Nissl staining. Taken together, the present results suggest that PGA₁ has the ability to protect striatal medium spiny neurons against NMDA receptor-mediated toxicity by inhibiting their apoptotic demise.

Since neuroprotection by PGA₁ has not been previously reported, the foregoing observations prompted an evaluation of its effects on an apoptotic cascade linked to the excitotoxin-induced death of striatal neurons. In earlier investigations, we observed that hyperstimulation of NMDA or kainic acid receptors induces NF-B activation through the selective degradation of IB- (Nakai et al., 1999a; Qin et al., 2000). The present results indicate that the in vivo administration of PGA₁ selectively inhibits QA-induced NF-B activation by blocking IB- degradation. The findings are in agreement with those deriving from in vitro studies (Rossi et al., 1997, 2000). PGA₁ had no effect on QA-induced AP-1, suggesting that the neuroprotective action of PGA₁ is not mediated by NMDA receptor blockade. PGA₁'s neuroprotective action also does not appear to involve cAMP, since QA and PGA₁ had no effect on cAMP levels in rat striatum (data not shown). PGA₁ also had no effect on Bcl-2 and Bax levels in rat striatum after QA injection (data not shown). The foregoing observations support the possibility that the neuroprotective effects of PGA₁ against excitotoxicity are mediated, at least in part, by NF-B inhibition and HSP induction. Nevertheless, it should be noted that PGA₁ has multiple pharmacological actions, including blood vessel dilation and inflammatory response inhibition, and whether any of these effects influence acute excitotoxic injury under the conditions of the present study is not known. Although several prostanoid receptors have been characterized, whether the neuroprotective action of PGA₁ is mediated through its receptors remains to be determined (Coleman et al., 1994).

Earlier studies have found that aspirin and sodium salicylate inhibit glutamate toxicity in cultured neurons through a process involving NF-B inhibition (Grilli et al., 1996; Ko et al., 1998). Since PGA₁ reduced both the severity of internucleosomal DNA fragmentation and the size of the lesion induced by QA in rat striatum, the present results are consistent with the possibility that the inhibition of glutamate receptor-stimulated NF-B activation protects against excitotoxin-induced neuronal injury (Qin et al., 1998).

The role of NF-B in the regulation of apoptosis is complicated. In certain circumstances, including tumor necrosis factor-induced apoptosis in dividing somatic cells, NF-B signaling is associated with cell survival (Antwerp et al., 1996; Beg and Baltimore, 1996; Mattson et al., 1997). But under different conditions, such as the excitotoxic destruction of postmitotic neurons, NF-B activation promotes apoptosis (Qin et al., 1998; Nakai et al., 1999a; Schneider et al., 1999). Whether these opposite actions reflect differences in the type or maturity of the cells being studied, in the apoptotic triggers being used, or whether the studies were conducted in vitro or in vivo remains to be determined. Interestingly, a recent investigation has found that NF-B can play an antiapoptotic or proapoptotic role within the same type of cells (T-cell hybridomas) in response to different apoptotic stimuli (Lin et al., 1999). NF-B-regulated cell cycle entry may be another critical factor in determining its differing effects on apoptosis in proliferating cells and postmitotic neurons. NF-B positively regulates cyclin D1 expression and stimulates G₀/G₁-to-S phase transition (Hinz et al., 1999). It now appears that cell cycle mediators contribute to the induction of neuronal apoptosis in response to ischemia, excitotoxins, oxidative stress, or nerve growth factor withdrawal (Freeman et al., 1994; Kranenburg et al., 1996; Park et al., 1998). Moreover, cell cycle inhibitors or cyclin-dependent kinase inhibitors attenuate apoptosis triggered by nerve growth factor withdrawal in cultured neurons (Freeman et al., 1994; Park et al., 1997). In other studies, we found that cyclin D1 is induced by the NMDA receptor agonist QA, and cyclin D1 induction can be reduced by the NF-B inhibitors, including PGA₁ (Z.-H. Qin, R.-W. Chen, Y. Wang, X. Wang, D.-M. Chuang, and T. N. Chase, in preparation). The inhibition of NF-B activation in the present study supports earlier reports showing PGA₁ causes cell cycle arrest (Goubin et al., 1986; Hughes-Fulford, 1994), which could serve as one of the mechanisms underlying neuroprotection by PGA₁.

HSPs may contribute to the protective effects of PGA₁. PGA₁ has been reported to increase protein levels of HSPs (D'Onofrio et al., 1994; Lacal et al., 1994). Here we found that PGA₁ markedly increased levels of 70 and 72 HSPs in rat striatum. HSPs are a highly conserved, finely regulated, cellular defense mechanism known to be induced in various pathological states (Sloviter and Lowenstein, 1992). Indeed, 70- and 72-kDa HSPs can have neuroprotective effects against various insults, including glutamate toxicity (Lowenstein et al., 1991; Rordorf et al., 1991). Overexpression of 72-kDa HSP in vivo with viral vectors protects striatal and hippocampal neurons from ischemia- and kainic acid-induced damage (Yenari et al., 1998). HSP induction could thus serve as an important mediator of the ability of PGA₁ to protect striatal neurons against excitotoxin-induced apoptosis (Mailhos et al., 1993). HSPs are known to have multiple influences on cellular function, including protein folding and trafficking as well as intracellular signal transduction, and exactly how they might counteract QA-induced neuronal apoptotic cascades remains to be elucidated.

Excitotoxicity has been proposed to contribute the pathogenesis of a number of neurodegenerative disorders, including stroke, Huntington's disease, Parkinson's disease, AD, and amyotrophic lateral sclerosis. If correct, inhibiting the effects of glutamate receptor hyperstimulation could act to retard the degenerative process. Given the prominent role of NF-B and heat shock proteins in neuronal survival as well as the present finding that pretreatment or post-treatment with PGA₁ inhibits the QA-induced apoptotic death of these neurons, it is tempting to speculate that drugs capable of inhibiting NF-B cascade and inducing heat shock proteins may be useful in the treatment of neurodegenerative disorders where excitotoxic mechanisms contribute to pathogenesis.