- Email: [email protected]
Histological and Behavioral Protection by (−)-Nicotine against Quinolinic Acid-Induced Neurodegeneration in the Hippocampus
NEUROBIOLOGY OF LEARNING AND MEMORY ARTICLE NO.
69, 46–64 (1998)
NL973803
Histological and Behavioral Protection by (0)-Nicotine against Quinolinic Acid-Induced Neurodegeneration in the Hippocampus Alyssa B. O’Neill,* Sherry J. Morgan,† and Jorge D. Brioni*,1 *Neurological and Urological Diseases Research and †Pathology Department, Pharmaceutical Products Division, Abbott Laboratories, Abbott Park, Illinois 60064-3500
Injections of quinolinic acid (60, 180, and 600 nmol) in the dorsal hippocampus induced significant neurotoxicity that was evident 1 day after the injection. By day 3, pyramidal as well as granular cells were affected even at the lowest dose of quinolinic acid, an effect that persisted up to 20 days. Consistent with the histological findings, animals with bilateral injections in the dorsal hippocampus were cognitively impaired during acquisition and retention of spatial information in the water maze. A subacute treatment with (0)-nicotine (62 mmol/kg/day) delivered by subcutaneous minipumps prevented the histological and cognitive deficits induced by the bilateral quinolinic acid (60 nmol) injections. These data indicate that quinolinic acid can induce degeneration of both pyramidal as well as granule cells in the hippocampus, leading to cognitive impairments in the rat, and that activation of neuronal nicotinic acetylcholine receptors can prevent the neurodegenerative process induced by quinolinic acid. q 1998 Academic Press Key Words: hippocampus; learning; memory; nicotine; quinolinic acid.
Nicotinic acetylcholine receptor (nAChR) agonists can induce several pharmacological effects in rodents including hormone release, analgesia, memory facilitation and increased attention, by virtue of their specific binding to selective nAChR subtypes in the brain (Arneric, Sullivan, & Williams, 1995; Brioni, Decker, Sullivan, & Arneric, 1996; Levin, 1992). A relatively novel finding has been the demonstration that activation of neuronal nAChRs can induce neuroprotection in different models of neuronal injury. In vitro studies have revealed that (0)-nicotine protected striatal and cortical neurons against NMDA-induced neurotoxicity but not against kainic-induced toxicity (Akaike, Tamura, Yokaota, Shimohama, & Kimura, 1994; Marin, Maus, Desagher, Glowinski, & Premont, 1994). This neuroprotective effect of (0)-nicotine is mimicked by ABT-418 and GTS-21, two novel cholinergic channel activators; ABT-418 reversed glutamate-induced toxicity in primary cultures of rat cortical neurons and human neuroblastoma IMR32 cells (DonnellyRoberts, Xue, Arneric, & Sullivan, 1996), and GTS-21 prevented PC12 cell death after nerve growth factor removal (Martin, Panickar, King, Deyrup, Hunter, Wang, & Meyer, 1994). Similarly, in vivo studies have shown that (0)-nicotine reduced the loss of tyrosine-hydroxylase immunoreactive neurons after partial hemitransection of the mesostriatal dopaminergic pathway (Jan1
We thank Ms. Theresa Sykora and Mr. Steven Postl for technical support. Address reprint requests to Dr. Jorge D. Brioni, Neurological and Urological Diseases Research, Building AP10LL (47C), Abbott Laboratories, Abbott Park, IL 60064-3500. Fax: (847) 937-9195. 1074-7427/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
NLM 3803
/
46
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
47
son, Fuxe, Agnati, Katayama, Harfstrand, Anderson, & Goldstein, 1988; Owman, Fuxe, Jason, & Kohrstrom, 1989) and reduced cortical neuronal loss after ibotenic acid injections in the nucleus basalis magnocellularis in rats (SjakShie & Meyer, 1993). Quinolinic acid (pyridine-2,3-dicarboxylic acid) is an endogenous substance in the mammalian brain that can be synthesized from tryptophan by several enzymes involved in the kynurenine pathway (Reinhard, Erickson, & Flanagan, 1994; Stone, 1993). Although it has been described as an intermediate in the synthesis of nicotinamide adenine dinucleotide cofactors, current interest has focused on its neurotoxic properties and its eventual participation in several neurological disorders like Huntington’s disease and AIDS-dementia (Heyes, Saito, Major, Milstien, Markey, & Vickers, 1993; Lipton & Rosenberg, 1994; Schwarcz, Whetsell, & Mangano, 1982). Electrophysiological studies have revealed that quinolinic acid is as potent as glutamate or aspartate to stimulate rat cortical neurons, an effect blocked by the N-methyl-D-aspartate (NMDA) antagonist, 2 APV (Stone & Perkins, 1981). Injections of quinolinic acid (120 nmol) in several brain areas induced different degrees of neurodegeneration that might reflect the differential sensitivity of mammalian neurons to its toxic actions (Schwarcz & Kohler, 1983); the striatum, hippocampus, and globus pallidus were more sensitive to the toxic effect in comparison to the cerebellum, amygdala, septum and hypothalamus, with a selective vulnerability of pyramidal cells in comparison to granular cells in the hippocampus (Schwarcz & Kohler, 1983). In the present study we investigated the time course and potency of the neurotoxic actions of quinolinic acid after injections in the dorsal hippocampus and attempted to correlate the histological changes with the performance of rats in a spatial memory test that is very sensitive to hippocampal function. As the activation of neuronal nAChRs has been shown to provide neuroprotection in several models (Brioni, Morgan, O’Neill, Sykora, Postl, Pan, Sullivan, & Arneric, 1996), the ability of (0)-nicotine to reverse the histological and cognitive effects of quinolinic acid injections in the dorsal hippocampus was later investigated. MATERIALS AND METHODS Animal studies were conducted according to the guidelines of the American Association for the Accreditation of Laboratory Animal Care (AAALAC) and according to protocols approved by the Abbott Institutional Animal Care and Use Committee. Animals Male Long Evans rats (weighing 200 g on arrival) supplied by Harlan (Altamont, NY) were used. They were individually housed and acclimated to laboratory conditions for 1 week before the studies. Animals were kept in a climatecontrolled facility with a 12-h light/dark cycle. Testing was conducted during the light portion of the day. Surgery Intracerebral injections of quinolinic acid (18, 60, 180, and 600 nmol in 0.5 ml) were performed in animals anesthetized with 50 mg/kg pentobarbital
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
48
O’NEILL, MORGAN, AND BRIONI
(Nembutal, Abbott Laboratories). Injections were placed in the dorsal hippocampus under stereotaxic control (AP Å 03.3; ML, {2.5; DV, 03.6 from dura) and were made slowly through an injecting cannula attached by polyethylene tubing to a 10-ml syringe and a motor-driven minipump. A total volume of 0.5 ml was injected in a 4-min period and the cannula was left in place another 2 min to allow diffusion of the drug. When bilateral injections were needed for the behavioral studies, they were performed simultaneously. Animals were allowed to recover on a warming plate before returning them to the cage. Spatial Learning in the Water Maze Apparatus. A cylindrical water tank (60 cm high and 180 cm in diameter) was filled to a depth of 37 cm with 26 { 17C water rendered opaque by the addition of dry milk. Close to the rim of the tank along 1.50 m of its circumference, a white curtain that extended to the ceiling of the room served as a salient cue. Other cues were also available in the room. Four points equally spaced around the perimeter of the tank were arbitrarily designated as starting locations. On this basis, the tank was divided into four equal quadrants. Located in the center of one of the quadrants was a 13-cm wide circular platform. Cue training procedure. During this training version in the water maze, a black platform protruded 1 cm over the water surface and was visible to the animals. A trial was initiated by placing the rat in the water at one of the four starting locations along the perimeter of the pool, which varied from trial to trial in a quasirandom order (Brioni, Decker, Gamboa, Izquierdo, & McGaugh, 1990). The rats were allowed to swim until they located the escape platform and climbed onto it. If the rats did not locate the escape platform within 90 s, they were gently guided to it. The rats remained in the platform for 20 s before being removed. The following trial was conducted 5–10 s later. Animals received a total of eight trials in two sessions (the second session was conducted 4 h after the first one). Place training procedure. Three days after cue training was completed, animals were submitted to place training in the water maze. The platform remained in a fixed location throughout this experiment in order to evaluate reference memory in the rats (Morris, 1984). Animals received six trials per day, in two sets of 3 trials separated by an interval of 4 h. The rats received a total of 18 trials during 3 days of place training. Retention of the spatial information was assessed in a single probe trial conducted 24 h after the last place training session in the absence of the escape platform (free-swim trial). Animals were videotaped and data analyzed by the Poly-Track tracking software (San Diego Instruments, CA). Continuous Subcutaneous Treatment with (0)-Nicotine For the subacute treatment with (0)-nicotine, rats were implanted with Alza minipumps (Alzet, Model 2ML4, Alza Corporation, Palo Alto, CA) containing (0)-nicotine and delivering approximately 19 and 62 mmol/kg/day of (0)-nicotine. In the control group the minipumps delivered sterile physiological saline solution. Minipumps were surgically implanted subcutaneously. Surgery was conducted under sterile conditions using halothane anesthesia (4% in the induction period and 2% for maintenance). An incision was made in the back of
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
49
the animals and a pump was inserted subcutaneously. The wound was closed with wound-clips, and the animal was allow to recover on a heating pad before being returned to its home cage. Animals recovered from anesthesia within 5 min. Animals received bilateral injections of quinolinic acid in the dorsal hippocampus 7 days after implantation of the minipumps. Place training in the water maze started 7 days later and animals were sacrificed after the freeswim trial (10 days after the quinolinic injections). Histological Procedures Rats were deeply anesthetized with 100 mg/kg i.p. pentobarbital prior to being perfused transcardially with saline solution followed by 10% phosphatebuffered formalin. The brains were then removed and immersion-fixed for 24 h in the same fixative. The brains were placed in a tissue slicer with 3-mm gradations, and after wet-cutting, the brain slices were processed and embedded in paraffin. Hippocampal sections (6 mm) were cut coronally, mounted, and stained with hematoxilin and eosin. A detailed evaluation of histologic changes were performed on each specimen by an experimenter (S.J.M.) unaware of the treatment conditions. In addition, cell counts of viable neurons were performed on the specimens from the experiment that compared the effect of (0)-nicotine on quinolinic-induced lesions. The mean of four matched coronal sections from different rostrocaudal levels of the dorsal hippocampus of each subject was measured along a 500-mm length by means of an eye piece graticule. For the CA1–CA4 regions, the counts were made with a 201 objective; for the dorsal and ventral dentate, the counts were made with a 401 objective. Cell counts were expressed as the percentage of the number of viable cells in comparison to the control group. Drugs Quinolinic acid (Sigma Chemical Co., St. Louis, MO) was prepared in physiological saline solution and adjusted to pH 5.8 with 0.5 N NaOH. (0)-Nicotine bitartrate (Sigma) was dissolved in saline solution. Fresh solutions were prepared every day. Statistics Data were analyzed by one- or two-way analyses of variance (ANOVA) followed by the Fisher Protected Least Significance Difference test for individual means comparisons. RESULTS Effect of Quinolinic Acid Injections in the Hippocampus A group of rats was injected unilaterally with saline or quinolinic acid (60, 180, and 600 nmol) and evaluated histologically 1 and 3 days after the injection. Histologic examination of the hippocampus of rats necropsied 1 day postinjec-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
50
O’NEILL, MORGAN, AND BRIONI
FIG. 1. Neurotoxic effect of quinolinic acid 1 day after a unilateral injection in the dorsal hippocampus. (A) Control; (B) 60 nmol; (C) 180 nmol; (D) 600 nmol (120). Bar Å 200 mm.
tion revealed a dose-related neuronal loss (Fig. 1). Those rats that received 60 nmol of quinolinic acid revealed a unilateral extensive loss of hippocampal neurons. The most extensive loss was noted in the pyramidal regions, whereas there was some degree of neuronal retention in the dorsal dentate and, to a
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
51
FIG. 1—Continued
lesser extent, the ventral dentate (Fig. 1B). Cells in the CA4, CA3, CA2, and CA1 pyramidal region which were interpreted as necrotic were typical of early stage ischemic neurons, being characterized by the presence of an intensely eosinophilic, somewhat spindle shaped cytoplasm and an oval to round nucleus with variable retention of detail. In contrast, cells in the dentate region which
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
52
O’NEILL, MORGAN, AND BRIONI
FIG. 2. Neurotoxic effect of quinolinic acid 3 days after a unilateral injection in the dorsal hippocampus. (A) 60 nmol; (B) 180 nmol; (C) 600 nmol (120). Bar Å 200 mm.
were interpreted as necrotic had an imperceptible cytoplasm and a small, dense, round nucleus with no retention of detail. Other changes noted in these rats included moderate spongiosis (primarily perineuronal). Those rats that received 180 nmol of quinolinic acid (Fig. 1C) exhibited a unilateral complete loss of hippocampal neurons with some loss in the contralateral CA1 region as well. The morphology of the necrotic cells was similar to that as described for the rats that received 60 nmol. Perineuronal spongiosis was present although more severe than that noted in the 60-nmol group. The rats that received 600 nmol of quinolinic acid exhibited lesions which were similar, but more severe, than those that received 180 nmol (Fig. 1D). Histological changes noted in those rats necropsied 3 days postinjection revealed considerable progression of the lesions in all dosage groups (Fig. 2). Those rats that received 60 nmol of quinolinic acid showed a nearly complete loss of neurons; neuronal sparing, when present, was generally confined to the dorsal blade of the dentate gyrus. Morphologic changes in necrotic neurons were similar to those noted at 1 day postinjection with the exception of loss of nuclear detail in the ischemic neurons (CA4, CA3, CA2, and CA1). Hippocampal perineuronal spongiosis, as noted at 1 day postinjection was still present (Fig. 2A). Rats that received 180 or 600 nmol of quinolinic acid exhibited lesions that were similar, but more severe, than those noted at 1 day postinjection (Figs. 2B and 2C). Effect of Quinolinic Injections in the Hippocampus on Spatial Learning In order to determine the effect of quinolinic acid on cognitive performance, another group of rats received bilateral injections of quinolinic acid (18, 60,
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
53
FIG. 2—Continued
and 180 nmoles) and 7 days later were submitted to the water maze test. This dose range was selected based on the previous histological data after unilateral injections of quinolinic acid that showed severe lesions in the 600 nmol group. During cue training in the water maze, there were no differences in the latencies to reach the escape platform between the sham and the quinolinic
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
54
O’NEILL, MORGAN, AND BRIONI
FIG. 3. Effect of bilateral injections of quinolinic acid in the dorsal hippocampus on place learning in the water maze. Data represent the mean escape latencies of nine rats in each group. Training began 7 days after the lesions. In the place version of the maze, animals were submitted to two daily sessions and received three trials per session. **p õ .01; *** p õ .001 as compared to sham rats.
18- or 60-nmol groups in the last four trials; however, the quinolinic 180-nmol group exhibited significantly longer latencies to locate the visible platform (Control, 27.8 { 4.6 s; Quin 18, 21.2 { 5.3 s; Quin 60, 40.2 { 5.2 s; Quin 180, 53.7 { 3.7 s; p õ .05 as compared to control rats). Figure 3 shows the performance of the rats during place training in the water maze. Under our experimental conditions, by the third day sham rats were able to learn the test and to escape into the platform in approximately 10 s. Animals injected with quinolinic acid exhibited a significant impairment in the water maze. There was a significant quinolinic effect [F(3, 32) Å 32.3, p õ 0.001], a significant session effect [F(5, 160) Å 12.8, p õ .001], and a significant quinolinic 1 session interaction [F(15, 160) Å 2.3, p õ .01]. Animals that received 18 nmol performed as well as the sham group, while the groups receiving 60 and 180 nmol were significantly impaired in the acquisition of spatial information. Table 1A shows the performance of the rats during the free-swim trial conducted 24 h after the last training session. The original position of the platform was video-taped and the platform was removed from the tank. Quinolinicinjected animals exhibited significant longer latencies in comparison to the sham group [F(3, 32) Å 12.0, p õ .001] and a reduced number of annulus crossings [F(3, 32) Å 8.1, p õ .001], that reflect an impairment on retention of spatial information. Histological evaluation of the hippocampus conducted 20 days after the injection of quinolinic acid revealed a dose-related neuronal loss (Fig. 4). In those rats that received 18 nmol of quinolinic acid, there was a substantial bilateral neuronal necrosis (up to 90%) only in the CA3 and CA4 regions, as it occurred
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
55
TABLE 1 Performance in the Free-Swim Trial Conducted 24 h after Training in the Water Maze Latency (s)
Crossings
A Sham Quinolinic 18 mmol Quinolinic 60 mmol Quinolinic 180 mmol
10.4 10.0 25.7 26.6
{ { { {
1.8 3.1 3.0*** 2.7***
1.6 1.8 0.2 0.2
{ { { {
0.3 0.5 0.1** 0.1**
{ { { / { {
3.3 3.2 4.4 4.7* 2.4** 3.2
1.9 1.2 1.2 0.5 0.4 1.1
{ { { { { {
0.3 0.2 0.4 0.3*** 0.2*** 0.2†
B Sham NIC 19 mmol/kg/day NIC 62 mmol/kg/day Quinolinic 60 mmol Quin 60 / NIC 19 Quin 60 / NIC 62
9.1 10.1 15.1 20.3 26.3 14.8
Note. Data represent the mean { SEM latency to reach the platform location and annulus crossings during the 30-s free swim. * p õ .05, **p õ .01, ***p õ .001 as compared to sham rats. † p õ .05 as compared to the Quin 60 group.
to a lesser extent (10–20%) in the CA2, CA1, and ventral dentate regions (Fig. 4A). Occasional cells (õ5%) in the dorsal dentate were necrotic. Other changes noted primarily in the CA4 region included a marked infiltrate of gemistocytic astrocytes, moderate microgliosis, and moderate mineralization, involving both neurons and the neuropil. On the contrary, changes noted in rats that received 60 nmol quinolinic acid included near total (ú95%) loss of neurons with no evidence of preferential retention (Fig. 4B). Other hippocampal changes such as mineralization, microgliosis, gemistocytic astrocyte infiltration, and neovascularization were more severe than those seen in the 18-nmol group. A change noted in this group but not in the 18-nmol group included the presence of hippocampal parenchymal collapse. In rats that received 180 nmol quinolinic acid there was a total loss of neurons (Fig. 4C). Other hippocampal changes as noted in the 60-nmol group were apparent, but were more severe; for example, the hippocampal parenchymal collapse was so severe that orientation and identification of structures was difficult. The presence of cyst-like foci containing lipid-laden macrophages in the CA4 region was noticed in this group. Thus, although the extent of neuronal loss was similar to that noted at 3 days postinjection, evaluation at this time point revealed progression of the parenchymal changes (cellular infiltration and parenchymal collapse). Effect of a Subacute Treatment with (0)-Nicotine on Quinolinic-Induced Histological and Behavioral Changes In order to determine the neuroprotective effect of (0)-nicotine in quinolinicinjected rats, three groups of rats were implanted with chronic subcutaneous minipumps delivering physiological saline, nicotine 19 or 62 mmol/kg/day. Seven days after implantation of the minipumps, bilateral injections of quino-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
56
O’NEILL, MORGAN, AND BRIONI
FIG. 4. Morphological changes induced by of quinolinic acid 20 days after bilateral injections in the dorsal hippocampus. (A) 18 nmol; (B) 60 nmol; (C) 180 nmol (120). Bar Å 200 mm.
linic acid (60 nmol) or saline were carried out in the dorsal hippocampus. When animals were submitted to the cue version of the water maze there were no differences between the groups as they were all able to locate the escape platform (data not shown). Figure 5 shows the performance of the rats in the place version of the test conducted 7 days after the quinolinic acid injection. There was a significant effect of the drug treatment [F(5, 42) Å 12.2, p õ .001], a significant session effect [F(5, 210) Å 38.6, p õ .001], and a significant drug 1 session interaction [F(25, 210) Å 2.8, p õ .001]. An independent analysis of the nonlesioned animals that received saline or the (0)-nicotine doses revealed that the drug 1 session interaction was close to significance [F(5, 105) Å 1.83, p Å .064], as (0)-nicotine tended to improve the performance of sham animals. Sham rats were able to locate the escape platform in approximately 10 s by the last session, while the rats injected with quinolinic acid (60 nmol) were significantly impaired throughout all sessions. However, animals that received the subacute treatment with (0)-nicotine 62 mmol/kg/day performed significantly better during the last three sessions. In the free-swim trial conducted 24 h later (Table 1B), there was a significant effect on the escape latencies [F(5, 41) Å 3.1, p õ .05] and on the number of platform crossings [F(5, 41) Å 4.5, p õ .01] as quinolinic-injected rats performed significantly worse than the control group, with the exception of those rats treated with 62 mmol/kg/day (0)-nicotine. Rats that received (0)-nicotine alone (19 or 62 mg/kg/day) exhibited lesions indistinguishable from those of control rats (Fig. 6A). Changes common to these groups included the presence of injection tracts (linear gliosis, hemosiderosis, and occasional neovascularization) along with focal loss of neurons. Blood samples were collected from the animals receiving the subacute (0)-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
57
FIG. 4—Continued
nicotine treatment and the plasma levels were determined by HPLC/ED as previously described (Decker, Brioni, Sullivan, Buckley, Radek, Raskiewickz, Kang, Kim, Giardina, Williams, & Arneric, 1994). Plasma (0)-nicotine levels for the animals receiving 19 or 62 mmol/kg/day averaged 36.2 { 9.6 and 126.8 { 41.9 ng/ml, respectively.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
58
O’NEILL, MORGAN, AND BRIONI
FIG. 5. Reversal of quinolinic-induced cognitive impairments in the water maze by a subacute treatment with (0)-nicotine. Data represent the mean escape latencies of eight rats in each group. Place training began 7 days after the lesions. Animals were submitted to two daily sessions and received three trials per session. *p õ .05; **p õ .01; ***p õ .001 as compared to sham rats. // p õ .01; /// p õ .001 as compared to the quinolinic 60-nmol group.
Histologic examination of the hippocampus 10 days after 60-nmol quinolinic acid injections (Fig. 6B) was similar to the 60-nmol rats that were necropsied at 20 days (see above). The primary difference between these rats and those necropsied at 20 days was the presence of cyst-like foci in the CA4 region in those necropsied at this time point and hippocampal parenchymal collapse in those necropsied at 20 days. The hippocampal parenchymal collapse is likely a sequelae for the cystic-like structures in the CA4 region. Other changes (mineralization, microgliosis, gemistocytic astrocyte infiltration, neovascularization) present at 20 days were also present at this time point but were less pronounced. Quantification of viable hippocampal neurons was performed on specimens from each group necropsied at 10 days (Table 2), and the neuronal counts for all hippocampal regions were significantly lower in quinolinic-injected rats than in saline-injected rats [F(1, 4) Å 2,618, p õ .0001]. Rats that received 60 nmol quinolinic acid along with 19 mmol/kg/day (0)-nicotine exhibited morphological changes that were nearly identical to that noted in the rats that received 60 nmol quinolinic acid alone, and the neuron counts in this group were similar to those in the rats that received quinolinic acid alone. Rats that received 60 nmol quinolinic acid along with 62 mmol/kg/day (0)-nicotine exhibited a definite sparing of the neuronal loss (Table 2; Fig. 6C). With the exception of the ventral dentate region, all regions of the hippocampus had significantly higher neuron counts in this treatment group than seen in the rats that received quinolinic acid alone (all F ’s p õ .001).
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
59
DISCUSSION These data demonstrate that quinolinic acid can induce potent and long lasting neurotoxic actions on pyramidal as well as granular cells in the rat dorsal hippocampus. The hippocampal cell loss was associated with the performance of rats in a spatial memory test as lesioned rats were significantly impaired during acquisition and retention of spatial information in the water maze. However, a subacute treatment with (0)-nicotine delivered via subcutaneously implanted minipumps prevented the neurotoxic and cognitive-impairing effects of quinolinic acid. Quinolinic acid is an endogenous substance in the mammalian brain synthesized from tryptophan by several enzymes involved in the kynurenine pathway (Reinhard et al., 1994; Stone, 1993) that can exert neurotoxic actions in the striatum and in the hippocampus (Schwarcz et al., 1982). Further studies have shown regional variations in the sensitivity to quinolinic acid (120 nmol) as the striatum, hippocampus, and cortex were more sensitive than the amygdala, septum, or hypothalamus, whereas negligible effects were noted in the cerebellum (Perkins & Stone, 1983; Schwarcz & Kohler, 1983). When injected in the basal forebrain, quinolinic acid was less potent than kainic acid but more potent than quisqualate and ibotenic acid (Winn, Stone, Latimer, Hastings, & Clark, 1991). Consistent with previous findings with other neurotoxic agents like kainic acid or quisqualic acid, injections of quinolinic acid in selective brain areas induced marked behavioral dysfunction in rats. Injections of quinolinic acid in the striatum have been shown to reduce sensorimotor gating (Kodsi & Swerdlow, 1994), impair active avoidance learning (Vecsei & Beal, 1991), and impair the performance of rats in spatial memory tests (Block, Kunkel, & Schwarz, 1993). Injections in the nucleus basalis magnocellularis impaired T-maze alternation and impaired working memory in the radial arm-maze consistent with a significant decrease in cortical choline acetyltransferase activity in lesioned rats (Beninger, Wirsching, Jhamandas, Boegman, & ElDefrawy, 1986; Wirsching, Beninger, Jhamandas, Boegman, & Bialik, 1989). With regard to the hippocampus, neuronal loss after quinolinic injections have been observed 4 days (Schwarcz & Kohler, 1983; Schwarcz et al., 1982) and 7 days after the injections (Foster & Woodruff, 1988). The present data demonstrate that quinolinic-induced cell death can be detected as early as 1 day, with pyramidal cells being more sensitive than granular cells, as indicated by the sparing of the ventral granular layer at this time point. The lesion progressed as a complete loss of pyramidal and granular neurons was observed 3 days after the injection. Although it was reported that quinolinic injections in the hippocampus induced no behavioral effects in rats despite the presence of an extensive histological damage (Schwarcz & Kohler, 1983), a significant impairment during acquisition of spatial information was noted in our experiments in those groups receiving bilateral injections of 60 and 180 nmol. As the 60-nmol group did not show any impairment in the cue version of the test, these data indicate that the behavioral effects are not due to sensorimotor impairment in the rats but due to a specific place learning deficit. The histological evaluation conducted 20 days after the injections showed that the morphological characteristic observed 1 and 3 days after the injections were still present or even exacerbated, indicating the permanent nature of the lesion induced by quinolinic acid.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
60
O’NEILL, MORGAN, AND BRIONI
FIG. 6. Neuroprotective actions of (0)-nicotine against quinolinic acid-induced toxicity. (A) Control; (B) quinolinic acid 60 nmol; (C) quinolinic acid 60 nmol / 62 mmol/kg/day (0)-nicotine (120). Bar Å 200 mm.
The search for pharmacological agents to counteract the ongoing neurodegenerative process present in Parkinson’s disease, Alzheimer’s disease, Huntington’s disease, amyotrophic lateral sclerosis, AIDS-dementia, epilepsy, or ischemia is a major scientific challenge at the present time. The list of potential targets include several classes of drugs such as NMDA receptors antagonists, GABAergic agonists, kynurenine analogs, antioxidants, adenosine agonists, growth factors, and Ca2/ channel antagonists. The in vivo and in vitro neuroprotective effects of (0)-nicotine in several models of neurodegeneration suggest that the neuronal nAChR could represent a novel therapeutic target (Brioni et al., 1996). Subchronic administration of (0)-nicotine reduced the loss of tyroxine-hydroxylase immunoreactive neurons after partial hemitransection of the mesostriatal dopaminergic pathway and attenuated the histological and behavioral effect of perinatal asphyxia on central dopaminergic pathways (Chen, Ogren, Bjelke, Bolme, Eneroth, Gross, Loidl, Herrera-Marschitz, & Anderson, 1995; Janson, Fuxe, Agnati, Katayama, Harfstrand, Anderson, & Goldstein, 1988; Owman et al., 1989). However, the neuroprotective actions of (0)-nicotine are not universal, as in rats with fimbria–fornix transection (0)-nicotine did not prevent but actually increased the disappearance of septal neurons (Fuxe, Rosen, Lippoldt, Andbjer, Hasselrot, Finman, & Agnati, 1994). Several studies in cultured cells have corroborated the neuroprotective effect of (0)-nicotine in rats. (0)-Nicotine protected against NMDA-induced neurotoxicity in striatal and cortical neurons, while the novel cholinergic channel modulator, ABT-418, reversed glutamate-induced toxicity in rat cortical neurons and human neuroblastoma IMR32 cells (Akaike et al., 1994; Donnelly-Roberts et al., 1996; Marin et al., 1994). Another novel
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
61
FIG. 6—Continued
cholinergic channel modulator with neuroprotective properties is GTS-21 as this compound prevented PC12 cell death after removal of nerve growth factor from the medium (Martin et al., 1994). In order to determine the potential neuroprotective effect of (0)-nicotine against quinolinic-induced lesions, a group of rats were implanted with subcu-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
62
O’NEILL, MORGAN, AND BRIONI
TABLE 2 Quantitative Analysis of the Effect of (0)-Nicotine on Quinolinic-Induced Neurodegeneration in the Hippocampus Pyramidal layer Treatment
CA1
CA2
Saline / saline 100.0 NIC 19 / saline 100.5 { 3.3 NIC 62 / saline 94.8 { 1.5 Saline / Quin 60 13.1 { .6*** NIC 19 / Quin 60 15.2 { 3.9*** NIC 62 / Quin 60 37.4 { 10.8***,†
Dentate gyrus
CA3
100.0 100.0 89.7 { 13.2 110.6 { 3.5 84.2 { 6.6 99.8 { 2.1 9.7 { 1.8*** 11.9 { 1.7*** 10.2 { 1.5*** 12.3 { 2.6*** 48.6 { 18.7*,† 31.1 { 7.6***,†
CA4
99.1 96.2 2.2 6.7 21.8
Dorsal
100.0 { 3.9 { 2.3 { .8*** { 4.7*** { 5.3***,††
95.0 98.1 10.8 12.9 46.1
100.0 { 3.4 { 1.5 { 2.3*** { 4.7*** { 15.7**,†
Ventral 100.0 92.7 { 3.7 93.0 { 3.5 4.2 { 1.4*** 15.3 { 9.9*** 21.5 { 7.1***
Note. Number of viable neurons for each hippocampal subfield expressed as a percentage of the saline–saline group. The mean of four matched coronal sections from the dorsal hippocampus of each subject were averaged across the treatment group (n Å 2 in each saline-injected group and n Å 4 in each quinolinic-injected group). * p õ .05, **p õ .01, ***p õ .001 as compared to saline-injected rats. † p õ .05, ††p õ .01 as compared to quinolinic-injected rats.
taneous minipumps delivering 19 or 62 mmol/kg/day (0)-nicotine and 1 week later injected with quinolinic acid in the dorsal hippocampus. Behavioral studies in the water maze indicated that the chronic treatment with (0)-nicotine tended to facilitate acquisition of spatial information in nonlesioned rats, consistent with previous reports in the field (Levin, 1992). Quinolinic injected rats receiving chronic vehicle solution or 19 mmol/kg/day (0)-nicotine were impaired throughout the sessions, but the group receiving 62 mmol/kg/day (0)-nicotine exhibited a significant improvement in the last three sessions in comparison to the lesioned group. The histological analysis in those animals was consistent with the behavioral findings as neuroprotection was observed in those animals receiving the larger dose of (0)-nicotine. Preliminary data from our laboratory indicate that 10 to 20% of the neurons undergoing degeneration exhibit apoptosis after the 60-nmol injection of quinolinic acid (O’Neill, Morgan, Sykora, Postl, & Brioni, 1996). Whether the neuroprotective actions of (0)-nicotine can preferentially block apoptosis or necrotic type mechanisms of neuronal death is unclear at the present time but is a subject that merits further investigation. It is unknown which nAChR subunits mediate the neuroprotective actions of (0)-nicotine, although the ability of methyllycaconitine and a-bungarotoxin to block (0)-nicotine’s protection against glutamate-induced toxicity might indicate the participation of the a7 nAChR subunit (Donnelly-Roberts et al., 1996). Future work aiming to identify the specific nAChR subunit/s involved in the neuroprotective effect of nAChR agonists will allow the identification of novel ligands with greater potency and efficacy. As quinolinic acid is an endogenous substance that can exert neurotoxic effects, its eventual participation in several neurological and neuroinflammatory diseases like Huntington’s disease and AIDS-dementia has been proposed (Heyes et al., 1993; Lipton & Rosenberg, 1994). The neuroprotective actions of nAChR ligands indicate that novel nAChR ligands with neuroprotective efficacy might be useful for the treatment of these disorders associated with increased quinolinic acid levels in the brain.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
63
REFERENCES Akaike, A., Tamura, Y., Yokaota, T., Shimohama, S., & Kimura, J. (1994). Nicotine-induced protection of cultured cortical neurons against N-methyl-D-aspartate receptor-mediated glutamate cytotoxicity. Brain Research, 644, 181–187. Arneric, S., Sullivan, J., & Williams, M. (1995). Neuronal nicotinic acetylcholine receptors: Novel targets for CNS therapeutics. In F. Bloom & D. Kupfer (Eds.), Psychopharmacology: Fourth generation of progress (pp. 95–110). New York: Raven Press. Beninger, R. J., Wirsching, B. A., Jhamandas, K., Boegman, R. J., & El-Defrawy, S. R. (1986). Effects of altered cholinergic function on working and reference memory in the rat. Canadian Journal Physiology and Pharmacology, 64, 376–382. Block, F., Kunkel, M., & Schwarz, M. (1993). Quinolinic acid lesion of the striatum induces impairment in spatial learning and motor performance in rats. Neuroscience Letters, 149, 126–128. Brioni, J. D., Decker, M. W., Gamboa, L. P., Izquierdo, I., & McGaugh, J. L. (1990). Muscimol injections in the medial septum impair spatial learning. Brain Research, 522, 227–234. Brioni, J. D., Decker, M. W., Sullivan, J. P., & Arneric, S. P. (1996). The pharmacology of (0)nicotine and novel cholinergic channel activators. Advances in Pharmacology, 37, 153–214. Brioni, J. D., Morgan, S. J., O’Neill, A. B., Sykora, T. M., Postl, S. P., Pan, J. B., Sullivan, J. P., & Arneric, S. P. (1996). ‘‘In vivo’’ profile of novel nicotinic ligands with CNS selectivity. Medicinal Chemistry Research, 6, 487–510. Chen, Y., Ogren, S., Bjelke, B., Bolme, P., Eneroth, P., Gross, J., Loidl, F., Herrera-Marschitz, M., & Anderson, K. (1995). Nicotine treatment counteracts perinatal asphyxia-induced changes in the mesostriatal-limbic dopamine systems and in motor behaviour in the four-week-old male rat. Neuroscience, 68, 531–538. Decker, M. W., Brioni, J. D., Sullivan, J. P., Buckley, M. J., Radek, R. J., Raskiewickz, J. L., Kang, C. H., Kim, D. J. B., Giardina, W. J., Williams, M., & Arneric, S. P. (1994). ABT 418: A novel cholinergic ligand with cognition enhancing and anxiolytic activities. II. In vivo characterization. Journal of Pharmacology and Experimental Therapeutics, 270, 319–328. Donnelly-Roberts, D., Xue, I., Arneric, S., & Sullivan, J. (1996). In vitro neuroprotective properties of a novel cholinergic channel activator (ChCA), ABT-418. Brain Research, 719, 36–44. Foster, A. C., & Woodruff, G. N. (1988). Neuroprotective effects of MK-801 ‘‘in vivo’’: Selectivity and evidence for delayed degeneration mediated by NMDA receptor activation. Journal of Neuroscience, 8, 4745–4754. Fuxe, K., Rosen, L., Lippoldt, A., Andbjer, B., Hasselrot, U., Finman, U., & Agnati, L. (1994). Chronic continuous infusion of nicotine increases the dissapearance of choline acetyltransferase immunoreactivity in the cholinergic cell bodies of the medial septal nucleus following a partial unilateral transection of the fimbria fornix. Clinical Investigations, 72, 262–268. Heyes, M. P., Saito, K., Major, E. O., Milstien, S., Markey, S. P., & Vickers, J. H. (1993). A mechanism of quinolinic acid formation by brain in inflammatory neurological disease. Brain, 116, 1425–1450. Janson, A., Fuxe, K., Agnati, L., Katayama, I., Harfstrand, A., Anderson, K., & Goldstein, M. (1988). Chronic nicotine treatment counteracts the disappearance of tyrosine-hydroxylaseinmunoreactive nerve cell bodies, dendrites and terminals in the mesostriatal dopamine system of the male rat after partial hemitransection. Brain Research, 455, 332–345. Kodsi, M. H., & Swerdlow, N. R. (1994). Quinolinic acid lesions of the ventral striatum reduce sensorimotor gating of acoustic startle in rats. Brain Research, 643, 59–65. Levin, E. D. (1992). Nicotinic systems and cognitive function. Psychopharmacology, 108, 417–431. Lipton, S. A., & Rosenberg, P. A. (1994). Excitatory amino acids as a final common pathway for neurological disorders. New England Journal of Medicine, 330, 613–622. Marin, P., Maus, M., Desagher, S., Glowinski, J., & Premont, J. (1994). Nicotine protects cultured striatal neurons against N-methyl-D-aspartate receptor-mediated neurotoxicity. NeuroReport, 5, 1977–1980. Martin, E., Panickar, K., King, M., Deyrup, M., Hunter, B., Wang, G., & Meyer, E. (1994). Cytoprotective actions of 2,4-dimethoxybenzylidene anabaseine in differentiated PC12 cells and septal cholinergic neurons. Drug Development Research, 31, 135–141.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
64
O’NEILL, MORGAN, AND BRIONI
Morris, R. (1984). Developments of a water-maze procedure for studying spatial learning in the rat. Journal of Neuroscience Methods, 11, 47–60. O’Neill, A. B., Morgan, S. J., Sykora, T., Postl, S., & Brioni, J. D. (1996). Hippocampal neurodegeneration induced by quinolinic acid: Histological and behavioral protection by nicotine. Society for Neuroscience Abstracts, 22, 834.9. Owman, C., Fuxe, K., Jason, A., & Kahrstrom, J. (1989). Studies of protective actions of nicotine on neuronal and vascular functions in rats: Comparison between sympathetic noradrenergic and mesostriatal dopaminergic fiber system, and the effect of a dopamine agonist. Progress in Brain Research, 79, 267–276. Perkins, M. N., & Stone, T. W. (1983). Quinolinic acid: Regional variations in neuronal sensitivity. Brain Research, 259, 172–176. Reinhard, J. F., Erickson, J. B., & Flanagan, E. M. (1994). Quinolinic acid in neurological disease: Opportunities for novel drug discovery. Advances in Pharmacology, 30, 85–127. Schwarcz, R., & Kohler, C. (1983). Differential vulnerability of central neurons of the rat to quinolinic acid. Neuroscience Letters, 38, 85–90. Schwarcz, R., Whetsell, W. O., & Mangano, R. M. (1982). Quinolinic acid: an endogenous metabolite that produces axon-sparing lesions in rat brain. Science, 219, 316–318. Sjak-Shie, N., & Meyer, E. (1993). Effects of chronic nicotine and pilocarpine administration on neocortical neuronal density and [3H]GABA uptake in nucleus basalis lesioned rats. Brain Research, 624, 295–298. Stone, T. W. (1993). Neuropharmacology of quinolinic and kynurenic acid. Pharmacological Reviews, 45, 309–379. Stone, W., & Perkins, M. N. (1981). Quinolinic acid: A potent endogenous excitant at aminoacid receptors in CNS. European Journal of Pharmacology, 72, 411–412. Vecsei, L., & Beal, M. F. (1991). Comparative behavioral and neurochemical studies with striatal kainic acid or quinolinic acid lesioned rats. Pharmacology Biochemistry and Behavior, 39, 473–478. Winn, P., Stone, T. W., Latimer, M., Hastings, M. H., & Clark, A. J. M. (1991). A comparison of excitotoxic lesions of the basal forebrain by kainate, quinolinate, ibotenate, N-methyl-Daspartate or quisqualate, and the effects on toxicity of 2-amino-5-phophonovaleric acid and kynurenic acid in the rat. British Journal of Pharmacology, 102, 904–908. Wirsching, B. A., Beninger, R. J., Jhamandas, K., Boegman, R. J., & Bialik, M. (1989). Kynurenic acid protects against the neurochemical and behavioral effects of unilateral quinolinic acid injections into the nucleus basalis of rats. Behavioral Neuroscience, 103, 90–97.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
69, 46–64 (1998)
NL973803
Histological and Behavioral Protection by (0)-Nicotine against Quinolinic Acid-Induced Neurodegeneration in the Hippocampus Alyssa B. O’Neill,* Sherry J. Morgan,† and Jorge D. Brioni*,1 *Neurological and Urological Diseases Research and †Pathology Department, Pharmaceutical Products Division, Abbott Laboratories, Abbott Park, Illinois 60064-3500
Injections of quinolinic acid (60, 180, and 600 nmol) in the dorsal hippocampus induced significant neurotoxicity that was evident 1 day after the injection. By day 3, pyramidal as well as granular cells were affected even at the lowest dose of quinolinic acid, an effect that persisted up to 20 days. Consistent with the histological findings, animals with bilateral injections in the dorsal hippocampus were cognitively impaired during acquisition and retention of spatial information in the water maze. A subacute treatment with (0)-nicotine (62 mmol/kg/day) delivered by subcutaneous minipumps prevented the histological and cognitive deficits induced by the bilateral quinolinic acid (60 nmol) injections. These data indicate that quinolinic acid can induce degeneration of both pyramidal as well as granule cells in the hippocampus, leading to cognitive impairments in the rat, and that activation of neuronal nicotinic acetylcholine receptors can prevent the neurodegenerative process induced by quinolinic acid. q 1998 Academic Press Key Words: hippocampus; learning; memory; nicotine; quinolinic acid.
Nicotinic acetylcholine receptor (nAChR) agonists can induce several pharmacological effects in rodents including hormone release, analgesia, memory facilitation and increased attention, by virtue of their specific binding to selective nAChR subtypes in the brain (Arneric, Sullivan, & Williams, 1995; Brioni, Decker, Sullivan, & Arneric, 1996; Levin, 1992). A relatively novel finding has been the demonstration that activation of neuronal nAChRs can induce neuroprotection in different models of neuronal injury. In vitro studies have revealed that (0)-nicotine protected striatal and cortical neurons against NMDA-induced neurotoxicity but not against kainic-induced toxicity (Akaike, Tamura, Yokaota, Shimohama, & Kimura, 1994; Marin, Maus, Desagher, Glowinski, & Premont, 1994). This neuroprotective effect of (0)-nicotine is mimicked by ABT-418 and GTS-21, two novel cholinergic channel activators; ABT-418 reversed glutamate-induced toxicity in primary cultures of rat cortical neurons and human neuroblastoma IMR32 cells (DonnellyRoberts, Xue, Arneric, & Sullivan, 1996), and GTS-21 prevented PC12 cell death after nerve growth factor removal (Martin, Panickar, King, Deyrup, Hunter, Wang, & Meyer, 1994). Similarly, in vivo studies have shown that (0)-nicotine reduced the loss of tyrosine-hydroxylase immunoreactive neurons after partial hemitransection of the mesostriatal dopaminergic pathway (Jan1
We thank Ms. Theresa Sykora and Mr. Steven Postl for technical support. Address reprint requests to Dr. Jorge D. Brioni, Neurological and Urological Diseases Research, Building AP10LL (47C), Abbott Laboratories, Abbott Park, IL 60064-3500. Fax: (847) 937-9195. 1074-7427/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
NLM 3803
/
46
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
47
son, Fuxe, Agnati, Katayama, Harfstrand, Anderson, & Goldstein, 1988; Owman, Fuxe, Jason, & Kohrstrom, 1989) and reduced cortical neuronal loss after ibotenic acid injections in the nucleus basalis magnocellularis in rats (SjakShie & Meyer, 1993). Quinolinic acid (pyridine-2,3-dicarboxylic acid) is an endogenous substance in the mammalian brain that can be synthesized from tryptophan by several enzymes involved in the kynurenine pathway (Reinhard, Erickson, & Flanagan, 1994; Stone, 1993). Although it has been described as an intermediate in the synthesis of nicotinamide adenine dinucleotide cofactors, current interest has focused on its neurotoxic properties and its eventual participation in several neurological disorders like Huntington’s disease and AIDS-dementia (Heyes, Saito, Major, Milstien, Markey, & Vickers, 1993; Lipton & Rosenberg, 1994; Schwarcz, Whetsell, & Mangano, 1982). Electrophysiological studies have revealed that quinolinic acid is as potent as glutamate or aspartate to stimulate rat cortical neurons, an effect blocked by the N-methyl-D-aspartate (NMDA) antagonist, 2 APV (Stone & Perkins, 1981). Injections of quinolinic acid (120 nmol) in several brain areas induced different degrees of neurodegeneration that might reflect the differential sensitivity of mammalian neurons to its toxic actions (Schwarcz & Kohler, 1983); the striatum, hippocampus, and globus pallidus were more sensitive to the toxic effect in comparison to the cerebellum, amygdala, septum and hypothalamus, with a selective vulnerability of pyramidal cells in comparison to granular cells in the hippocampus (Schwarcz & Kohler, 1983). In the present study we investigated the time course and potency of the neurotoxic actions of quinolinic acid after injections in the dorsal hippocampus and attempted to correlate the histological changes with the performance of rats in a spatial memory test that is very sensitive to hippocampal function. As the activation of neuronal nAChRs has been shown to provide neuroprotection in several models (Brioni, Morgan, O’Neill, Sykora, Postl, Pan, Sullivan, & Arneric, 1996), the ability of (0)-nicotine to reverse the histological and cognitive effects of quinolinic acid injections in the dorsal hippocampus was later investigated. MATERIALS AND METHODS Animal studies were conducted according to the guidelines of the American Association for the Accreditation of Laboratory Animal Care (AAALAC) and according to protocols approved by the Abbott Institutional Animal Care and Use Committee. Animals Male Long Evans rats (weighing 200 g on arrival) supplied by Harlan (Altamont, NY) were used. They were individually housed and acclimated to laboratory conditions for 1 week before the studies. Animals were kept in a climatecontrolled facility with a 12-h light/dark cycle. Testing was conducted during the light portion of the day. Surgery Intracerebral injections of quinolinic acid (18, 60, 180, and 600 nmol in 0.5 ml) were performed in animals anesthetized with 50 mg/kg pentobarbital
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
48
O’NEILL, MORGAN, AND BRIONI
(Nembutal, Abbott Laboratories). Injections were placed in the dorsal hippocampus under stereotaxic control (AP Å 03.3; ML, {2.5; DV, 03.6 from dura) and were made slowly through an injecting cannula attached by polyethylene tubing to a 10-ml syringe and a motor-driven minipump. A total volume of 0.5 ml was injected in a 4-min period and the cannula was left in place another 2 min to allow diffusion of the drug. When bilateral injections were needed for the behavioral studies, they were performed simultaneously. Animals were allowed to recover on a warming plate before returning them to the cage. Spatial Learning in the Water Maze Apparatus. A cylindrical water tank (60 cm high and 180 cm in diameter) was filled to a depth of 37 cm with 26 { 17C water rendered opaque by the addition of dry milk. Close to the rim of the tank along 1.50 m of its circumference, a white curtain that extended to the ceiling of the room served as a salient cue. Other cues were also available in the room. Four points equally spaced around the perimeter of the tank were arbitrarily designated as starting locations. On this basis, the tank was divided into four equal quadrants. Located in the center of one of the quadrants was a 13-cm wide circular platform. Cue training procedure. During this training version in the water maze, a black platform protruded 1 cm over the water surface and was visible to the animals. A trial was initiated by placing the rat in the water at one of the four starting locations along the perimeter of the pool, which varied from trial to trial in a quasirandom order (Brioni, Decker, Gamboa, Izquierdo, & McGaugh, 1990). The rats were allowed to swim until they located the escape platform and climbed onto it. If the rats did not locate the escape platform within 90 s, they were gently guided to it. The rats remained in the platform for 20 s before being removed. The following trial was conducted 5–10 s later. Animals received a total of eight trials in two sessions (the second session was conducted 4 h after the first one). Place training procedure. Three days after cue training was completed, animals were submitted to place training in the water maze. The platform remained in a fixed location throughout this experiment in order to evaluate reference memory in the rats (Morris, 1984). Animals received six trials per day, in two sets of 3 trials separated by an interval of 4 h. The rats received a total of 18 trials during 3 days of place training. Retention of the spatial information was assessed in a single probe trial conducted 24 h after the last place training session in the absence of the escape platform (free-swim trial). Animals were videotaped and data analyzed by the Poly-Track tracking software (San Diego Instruments, CA). Continuous Subcutaneous Treatment with (0)-Nicotine For the subacute treatment with (0)-nicotine, rats were implanted with Alza minipumps (Alzet, Model 2ML4, Alza Corporation, Palo Alto, CA) containing (0)-nicotine and delivering approximately 19 and 62 mmol/kg/day of (0)-nicotine. In the control group the minipumps delivered sterile physiological saline solution. Minipumps were surgically implanted subcutaneously. Surgery was conducted under sterile conditions using halothane anesthesia (4% in the induction period and 2% for maintenance). An incision was made in the back of
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
49
the animals and a pump was inserted subcutaneously. The wound was closed with wound-clips, and the animal was allow to recover on a heating pad before being returned to its home cage. Animals recovered from anesthesia within 5 min. Animals received bilateral injections of quinolinic acid in the dorsal hippocampus 7 days after implantation of the minipumps. Place training in the water maze started 7 days later and animals were sacrificed after the freeswim trial (10 days after the quinolinic injections). Histological Procedures Rats were deeply anesthetized with 100 mg/kg i.p. pentobarbital prior to being perfused transcardially with saline solution followed by 10% phosphatebuffered formalin. The brains were then removed and immersion-fixed for 24 h in the same fixative. The brains were placed in a tissue slicer with 3-mm gradations, and after wet-cutting, the brain slices were processed and embedded in paraffin. Hippocampal sections (6 mm) were cut coronally, mounted, and stained with hematoxilin and eosin. A detailed evaluation of histologic changes were performed on each specimen by an experimenter (S.J.M.) unaware of the treatment conditions. In addition, cell counts of viable neurons were performed on the specimens from the experiment that compared the effect of (0)-nicotine on quinolinic-induced lesions. The mean of four matched coronal sections from different rostrocaudal levels of the dorsal hippocampus of each subject was measured along a 500-mm length by means of an eye piece graticule. For the CA1–CA4 regions, the counts were made with a 201 objective; for the dorsal and ventral dentate, the counts were made with a 401 objective. Cell counts were expressed as the percentage of the number of viable cells in comparison to the control group. Drugs Quinolinic acid (Sigma Chemical Co., St. Louis, MO) was prepared in physiological saline solution and adjusted to pH 5.8 with 0.5 N NaOH. (0)-Nicotine bitartrate (Sigma) was dissolved in saline solution. Fresh solutions were prepared every day. Statistics Data were analyzed by one- or two-way analyses of variance (ANOVA) followed by the Fisher Protected Least Significance Difference test for individual means comparisons. RESULTS Effect of Quinolinic Acid Injections in the Hippocampus A group of rats was injected unilaterally with saline or quinolinic acid (60, 180, and 600 nmol) and evaluated histologically 1 and 3 days after the injection. Histologic examination of the hippocampus of rats necropsied 1 day postinjec-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
50
O’NEILL, MORGAN, AND BRIONI
FIG. 1. Neurotoxic effect of quinolinic acid 1 day after a unilateral injection in the dorsal hippocampus. (A) Control; (B) 60 nmol; (C) 180 nmol; (D) 600 nmol (120). Bar Å 200 mm.
tion revealed a dose-related neuronal loss (Fig. 1). Those rats that received 60 nmol of quinolinic acid revealed a unilateral extensive loss of hippocampal neurons. The most extensive loss was noted in the pyramidal regions, whereas there was some degree of neuronal retention in the dorsal dentate and, to a
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
51
FIG. 1—Continued
lesser extent, the ventral dentate (Fig. 1B). Cells in the CA4, CA3, CA2, and CA1 pyramidal region which were interpreted as necrotic were typical of early stage ischemic neurons, being characterized by the presence of an intensely eosinophilic, somewhat spindle shaped cytoplasm and an oval to round nucleus with variable retention of detail. In contrast, cells in the dentate region which
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
52
O’NEILL, MORGAN, AND BRIONI
FIG. 2. Neurotoxic effect of quinolinic acid 3 days after a unilateral injection in the dorsal hippocampus. (A) 60 nmol; (B) 180 nmol; (C) 600 nmol (120). Bar Å 200 mm.
were interpreted as necrotic had an imperceptible cytoplasm and a small, dense, round nucleus with no retention of detail. Other changes noted in these rats included moderate spongiosis (primarily perineuronal). Those rats that received 180 nmol of quinolinic acid (Fig. 1C) exhibited a unilateral complete loss of hippocampal neurons with some loss in the contralateral CA1 region as well. The morphology of the necrotic cells was similar to that as described for the rats that received 60 nmol. Perineuronal spongiosis was present although more severe than that noted in the 60-nmol group. The rats that received 600 nmol of quinolinic acid exhibited lesions which were similar, but more severe, than those that received 180 nmol (Fig. 1D). Histological changes noted in those rats necropsied 3 days postinjection revealed considerable progression of the lesions in all dosage groups (Fig. 2). Those rats that received 60 nmol of quinolinic acid showed a nearly complete loss of neurons; neuronal sparing, when present, was generally confined to the dorsal blade of the dentate gyrus. Morphologic changes in necrotic neurons were similar to those noted at 1 day postinjection with the exception of loss of nuclear detail in the ischemic neurons (CA4, CA3, CA2, and CA1). Hippocampal perineuronal spongiosis, as noted at 1 day postinjection was still present (Fig. 2A). Rats that received 180 or 600 nmol of quinolinic acid exhibited lesions that were similar, but more severe, than those noted at 1 day postinjection (Figs. 2B and 2C). Effect of Quinolinic Injections in the Hippocampus on Spatial Learning In order to determine the effect of quinolinic acid on cognitive performance, another group of rats received bilateral injections of quinolinic acid (18, 60,
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
53
FIG. 2—Continued
and 180 nmoles) and 7 days later were submitted to the water maze test. This dose range was selected based on the previous histological data after unilateral injections of quinolinic acid that showed severe lesions in the 600 nmol group. During cue training in the water maze, there were no differences in the latencies to reach the escape platform between the sham and the quinolinic
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
54
O’NEILL, MORGAN, AND BRIONI
FIG. 3. Effect of bilateral injections of quinolinic acid in the dorsal hippocampus on place learning in the water maze. Data represent the mean escape latencies of nine rats in each group. Training began 7 days after the lesions. In the place version of the maze, animals were submitted to two daily sessions and received three trials per session. **p õ .01; *** p õ .001 as compared to sham rats.
18- or 60-nmol groups in the last four trials; however, the quinolinic 180-nmol group exhibited significantly longer latencies to locate the visible platform (Control, 27.8 { 4.6 s; Quin 18, 21.2 { 5.3 s; Quin 60, 40.2 { 5.2 s; Quin 180, 53.7 { 3.7 s; p õ .05 as compared to control rats). Figure 3 shows the performance of the rats during place training in the water maze. Under our experimental conditions, by the third day sham rats were able to learn the test and to escape into the platform in approximately 10 s. Animals injected with quinolinic acid exhibited a significant impairment in the water maze. There was a significant quinolinic effect [F(3, 32) Å 32.3, p õ 0.001], a significant session effect [F(5, 160) Å 12.8, p õ .001], and a significant quinolinic 1 session interaction [F(15, 160) Å 2.3, p õ .01]. Animals that received 18 nmol performed as well as the sham group, while the groups receiving 60 and 180 nmol were significantly impaired in the acquisition of spatial information. Table 1A shows the performance of the rats during the free-swim trial conducted 24 h after the last training session. The original position of the platform was video-taped and the platform was removed from the tank. Quinolinicinjected animals exhibited significant longer latencies in comparison to the sham group [F(3, 32) Å 12.0, p õ .001] and a reduced number of annulus crossings [F(3, 32) Å 8.1, p õ .001], that reflect an impairment on retention of spatial information. Histological evaluation of the hippocampus conducted 20 days after the injection of quinolinic acid revealed a dose-related neuronal loss (Fig. 4). In those rats that received 18 nmol of quinolinic acid, there was a substantial bilateral neuronal necrosis (up to 90%) only in the CA3 and CA4 regions, as it occurred
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
55
TABLE 1 Performance in the Free-Swim Trial Conducted 24 h after Training in the Water Maze Latency (s)
Crossings
A Sham Quinolinic 18 mmol Quinolinic 60 mmol Quinolinic 180 mmol
10.4 10.0 25.7 26.6
{ { { {
1.8 3.1 3.0*** 2.7***
1.6 1.8 0.2 0.2
{ { { {
0.3 0.5 0.1** 0.1**
{ { { / { {
3.3 3.2 4.4 4.7* 2.4** 3.2
1.9 1.2 1.2 0.5 0.4 1.1
{ { { { { {
0.3 0.2 0.4 0.3*** 0.2*** 0.2†
B Sham NIC 19 mmol/kg/day NIC 62 mmol/kg/day Quinolinic 60 mmol Quin 60 / NIC 19 Quin 60 / NIC 62
9.1 10.1 15.1 20.3 26.3 14.8
Note. Data represent the mean { SEM latency to reach the platform location and annulus crossings during the 30-s free swim. * p õ .05, **p õ .01, ***p õ .001 as compared to sham rats. † p õ .05 as compared to the Quin 60 group.
to a lesser extent (10–20%) in the CA2, CA1, and ventral dentate regions (Fig. 4A). Occasional cells (õ5%) in the dorsal dentate were necrotic. Other changes noted primarily in the CA4 region included a marked infiltrate of gemistocytic astrocytes, moderate microgliosis, and moderate mineralization, involving both neurons and the neuropil. On the contrary, changes noted in rats that received 60 nmol quinolinic acid included near total (ú95%) loss of neurons with no evidence of preferential retention (Fig. 4B). Other hippocampal changes such as mineralization, microgliosis, gemistocytic astrocyte infiltration, and neovascularization were more severe than those seen in the 18-nmol group. A change noted in this group but not in the 18-nmol group included the presence of hippocampal parenchymal collapse. In rats that received 180 nmol quinolinic acid there was a total loss of neurons (Fig. 4C). Other hippocampal changes as noted in the 60-nmol group were apparent, but were more severe; for example, the hippocampal parenchymal collapse was so severe that orientation and identification of structures was difficult. The presence of cyst-like foci containing lipid-laden macrophages in the CA4 region was noticed in this group. Thus, although the extent of neuronal loss was similar to that noted at 3 days postinjection, evaluation at this time point revealed progression of the parenchymal changes (cellular infiltration and parenchymal collapse). Effect of a Subacute Treatment with (0)-Nicotine on Quinolinic-Induced Histological and Behavioral Changes In order to determine the neuroprotective effect of (0)-nicotine in quinolinicinjected rats, three groups of rats were implanted with chronic subcutaneous minipumps delivering physiological saline, nicotine 19 or 62 mmol/kg/day. Seven days after implantation of the minipumps, bilateral injections of quino-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
56
O’NEILL, MORGAN, AND BRIONI
FIG. 4. Morphological changes induced by of quinolinic acid 20 days after bilateral injections in the dorsal hippocampus. (A) 18 nmol; (B) 60 nmol; (C) 180 nmol (120). Bar Å 200 mm.
linic acid (60 nmol) or saline were carried out in the dorsal hippocampus. When animals were submitted to the cue version of the water maze there were no differences between the groups as they were all able to locate the escape platform (data not shown). Figure 5 shows the performance of the rats in the place version of the test conducted 7 days after the quinolinic acid injection. There was a significant effect of the drug treatment [F(5, 42) Å 12.2, p õ .001], a significant session effect [F(5, 210) Å 38.6, p õ .001], and a significant drug 1 session interaction [F(25, 210) Å 2.8, p õ .001]. An independent analysis of the nonlesioned animals that received saline or the (0)-nicotine doses revealed that the drug 1 session interaction was close to significance [F(5, 105) Å 1.83, p Å .064], as (0)-nicotine tended to improve the performance of sham animals. Sham rats were able to locate the escape platform in approximately 10 s by the last session, while the rats injected with quinolinic acid (60 nmol) were significantly impaired throughout all sessions. However, animals that received the subacute treatment with (0)-nicotine 62 mmol/kg/day performed significantly better during the last three sessions. In the free-swim trial conducted 24 h later (Table 1B), there was a significant effect on the escape latencies [F(5, 41) Å 3.1, p õ .05] and on the number of platform crossings [F(5, 41) Å 4.5, p õ .01] as quinolinic-injected rats performed significantly worse than the control group, with the exception of those rats treated with 62 mmol/kg/day (0)-nicotine. Rats that received (0)-nicotine alone (19 or 62 mg/kg/day) exhibited lesions indistinguishable from those of control rats (Fig. 6A). Changes common to these groups included the presence of injection tracts (linear gliosis, hemosiderosis, and occasional neovascularization) along with focal loss of neurons. Blood samples were collected from the animals receiving the subacute (0)-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
57
FIG. 4—Continued
nicotine treatment and the plasma levels were determined by HPLC/ED as previously described (Decker, Brioni, Sullivan, Buckley, Radek, Raskiewickz, Kang, Kim, Giardina, Williams, & Arneric, 1994). Plasma (0)-nicotine levels for the animals receiving 19 or 62 mmol/kg/day averaged 36.2 { 9.6 and 126.8 { 41.9 ng/ml, respectively.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
58
O’NEILL, MORGAN, AND BRIONI
FIG. 5. Reversal of quinolinic-induced cognitive impairments in the water maze by a subacute treatment with (0)-nicotine. Data represent the mean escape latencies of eight rats in each group. Place training began 7 days after the lesions. Animals were submitted to two daily sessions and received three trials per session. *p õ .05; **p õ .01; ***p õ .001 as compared to sham rats. // p õ .01; /// p õ .001 as compared to the quinolinic 60-nmol group.
Histologic examination of the hippocampus 10 days after 60-nmol quinolinic acid injections (Fig. 6B) was similar to the 60-nmol rats that were necropsied at 20 days (see above). The primary difference between these rats and those necropsied at 20 days was the presence of cyst-like foci in the CA4 region in those necropsied at this time point and hippocampal parenchymal collapse in those necropsied at 20 days. The hippocampal parenchymal collapse is likely a sequelae for the cystic-like structures in the CA4 region. Other changes (mineralization, microgliosis, gemistocytic astrocyte infiltration, neovascularization) present at 20 days were also present at this time point but were less pronounced. Quantification of viable hippocampal neurons was performed on specimens from each group necropsied at 10 days (Table 2), and the neuronal counts for all hippocampal regions were significantly lower in quinolinic-injected rats than in saline-injected rats [F(1, 4) Å 2,618, p õ .0001]. Rats that received 60 nmol quinolinic acid along with 19 mmol/kg/day (0)-nicotine exhibited morphological changes that were nearly identical to that noted in the rats that received 60 nmol quinolinic acid alone, and the neuron counts in this group were similar to those in the rats that received quinolinic acid alone. Rats that received 60 nmol quinolinic acid along with 62 mmol/kg/day (0)-nicotine exhibited a definite sparing of the neuronal loss (Table 2; Fig. 6C). With the exception of the ventral dentate region, all regions of the hippocampus had significantly higher neuron counts in this treatment group than seen in the rats that received quinolinic acid alone (all F ’s p õ .001).
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
59
DISCUSSION These data demonstrate that quinolinic acid can induce potent and long lasting neurotoxic actions on pyramidal as well as granular cells in the rat dorsal hippocampus. The hippocampal cell loss was associated with the performance of rats in a spatial memory test as lesioned rats were significantly impaired during acquisition and retention of spatial information in the water maze. However, a subacute treatment with (0)-nicotine delivered via subcutaneously implanted minipumps prevented the neurotoxic and cognitive-impairing effects of quinolinic acid. Quinolinic acid is an endogenous substance in the mammalian brain synthesized from tryptophan by several enzymes involved in the kynurenine pathway (Reinhard et al., 1994; Stone, 1993) that can exert neurotoxic actions in the striatum and in the hippocampus (Schwarcz et al., 1982). Further studies have shown regional variations in the sensitivity to quinolinic acid (120 nmol) as the striatum, hippocampus, and cortex were more sensitive than the amygdala, septum, or hypothalamus, whereas negligible effects were noted in the cerebellum (Perkins & Stone, 1983; Schwarcz & Kohler, 1983). When injected in the basal forebrain, quinolinic acid was less potent than kainic acid but more potent than quisqualate and ibotenic acid (Winn, Stone, Latimer, Hastings, & Clark, 1991). Consistent with previous findings with other neurotoxic agents like kainic acid or quisqualic acid, injections of quinolinic acid in selective brain areas induced marked behavioral dysfunction in rats. Injections of quinolinic acid in the striatum have been shown to reduce sensorimotor gating (Kodsi & Swerdlow, 1994), impair active avoidance learning (Vecsei & Beal, 1991), and impair the performance of rats in spatial memory tests (Block, Kunkel, & Schwarz, 1993). Injections in the nucleus basalis magnocellularis impaired T-maze alternation and impaired working memory in the radial arm-maze consistent with a significant decrease in cortical choline acetyltransferase activity in lesioned rats (Beninger, Wirsching, Jhamandas, Boegman, & ElDefrawy, 1986; Wirsching, Beninger, Jhamandas, Boegman, & Bialik, 1989). With regard to the hippocampus, neuronal loss after quinolinic injections have been observed 4 days (Schwarcz & Kohler, 1983; Schwarcz et al., 1982) and 7 days after the injections (Foster & Woodruff, 1988). The present data demonstrate that quinolinic-induced cell death can be detected as early as 1 day, with pyramidal cells being more sensitive than granular cells, as indicated by the sparing of the ventral granular layer at this time point. The lesion progressed as a complete loss of pyramidal and granular neurons was observed 3 days after the injection. Although it was reported that quinolinic injections in the hippocampus induced no behavioral effects in rats despite the presence of an extensive histological damage (Schwarcz & Kohler, 1983), a significant impairment during acquisition of spatial information was noted in our experiments in those groups receiving bilateral injections of 60 and 180 nmol. As the 60-nmol group did not show any impairment in the cue version of the test, these data indicate that the behavioral effects are not due to sensorimotor impairment in the rats but due to a specific place learning deficit. The histological evaluation conducted 20 days after the injections showed that the morphological characteristic observed 1 and 3 days after the injections were still present or even exacerbated, indicating the permanent nature of the lesion induced by quinolinic acid.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
60
O’NEILL, MORGAN, AND BRIONI
FIG. 6. Neuroprotective actions of (0)-nicotine against quinolinic acid-induced toxicity. (A) Control; (B) quinolinic acid 60 nmol; (C) quinolinic acid 60 nmol / 62 mmol/kg/day (0)-nicotine (120). Bar Å 200 mm.
The search for pharmacological agents to counteract the ongoing neurodegenerative process present in Parkinson’s disease, Alzheimer’s disease, Huntington’s disease, amyotrophic lateral sclerosis, AIDS-dementia, epilepsy, or ischemia is a major scientific challenge at the present time. The list of potential targets include several classes of drugs such as NMDA receptors antagonists, GABAergic agonists, kynurenine analogs, antioxidants, adenosine agonists, growth factors, and Ca2/ channel antagonists. The in vivo and in vitro neuroprotective effects of (0)-nicotine in several models of neurodegeneration suggest that the neuronal nAChR could represent a novel therapeutic target (Brioni et al., 1996). Subchronic administration of (0)-nicotine reduced the loss of tyroxine-hydroxylase immunoreactive neurons after partial hemitransection of the mesostriatal dopaminergic pathway and attenuated the histological and behavioral effect of perinatal asphyxia on central dopaminergic pathways (Chen, Ogren, Bjelke, Bolme, Eneroth, Gross, Loidl, Herrera-Marschitz, & Anderson, 1995; Janson, Fuxe, Agnati, Katayama, Harfstrand, Anderson, & Goldstein, 1988; Owman et al., 1989). However, the neuroprotective actions of (0)-nicotine are not universal, as in rats with fimbria–fornix transection (0)-nicotine did not prevent but actually increased the disappearance of septal neurons (Fuxe, Rosen, Lippoldt, Andbjer, Hasselrot, Finman, & Agnati, 1994). Several studies in cultured cells have corroborated the neuroprotective effect of (0)-nicotine in rats. (0)-Nicotine protected against NMDA-induced neurotoxicity in striatal and cortical neurons, while the novel cholinergic channel modulator, ABT-418, reversed glutamate-induced toxicity in rat cortical neurons and human neuroblastoma IMR32 cells (Akaike et al., 1994; Donnelly-Roberts et al., 1996; Marin et al., 1994). Another novel
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
61
FIG. 6—Continued
cholinergic channel modulator with neuroprotective properties is GTS-21 as this compound prevented PC12 cell death after removal of nerve growth factor from the medium (Martin et al., 1994). In order to determine the potential neuroprotective effect of (0)-nicotine against quinolinic-induced lesions, a group of rats were implanted with subcu-
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
62
O’NEILL, MORGAN, AND BRIONI
TABLE 2 Quantitative Analysis of the Effect of (0)-Nicotine on Quinolinic-Induced Neurodegeneration in the Hippocampus Pyramidal layer Treatment
CA1
CA2
Saline / saline 100.0 NIC 19 / saline 100.5 { 3.3 NIC 62 / saline 94.8 { 1.5 Saline / Quin 60 13.1 { .6*** NIC 19 / Quin 60 15.2 { 3.9*** NIC 62 / Quin 60 37.4 { 10.8***,†
Dentate gyrus
CA3
100.0 100.0 89.7 { 13.2 110.6 { 3.5 84.2 { 6.6 99.8 { 2.1 9.7 { 1.8*** 11.9 { 1.7*** 10.2 { 1.5*** 12.3 { 2.6*** 48.6 { 18.7*,† 31.1 { 7.6***,†
CA4
99.1 96.2 2.2 6.7 21.8
Dorsal
100.0 { 3.9 { 2.3 { .8*** { 4.7*** { 5.3***,††
95.0 98.1 10.8 12.9 46.1
100.0 { 3.4 { 1.5 { 2.3*** { 4.7*** { 15.7**,†
Ventral 100.0 92.7 { 3.7 93.0 { 3.5 4.2 { 1.4*** 15.3 { 9.9*** 21.5 { 7.1***
Note. Number of viable neurons for each hippocampal subfield expressed as a percentage of the saline–saline group. The mean of four matched coronal sections from the dorsal hippocampus of each subject were averaged across the treatment group (n Å 2 in each saline-injected group and n Å 4 in each quinolinic-injected group). * p õ .05, **p õ .01, ***p õ .001 as compared to saline-injected rats. † p õ .05, ††p õ .01 as compared to quinolinic-injected rats.
taneous minipumps delivering 19 or 62 mmol/kg/day (0)-nicotine and 1 week later injected with quinolinic acid in the dorsal hippocampus. Behavioral studies in the water maze indicated that the chronic treatment with (0)-nicotine tended to facilitate acquisition of spatial information in nonlesioned rats, consistent with previous reports in the field (Levin, 1992). Quinolinic injected rats receiving chronic vehicle solution or 19 mmol/kg/day (0)-nicotine were impaired throughout the sessions, but the group receiving 62 mmol/kg/day (0)-nicotine exhibited a significant improvement in the last three sessions in comparison to the lesioned group. The histological analysis in those animals was consistent with the behavioral findings as neuroprotection was observed in those animals receiving the larger dose of (0)-nicotine. Preliminary data from our laboratory indicate that 10 to 20% of the neurons undergoing degeneration exhibit apoptosis after the 60-nmol injection of quinolinic acid (O’Neill, Morgan, Sykora, Postl, & Brioni, 1996). Whether the neuroprotective actions of (0)-nicotine can preferentially block apoptosis or necrotic type mechanisms of neuronal death is unclear at the present time but is a subject that merits further investigation. It is unknown which nAChR subunits mediate the neuroprotective actions of (0)-nicotine, although the ability of methyllycaconitine and a-bungarotoxin to block (0)-nicotine’s protection against glutamate-induced toxicity might indicate the participation of the a7 nAChR subunit (Donnelly-Roberts et al., 1996). Future work aiming to identify the specific nAChR subunit/s involved in the neuroprotective effect of nAChR agonists will allow the identification of novel ligands with greater potency and efficacy. As quinolinic acid is an endogenous substance that can exert neurotoxic effects, its eventual participation in several neurological and neuroinflammatory diseases like Huntington’s disease and AIDS-dementia has been proposed (Heyes et al., 1993; Lipton & Rosenberg, 1994). The neuroprotective actions of nAChR ligands indicate that novel nAChR ligands with neuroprotective efficacy might be useful for the treatment of these disorders associated with increased quinolinic acid levels in the brain.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
HIPPOCAMPAL NEURODEGENERATION AND PROTECTION BY (0)-NICOTINE
63
REFERENCES Akaike, A., Tamura, Y., Yokaota, T., Shimohama, S., & Kimura, J. (1994). Nicotine-induced protection of cultured cortical neurons against N-methyl-D-aspartate receptor-mediated glutamate cytotoxicity. Brain Research, 644, 181–187. Arneric, S., Sullivan, J., & Williams, M. (1995). Neuronal nicotinic acetylcholine receptors: Novel targets for CNS therapeutics. In F. Bloom & D. Kupfer (Eds.), Psychopharmacology: Fourth generation of progress (pp. 95–110). New York: Raven Press. Beninger, R. J., Wirsching, B. A., Jhamandas, K., Boegman, R. J., & El-Defrawy, S. R. (1986). Effects of altered cholinergic function on working and reference memory in the rat. Canadian Journal Physiology and Pharmacology, 64, 376–382. Block, F., Kunkel, M., & Schwarz, M. (1993). Quinolinic acid lesion of the striatum induces impairment in spatial learning and motor performance in rats. Neuroscience Letters, 149, 126–128. Brioni, J. D., Decker, M. W., Gamboa, L. P., Izquierdo, I., & McGaugh, J. L. (1990). Muscimol injections in the medial septum impair spatial learning. Brain Research, 522, 227–234. Brioni, J. D., Decker, M. W., Sullivan, J. P., & Arneric, S. P. (1996). The pharmacology of (0)nicotine and novel cholinergic channel activators. Advances in Pharmacology, 37, 153–214. Brioni, J. D., Morgan, S. J., O’Neill, A. B., Sykora, T. M., Postl, S. P., Pan, J. B., Sullivan, J. P., & Arneric, S. P. (1996). ‘‘In vivo’’ profile of novel nicotinic ligands with CNS selectivity. Medicinal Chemistry Research, 6, 487–510. Chen, Y., Ogren, S., Bjelke, B., Bolme, P., Eneroth, P., Gross, J., Loidl, F., Herrera-Marschitz, M., & Anderson, K. (1995). Nicotine treatment counteracts perinatal asphyxia-induced changes in the mesostriatal-limbic dopamine systems and in motor behaviour in the four-week-old male rat. Neuroscience, 68, 531–538. Decker, M. W., Brioni, J. D., Sullivan, J. P., Buckley, M. J., Radek, R. J., Raskiewickz, J. L., Kang, C. H., Kim, D. J. B., Giardina, W. J., Williams, M., & Arneric, S. P. (1994). ABT 418: A novel cholinergic ligand with cognition enhancing and anxiolytic activities. II. In vivo characterization. Journal of Pharmacology and Experimental Therapeutics, 270, 319–328. Donnelly-Roberts, D., Xue, I., Arneric, S., & Sullivan, J. (1996). In vitro neuroprotective properties of a novel cholinergic channel activator (ChCA), ABT-418. Brain Research, 719, 36–44. Foster, A. C., & Woodruff, G. N. (1988). Neuroprotective effects of MK-801 ‘‘in vivo’’: Selectivity and evidence for delayed degeneration mediated by NMDA receptor activation. Journal of Neuroscience, 8, 4745–4754. Fuxe, K., Rosen, L., Lippoldt, A., Andbjer, B., Hasselrot, U., Finman, U., & Agnati, L. (1994). Chronic continuous infusion of nicotine increases the dissapearance of choline acetyltransferase immunoreactivity in the cholinergic cell bodies of the medial septal nucleus following a partial unilateral transection of the fimbria fornix. Clinical Investigations, 72, 262–268. Heyes, M. P., Saito, K., Major, E. O., Milstien, S., Markey, S. P., & Vickers, J. H. (1993). A mechanism of quinolinic acid formation by brain in inflammatory neurological disease. Brain, 116, 1425–1450. Janson, A., Fuxe, K., Agnati, L., Katayama, I., Harfstrand, A., Anderson, K., & Goldstein, M. (1988). Chronic nicotine treatment counteracts the disappearance of tyrosine-hydroxylaseinmunoreactive nerve cell bodies, dendrites and terminals in the mesostriatal dopamine system of the male rat after partial hemitransection. Brain Research, 455, 332–345. Kodsi, M. H., & Swerdlow, N. R. (1994). Quinolinic acid lesions of the ventral striatum reduce sensorimotor gating of acoustic startle in rats. Brain Research, 643, 59–65. Levin, E. D. (1992). Nicotinic systems and cognitive function. Psychopharmacology, 108, 417–431. Lipton, S. A., & Rosenberg, P. A. (1994). Excitatory amino acids as a final common pathway for neurological disorders. New England Journal of Medicine, 330, 613–622. Marin, P., Maus, M., Desagher, S., Glowinski, J., & Premont, J. (1994). Nicotine protects cultured striatal neurons against N-methyl-D-aspartate receptor-mediated neurotoxicity. NeuroReport, 5, 1977–1980. Martin, E., Panickar, K., King, M., Deyrup, M., Hunter, B., Wang, G., & Meyer, E. (1994). Cytoprotective actions of 2,4-dimethoxybenzylidene anabaseine in differentiated PC12 cells and septal cholinergic neurons. Drug Development Research, 31, 135–141.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM
64
O’NEILL, MORGAN, AND BRIONI
Morris, R. (1984). Developments of a water-maze procedure for studying spatial learning in the rat. Journal of Neuroscience Methods, 11, 47–60. O’Neill, A. B., Morgan, S. J., Sykora, T., Postl, S., & Brioni, J. D. (1996). Hippocampal neurodegeneration induced by quinolinic acid: Histological and behavioral protection by nicotine. Society for Neuroscience Abstracts, 22, 834.9. Owman, C., Fuxe, K., Jason, A., & Kahrstrom, J. (1989). Studies of protective actions of nicotine on neuronal and vascular functions in rats: Comparison between sympathetic noradrenergic and mesostriatal dopaminergic fiber system, and the effect of a dopamine agonist. Progress in Brain Research, 79, 267–276. Perkins, M. N., & Stone, T. W. (1983). Quinolinic acid: Regional variations in neuronal sensitivity. Brain Research, 259, 172–176. Reinhard, J. F., Erickson, J. B., & Flanagan, E. M. (1994). Quinolinic acid in neurological disease: Opportunities for novel drug discovery. Advances in Pharmacology, 30, 85–127. Schwarcz, R., & Kohler, C. (1983). Differential vulnerability of central neurons of the rat to quinolinic acid. Neuroscience Letters, 38, 85–90. Schwarcz, R., Whetsell, W. O., & Mangano, R. M. (1982). Quinolinic acid: an endogenous metabolite that produces axon-sparing lesions in rat brain. Science, 219, 316–318. Sjak-Shie, N., & Meyer, E. (1993). Effects of chronic nicotine and pilocarpine administration on neocortical neuronal density and [3H]GABA uptake in nucleus basalis lesioned rats. Brain Research, 624, 295–298. Stone, T. W. (1993). Neuropharmacology of quinolinic and kynurenic acid. Pharmacological Reviews, 45, 309–379. Stone, W., & Perkins, M. N. (1981). Quinolinic acid: A potent endogenous excitant at aminoacid receptors in CNS. European Journal of Pharmacology, 72, 411–412. Vecsei, L., & Beal, M. F. (1991). Comparative behavioral and neurochemical studies with striatal kainic acid or quinolinic acid lesioned rats. Pharmacology Biochemistry and Behavior, 39, 473–478. Winn, P., Stone, T. W., Latimer, M., Hastings, M. H., & Clark, A. J. M. (1991). A comparison of excitotoxic lesions of the basal forebrain by kainate, quinolinate, ibotenate, N-methyl-Daspartate or quisqualate, and the effects on toxicity of 2-amino-5-phophonovaleric acid and kynurenic acid in the rat. British Journal of Pharmacology, 102, 904–908. Wirsching, B. A., Beninger, R. J., Jhamandas, K., Boegman, R. J., & Bialik, M. (1989). Kynurenic acid protects against the neurochemical and behavioral effects of unilateral quinolinic acid injections into the nucleus basalis of rats. Behavioral Neuroscience, 103, 90–97.
AID
NLM 3803
/
6v14$$$121 01-30-98 13:27:03
nlmoa
AP: NLM