Long-term effects of neonatal MK-801 treatment on prepulse inhibition in young adult rats
- First Online:
- Cite this article as:
- Uehara, T., Sumiyoshi, T., Seo, T. et al. Psychopharmacology (2009) 206: 623. doi:10.1007/s00213-009-1527-2
- 158 Views
Blockade of N-methyl-d-asparate (NMDA) receptors has been shown to produce some of the abnormal behaviors related to symptoms of schizophrenia in rodents and human. Neonatal treatment of rats with non-competitive NMDA antagonists has been shown to induce behavioral abnormality in a later period.
The aim of this study was to determine whether brief disruption of NMDA receptor function during a critical stage of development is sufficient to produce sensorimotor-gating deficits in the late adolescence or early adulthood in the rat.
Male pups received the NMDA receptor blocker MK-801 (0.13 or 0.20 mg/kg), or an equal volume of saline on postnatal day (PD) 7 through 10. The animals were tested twice for prepulse inhibition (PPI) and locomotor activity in pre- (PD 35-38) and post- (PD 56-59) puberty.
Neonatal exposure to both doses MK-801 disrupted PPI in the adolescence and early adulthood. Low-dose MK-801 elicited long-term effects on startle amplitudes, whereas high-dose MK-801 did not. Neither dose of MK-801 showed a significant effect on spontaneous locomotor activity, whereas the high dose attenuated rearing.
The results of this study suggest neonatal exposure to MK-801 disrupted sensorimotor gating in the adolescence and early adulthood stages. These findings indicate that rats transiently exposed to NMDA blockers in neonatal periods are useful for the study of the pathophysiology and treatment of schizophrenia.
KeywordsNMDA receptorMK-801NeonatalPrepulse inhibitionLocomotor activityRatAnimal modelSchizophrenia
Non-competitive antagonists at the N-methyl-d-asparate (NMDA) receptor have been shown to induce schizophrenia-like symptoms, i.e., positive and negative symptoms, and cognitive dysfunction in normal subjects (Jentsch and Roth 1999). These observations lead to the concept that rodents treated with NMDA receptor antagonists provide an animal model of schizophrenia (Bubenikova-Valesova et al. 2008; Jentsch and Roth 1999). In accord with this view, numerous studies reported that NMDA receptor antagonists, such as phencyclidine (PCP), MK-801, and ketamine, produce behavioral changes reminiscent of symptoms of schizophrenia i.e. hyperlocomotion, stereotypy, information-processing deficits, impairments of cognitive function and social interactions (Breese et al. 2002; Bubenikova-Valesova et al. 2008; Moghaddam and Jackson 2003).
It has been proposed that schizophrenia is a neurodevelopmental disorder caused by insults in the second trimester (Roberts 1991), which results in an abnormal development of the frontal cortex. This developmental change becomes clinically overt at the time of sexual maturation, and is associated with subsequent dysregulation of the neural system (Weinberger 1995). Morphological studies have demonstrated a reduction in the volume of the prefrontal cortex and the temporal lobes in schizophrenia (Kawasaki et al. 2004; Suzuki et al. 2005). These findings lead to the hypothesis that latent dysfunction of the temporal regions becomes overt by additional pathological changes in the frontal lobes, leading to manifestation of positive psychotic symptoms (Kurachi 2003a, b; Siever and Davis 2004). This hypothesis is supported by animal studies which found inactivation of the medial prefrontal cortex (mPFC), in subjects with structural abnormalities in the entorhinal cortex leads to dysregulation of dopaminergic neurotransmissions in the limbic regions, such as the amygdala and nucleus accumbens (Uehara et al. 2000, 2003, 2004, 2007).
Several researchers have attempted to develop animal models of schizophrenia using rats based on the neurodevelopmental hypothesis or the NMDA-receptor-dysfunction hypothesis (Beninger et al. 2002; Harris et al. 2003; Rasmussen et al. 2007; Stefani and Moghaddam 2005; Takahashi et al. 2006; Wang et al. 2001). Treatment of neonatal rats with non-competitive NMDA antagonists (e.g., MK-801, phencyclidine) led to behavioral abnormalities related to clinical symptoms of schizophrenia (e.g., locomotor activity, prepulse inhibition; Prepulse inhibition (PPI), spatial memory test) in the adult stage. Moreover, Stefani et al. reported that neonatal treatment with MK-801 (0.20 mg/kg/day) between postnatal day (PD) 7 and 10 led to frontal dysfunction in adult rats, such as impairment of the set-sifting test (Stefani et al. 2003; Stefani and Moghaddam 2005).
A sudden strong acoustic stimulus produces acoustic startle response in both humans and rodents. PPI is a phenomenon defined as reduction in startle reflex by prior presentation of a weak, non-startling stimulus (Graham 1975; Hoffman and Searle 1968), and has been used as a measure of sensorimotor gating. PPI can be measured both in human and animals, and is disrupted in patients with schizophrenia (Braff et al. 1992, 1999). NMDA antagonists have been shown to produce PPI deficits in animals (Bubenikova et al. 2005; Seo et al. 2008). However, conflicting results exist about long-term effects of postnatal NMDA receptor blockade on PPI. Some studies demonstrated that postnatal exposure to NMDA antagonists disrupts PPI in adult stage (Takahashi et al. 2006; Wang et al. 2001), whereas others did not find such an effect (Harris et al. 2003; Rasmussen et al. 2007).
The aim of this study was to determine whether a brief disruption of NMDA receptor function by MK-801 during the neonatal stage would produce sensorimotor-gating deficits in the late adolescence or early adulthood. For this purpose, we administered MK-801 to rats on PD 7–10 (Stefani and Moghaddam 2005), because this animal model has been shown to elicit frontal dysfunction at these later stages, an important component of the pathophysiology of schizophrenia (Kurachi 2003a, b; Siever and Davis 2004; Weinberger 1995).
Materials and methods
Female Wistar rats obtained at 14 days of pregnancy (Japan SLC, Japan) were housed individually at 24 ± 2°C under a 12-h light (0700–1900 h)–12-dark cycle with free access to food and water. At the time of weaning (PD 25), the animals were grouped into four to six in each same treatment, described in the next paragraph, in a cage with free access to food and water. The procedures complied with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All experiments were reviewed and approved by the Committee of Animal Research, University of Toyama School of Medicine.
On PD 7, male pups (10–15 g) were randomly assigned into three groups. They received MK-801 (Sigma-Aldrich, St. Louis, MO, 0.13 or 0.20 mg/kg s.c.) or equal volume of saline (vehicle group). From PD 7 to PD 10, pups were injected with drugs, between 8:00–10:00. All pups from the same litter received the same treatment (low-dose MK-801; two litters, high-dose MK-801; two litters, vehicle; two litters) to avoid the influence of cross-fostering and minimize maternal care effects. There were no statistical litter effects on all parameters in this study.
Apparatus and procedure of PPI
Rats in each group were tested twice for PPI on PD 35 and PD 56. The procedure of PPI measurement was based on our previous studies (Seo et al. 2008; Uehara et al. 2007). All testing occurred within startle chambers (Ohara & Co., LTD, Tokyo), which was housed in a sound-attenuated room with a 60 dB ambient noise level. Each startle chamber consisted of a Plexiglas cylinder 9.4 cm in internal diameter resting on an 11 cm × 22 cm Plexiglas stand. Acoustic stimuli and background noise were given via speakers mounted 12.2 cm above the Plexiglas cylinders, controlled with a computer box (Ohara & Co., LTD, Tokyo). A piezoelectric device mounted below the Plexiglas stand detected and transduced motion within the cylinder.
Rats were placed in a startle chamber. Five minutes after the acclimation period, they were exposed to six blocks of four different stimulus types, i.e., pulse-alone: 40 ms 120 dB white noise bursts; prepulse-pulse: 20 ms, white noise pulse of 70, 74, 78, 82 and 86 dB followed by 20 ms 120 dB white noise pulse at a fixed inter-stimulus interval (ISI) of 100 ms. Trials were presented in randomized order, with 20, 25, and 30 s randomized interval.
After the PPI examination, spontaneous activity was measured for 30 min in an ambulation observation chamber (blackened vinyl chloride cages, 40 cm × 40 cm × 40 cm; AMB-3001, Ohara & Co., Ltd., Tokyo, Japan) equipped with 6 × 6 photoelectric light sources spaced at 7-cm intervals and 2.5 cm (for locomotor activity) above the floor (AMB-2020, Ohara & Co., Ltd.; Sumiyoshi et al. 2004; Uehara et al. 2007). Rearing was measured photoelectric light sources spaced at 12 cm (on PD 36–38) or 19 cm (on PD 57–59) above the floor. Interruptions of light beams were registered as activity counts, and were summarized every 5 min by the Logger Interface control system (IF-10-LOG, Ohara & Co., Ltd.). For convenience, the test days are referred to as PD 35 and PD 56 throughout the manuscript.
Presentation of the results and statistics
Data were analyzed by analysis of variance (ANOVA) using SPSS software (version 16.0 J for Mac, SPSS Inc.). For comparison of body weight, two-way repeated measures ANOVA was performed with treatment status (Status = low-dose MK-801, high-dose MK-801, vehicle) as between-subject factor and age (PD 35 and PD 56) as repeated measures variable, followed by Tukey’s HSD test.
PPI data were presented as percentage of PPI (%PPI), which was calculated from startle amplitudes (SAs) using the following formula: %PPI = 100 − [(SA for prepulse-pulse trials) / (SA for pulse-alone trials)] × 100. Between-group comparisons were performed by three-way repeated measures ANOVA with treatment status (Status = low-dose MK-801, high-dose MK-801, vehicle) as between-subject factor, whereas prepulse intensity and age (PD 35 and PD 56) were treated as repeated measures variables. If treatment × age interaction was significant, two-way repeated measures ANOVA were performed with treatment status as between subject and prepulse intensity as repeated measures variables at each age (PD 35 and PD 56) separately. For comparisons of SAs, two-way repeated measures ANOVA was performed with treatment status as between-subject factor and age was treated as repeated measures variable. Post-hoc comparisons were made between all treatment groups with Tukey’s HSD test.
For comparisons of locomotor activity, we divided the time for activity counts into two time-intervals (0–15 min and 16–30 min). Three-way repeated measures ANOVA was performed with treatment status as between-subject factor, whereas age (PD 35 and PD 56) and time (first-half interval and second-half interval) as repeated measures variable followed by Tukey’s HSD test. A probability (P) of less than 0.05 was considered to be significant.
Effect of neonatal MK-801 treatment on body weight across development
Body weight across development
Vehicle (n = 10)
126.2 ± 1.5
295.2 ± 3.7
MK-801 (0.13 mg/kg) (n = 12)
114.2 ± 1.6
282.8 ± 3.5
MK-801 (0.20 mg/kg) (n = 13)
86.8 ± 2.0
224.6 ± 4.3
Effect of neonatal MK-801 treatment on startle and prepulse inhibition
Subsequent analysis was conducted to examine main effect of treatment on PD 35 and PD 56 separately. On PD 35, a significant main effect of treatment status [F(2,32) = 3.60, P = 0.04], but not treatment status × prepulse intensity interaction [F(2,32) = 2.13, P = 0.14] on PPI was noted. Tukey’s HSD test revealed that low-dose MK-801 reduced PPI compared to vehicle treatment (P = 0.03). However, there was no significant difference in PPI between vehicle and high MK-801 groups (P = 0.49). On PD 56, a significant main effect of treatment status [F(2,32) = 7.45, P = 0.002] and treatment status × prepulse intensity interaction [F(2,32) = 3.47, P = 0.04] on PPI was found. Tukey’s HSD test demonstrated a significant reduction of PPI in high-dose MK-801 group compared to vehicle group (P = 0.001). There was no difference in PPI between vehicle and low MK-801 groups (P = 0.15) on PD 56 (Fig. 2).
Effect of neonatal MK-801 treatment on locomotor activity
The present study demonstrates that transient, neonatal exposure to MK-801 induces disruption of sensorimotor gating in the adolescence and early adulthood. Low-dose MK-801 (0.13 mg/kg) elicited long-term effects on SAs, whereas a higher dose (0.2 mg/kg) did not on PD35 and 56. Treatment with low-dose MK-801 led to reduction of PPI without any effect on SAs on PD 35. PPI on PD 56 was not affected by low-dose MK-801, whereas SAs were increased. By contrast, high-dose MK-801 disrupted PPI only on PD 56 without any effect on SAs. On the other hand, either dose MK-801 did not affect spontaneous locomotor activity, although only the high dose decreased rearing activity.
Neonatal MK-801 treatment inhibited weight gain across development, replicating the results of a previous study (Stefani and Moghaddam 2005). Results in this study further revealed the effect of MK-801 on body weight is dose-dependent. In this study, all pups from the same litter received the same treatment to avoid the influence of cross-fostering and minimize maternal care effects. There were no statistical litter effects on all parameters (data not shown). In fact, only two litters were used in each treatment group, which may not be enough to detect any litter effects.
In this study, both doses of MK-801 disrupted PPI in pre- (PD 35) and post- (PD 56) puberty. Conflicting results have been reported on long-term effects of postnatal NMDA receptor blockade on PPI. Treatment with PCP (10 mg/kg, s.c.) on PD 7, 9, and 11 disrupted PPI on PD 24-26 (Wang et al. 2001), but not on PD 32, 39, 51, 72, and 88 (Rasmussen et al. 2007). In the former study, olanzapine reversed PCP-induced PPI disruption (Wang et al. 2001). MK-801 treatment (0.5 or 1.0 mg/kg, i.p.) on PD 3 did not affect PPI on PD 35 and 56, whereas it reduced sensitivity to prepulse intensity changes on PD 56 (post-puberty), but not on PD 35 (Beninger et al. 2002). On the other hand, MK-801 (0.5 mg/kg, s.c.) treatment on PD 7 did not affect PPI on PD 56 (Harris et al. 2003). These results suggest that postnatal treatment with NMDA antagonists diminishes sensorimotor gating transiently around the time of puberty (Rasmussen et al. 2007).
Results of the current study revealed that administration of MK-801 on PD 7–10 elicits prolonged effect on disrupted PPI. Treatment with PCP (2 or 10 mg/kg, s.c.) from PD 3 to PD 16 has been shown to diminish PPI in 8-week-old rats (Takahashi et al. 2006). Overall, the ability of NMDA receptor blockade during the neonatal period to disrupt PPI may depends on treatment duration. It is also worthwhile to note that studies using Sprague–Dawley (SD) rats did not demonstrate PPI disruption induced by postnatal treatment of NMDA antagonists, unlike the results of current study using Wistar rats. This may represent stain-dependent effects of PCP on PPI, as has been reported in an acute study (Varty and Higgins 1994).
MK-801 treatment (0.20 mg/kg/day) between PD 7 and PD 10 has been shown to impair performance on the set-shifting test (Stefani et al. 2003; Stefani and Moghaddam 2005), a measure of medial prefrontal cortex (mPFC) function which is dependent on NMDA receptor-mediated neurotransmission (Stefani and Moghaddam 2003). These previous results indicate that MK-801 treatment in the neonatal period impairs NMDA receptor function in the mPFC in matured rats. On the other hand, PPI is thought to be regulated by the prefrontocortico-limbic-striato-pallidal circuit that connects to the primary acoustic startle response pathway through mesopontine and nigral projections (Swerdlow et al. 2001). NMDA receptors in the mPFC, amygdala, and hippocampus play an important role in this circuit (Bakshi and Geyer 1998; Schwabe and Koch 2004). These lines of evidence suggest the role of NMDA receptor blockade by neonatal exposure to MK-801 in the manifestation of PPI disruption in a later period, as observed in this study.
Prenatal administration (E15-E18) of MK-801 has been reported to decrease the number of parvalbumin-positive GABAergic interneurons in the mPFC in adult rats (PD 35 and 63; Abekawa et al. 2007). Manipulations to decrease GABAergic transmissions in the mPFC have been shown to disrupt PPI, presumably by disinhibition of descending glutamatergic fibers (Japha and Koch 1999; Schwabe and Koch 2004; Swerdlow et al. 2001). Further studies are warranted to determine if histochemical changes of parvalbumin-positive GABAergic interneurons are present in the animal model studied here.
Low-dose MK-801 enhanced SAs on PD 56, whereas high-dose MK-801 did not affect them. This finding appears not consistent with previous studies reporting that neonatal NMDA antagonist administration did not affect SAs in the later stage (Rasmussen et al. 2007; Wang et al. 2001). However, it was also reported that postnatal treatment with (E)-2-amino-4-methyl-5-phostphno-3-pentenoic acid (CGP 40116), a competitive antagonist at NMDA receptors, increased SAs in early adult stage (Wedzony et al. 2008). The increase in the amplitude of the startle reflex may result from supersensitivity of glutamatergic receptors in the ventral nucleus of the lateral lemniscus (VLL) and the caudal pontine reticular nucleus (PnC; Krase et al. 1993; Spiera and Davis 1988). Moreover, repeated (PD 7,9,11), but not single (PD 7), injection of PCP reduces the number of the NMDA receptors in the striatum, whereas both sub-chronic and single injection of PCP increases the number of these receptors in the frontal cortex (Anastasio and Johnson 2008). These results suggest that functional changes of NMDA receptors induced by neonatal treatment with NMDA antagonists depends on dose and region.
Neonatal exposure to MK-801 did not affect locomotor activity. This finding is in line with previous studies reporting that neonatal treatment with NMDA receptor antagonists did not affect novelty-induced and spontaneous locomotor activity both in pre- and post-pubertal periods (Beninger et al. 2002; Harris et al. 2003; Stefani and Moghaddam 2005). Rats treated with NMDA receptor antagonists in the neonatal period showed enhanced PCP (Wang et al. 2001) or amphetamine-induced (Beninger et al. 2002) locomotor activity. On the other hand, we found enhancement of rearing on PD 56 in rats treated with low-dose MK-801. MK-801 treatment (0.5 and 1.0 mg/kg) on PD 3 has been shown to enhance amphetamine-induced rearing activity at PD 35 and PD 56 (Beninger et al. 2002). These findings suggest that NMDA receptor blockade at a neonatal period caused behavioral changes related to dopamine supersensitivity around puberty.
In conclusion, neonatal exposure to MK-801 disrupted PPI in the adolescence and early adulthood, possibly associated with NMDA receptor dysfunction in the mPFC. In view of the neurodevelopmental and NMDA hypotheses of schizophrenia, these findings indicate that rats transiently exposed to NMDA blockers in neonatal periods are useful for the study of the pathophysiology and treatment of schizophrenia.
The authors gratefully acknowledge the insightful comments and criticism by Dr. M. Tsunoda and Dr. K. Tanaka.
This study was supported by a Grant-in-Aid for Scientific Research from Japan Society for the Promotion of Science (No. 20591363).