Two kinds of mitogen-activated protein kinase phosphatases, MKP-1 and MKP-3, are differentially activated by acute and chronic methamphetamine treatment in the rat brain

Authors


Address correspondence and reprint requests to Dr Manabu Takaki, Department of Neuropsychiatry, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700–8558, Japan. E-mail: manabuta@cc.okayama-u.ac.jp

Abstract

Two functionally different MAP kinase phosphatases (MKPs) were investigated to clarify their roles in behavioral sensitization to methamphetamine (METH). MKP-1 mRNA levels increased substantially by about 60–300% in a range of brain regions, including several cortices, the striatum and thalamus 0.5–1 h after acute METH administration. After chronic METH administration its increase was less pronounced, but a more than 50% increase was still seen in the frontal cortex. MKP-1 protein levels also increased 3 h after acute or chronic METH administration. MKP-3 mRNA levels increased by about 30–50% in several cortices, the striatum and hippocampus 1 h after acute METH administration, but only in the hippocampus CA1 after chronic METH administration. Pre-treatment with the D1 dopamine receptor antagonist, SCH23390, attenuated the METH-induced increase of MKP-1 and MKP-3 mRNA in every brain region, while pre-treatment with the NMDA receptor antagonist, MK-801, attenuated it in some regions. These findings suggest that in METH-induced sensitization, MKP-1 and MKP-3 play important roles in the neural plastic modification in widespread brain regions in the earlier induction process, but in the later maintenance process, they do so only in restricted brain regions such as MKP-1 in the frontal cortices and MKP-3 in the hippocampus.

Abbreviations used
AMPH

amphetamine

Arc

activity-regulated cytoskeleton-associated protein

ERK

extracellular signal-regulated kinase

MAP

mitogen-activated protein

MKPs

MAP kinase phosphates

METH

methamphetamine

PBS

phosphate-buffered saline

TBS

Tris-buffered saline

t-PA

tissue plasminogen activator

SAPK/JNK

stress-activated protein kinase/c-Jun N-terminal kinase

TTBS

Tween Tris-buffered saline

VTA

ventral tegmental area.

Abuse of psychostimulants such as amphetamine (AMPH), methamphetamine (METH) and cocaine leads to the gradual development of psychotic symptoms that resemble those of paranoid schizophrenia (Sato et al. 1983). Patients who have exhibited psychostimulant-induced psychosis are prone to relapse either following a re-injection of the psychostimulant or being subjected to stress. In rodents, it has been shown consistently that repeated administration of psychostimulants leads to progressive augmentation of stereotypical behavior, a phenomenon that is known as behavioral sensitization (Robinson and Becker 1986; Akiyama et al. 1994). Sensitized animals remain hypersensitive to a subsequent dose of psychostimulants, with regard to the psychomotor activating and rewarding effects, for months to years (Paulson et al. 1991; Robinson and Kolb 1997). The use of experimental behavioral sensitization, which is apparently analogous to psychostimulant-induced psychosis in humans, has enabled a great deal of research aimed at elucidating the neural and molecular mechanisms underlying this liability. Accumulating evidence suggests that the behavioral sensitization phenomenon induced by psychostimulants should be accompanied by long-lasting neural plasticity, which may involve structural modifications in the nervous system, such as lasting alterations in the patterns and density of synaptic connectivity (Robinson and Becker 1986). A microscopy study using Golgi-stained tissue has revealed that exposure to chronic AMPH produces a long-lasting increase in the length of dendrites, the density of dendritic spines and the number of branched spines on the major output cells of the nucleus accumbens and prefrontal cortex (Robinson and Kolb 1997). However, the molecular basis for such morphological changes as a result of treatment with psychostimulants is not well understood. There are only a few studies in which the expressional changes of plasticity-related genes resulting from treatment with psychostimulants have been investigated. For example, tissue plasminogen activator (t-PA) mRNA, which may participate in axonal growth and the formation and rearrangement of synaptic contacts, increases in some cortices after acute administration of METH and cocaine (Hashimoto et al. 1998). Acute cocaine administration causes an increase in synaptotagmin IV mRNA, a synaptic vesicle protein, in the striatum (Denovan-Wright et al. 1998). Acute and chronic METH administration increases the levels of activity-regulated cytoskeleton-associated protein (Arc) mRNA, an effector immediate early gene, in the striatum, cerebral cortices and hippocampus CA1 (Kodama et al. 1998). The expression of synaptophysin mRNA, a major integral membrane protein of small presynaptic vesicles (Marqueze-Pouey et al. 1991), was reported to increase in the nucleus accumbens, prefrontal and temporal cortices after acute METH administration, while that of stathmin mRNA, a growth-associated cytosolic phosphoprotein (Sobel 1991) was reported to increase in the prefrontal cortex after acute METH administration (Takaki et al. 2001). These studies indicated that t-PA, synaptotagmin IV, synaptophysin and stathmin could be implicated in the early induction process of psychostimulant-induced neural plasticity, while Arc mRNA could be implicated in both the early induction and later maintenance processes.

Mitogen-activated protein (MAP) kinases play an important role in cell growth, proliferation and apoptosis (Bokemeyer et al. 1996; Muda et al. 1996b, 1998; Boschert et al. 1998). MAP kinases are activated by their phosphorylation and selectively inactivated by MAP kinase phosphatases (MKPs), which are themselves induced or activated by MAP kinases (Bokemeyer et al. 1996; Muda et al. 1996a, 1996b; Brondello et al. 1997). Several lines of evidence indicate that MAP kinases and MKPs are involved in psychostimulants-induced behavioral sensitization (Berhow et al. 1996; Thiriet et al. 1998; Pierce et al. 1999; Valjent et al. 2000). To elucidate the mechanism of psychostimulants-induced behavioral sensitization further, we used in situ hybridization and western blot analysis to investigate the effect of acute and chronic METH administration on mRNA and protein levels of two kinds of MKPs, MKP-1 and MKP-3, which act on different substrates, in the rat brain. We also investigated the effect of pharmacological manipulation on MKP-1 and MKP-3 mRNAs using the D1 dopamine receptor antagonist, SCH23390, and the NMDA receptor antagonist, MK-801, which have been reported previously to inhibit psychostimulant-induced behavioral sensitization (Karler et al. 1989; Ujike et al. 1989; Stewart and Druhan 1993).

Materials and methods

Animals

Male Sprague–Dawley rats (Charles River, Yokohama, Japan) weighing 220–240 g were housed under a 12-h light/12-h dark cycle (light on at 0700 h, light off at 1900 h) and at constant temperature (25°C) and humidity, and allowed free access to food and water. All of the animals used were handled gently for 3 min, once daily, for one week before being subjected to drug treatment. All procedures used on the animals were in strict accordance with the Guidelines for Animal Experiments of Okayama University Medical School.

Treatment protocol

For the acute treatment time-course experiment, the animals were decapitated in the METH-naive state and 0.5, 1, 3, 6 and 24 h (n = 6 for each time point) after a single injection (i.p.) of 4 mg/kg METH.

For the pharmacological blockade experiments, the animals were treated with either 0.5 mg/kg SCH23390, 0.25 mg/kg MK-801 or an equal volume of saline (SAL) i.p. injection, 30 min before i.p. injection with either 4 mg/kg METH or SAL, and all of the animals were killed 1 h later (n = 6 or 8 in each group). This resulted in six treatment groups: pre-treatment with SCH23390 followed by METH challenge (SCH-METH) or SAL challenge (SCH-SAL); pre-treatment with MK-801 followed by METH challenge (MK-METH) or SAL challenge (MK-SAL); and pre-treatment with SAL followed by METH challenge (SAL-METH) or SAL challenge (SAL-SAL).

For the chronic treatment experiment, the animals were treated with either 4 mg/kg METH or an equal volume of SAL i.p. injection once daily for 10 days. We have reported previously that this chronic METH treatment induces the development of behavioral sensitization, which is manifested as the augmentation of stereotypical behavior (Ujike et al. 1989). The animals were then killed either 1 h or 24 h after the last injection. This resulted in three treatment groups (n = 8 in each group): animals decapitated 1 h after the last injection of chronic METH administration (METH 1 h group); animals decapitated 24 h after the last injection of chronic METH administration (METH 24 h group); and animals decapitated 1 h after the last injection of chronic SAL administration (SAL 1 h group).

For western blot analysis of the effects of acute and chronic treatment, the rats were killed by decapitation 1 h or 3 h after either a single injection of 4 mg/kg METH or an equal volume of SAL, and after the last injection of chronic METH (4 mg/kg) or equal volume of SAL administration for 14 days.

Synthesis of oligonucleotide probes

The following antisense oligonucleotide probes complementary to the cDNA sequences of the two molecules were produced according to the method of Boschert et al. (1998): MKP-1 (accession number: X84004, 874–913) 5′-CAACGAGGCGATT GACTTTATAGACTCCATCAAGGATGCT–3′; and MKP-3 (accession number: X94185, 626–667) 5′-GGTGGCTTCAGTAA GTTCCAGGCCGAGTTCGCCCTGCACTGC–3′. A search of the GenBank database using the BLAST program revealed no significant homologous sequence with these oligonucleotide probes.

In situ hybridization

We performed in situ hybridization according to the method of Kodama et al. (1998). In preliminary experiments designed to acquire definite evidence for the sequence specificity of the signal, coronal brain sections of 10 µm thickness were incubated in the radiolabeled oligonucleotide probe in the presence of a 50 : 1 molar excess of the unlabeled oligonucleotide probe. Sections were wrapped in plastic film, exposed to a phosphorimaging plate (BAS-IP MS 2040, Fujifilm, Japan), scanned using a Bas station (BAS-2000 II, Fujifilm) for 1 day and then analyzed by image analysis software (Mac Bas version 2.2, Fujifilm). Brain regions to be integrated in the image analysis were determined by free-hand enclosure. Background signals were determined by a count of the area without brain tissue and the specific signals were calculated by the reduction of this background value. The average decay per minute (d.p.m.) values of the left and right sides of the orbital cortex, prefrontal cortex, parietal cortex, temporal cortex, striatum and the CA1 and CA3 regions of the hippocampus were quantified using the Bas system. These densities were converted into cpm/mg using a standard curve (log fit) generated from autoradiographic [14C] microscales (Amersham, Piscataway, NJ, USA). Selected slides were dipped in Hypercoat LM-1 (Amersham), and stored at 4°C for 12 weeks before being developed. The sections were counterstained with crecyl violet.

Western blot analysis

The brains were removed rapidly, separated into the prefrontal cortex, hippocampus, and striatum, and stored in Eppendorf microtubes at −80°C. Each brain sample was homogenized in 3 mL of ice-cold RIPA buffer [phosphate-buffered saline (PBS), 1% NP40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS)], 1% phenylmethylsulfonyl fluoride (Sigma, St Louis, MO, USA; 10 mg/mL in isopropanol), 3% aprotinin (Sigma) and 1% sodium orthovanadate (Sigma) per gram of brain tissue using a High Intensity Ultrasonic Liquid Processor (Sonics and Materials, New Town, CT, USA). The homogenates were centrifuged at 15 000 g in a refrigerated microcentrifuge for 20 min at 4°C. The supernatant fluid was pooled and used for the experiments.

Protein samples were mixed with an equal volume of the sample buffer (95% Laemmli sample buffer [Bio-Rad], 5% 2-mercaptoethanol) and boiled for 90 s. Samples (5 µL) were loaded into the wells and subjected to SDS-polyacrylamide gel electrophoresis (SDS–PAGE) on 10% acrylamide gel (Bio-Rad, Hercules, CA, USA) applied at 100 V for 90 min using SDS–PAGE transfer buffer (25 mm Tris, 0.1% SDS, 192 mm glycine). The proteins were then electrophoretically transferred onto nitrocellulose membranes (Bio-Rad) at 30 V overnight in a Bio-Rad wet blotting tank, using Towbin transfer buffer (25 mm Tris, 0.1% SDS, 192 mm glycine, 20% methanol). Membranes were rinsed in Tris-buffered saline (TBS) 20 mm Tris, 500 mm sodium chloride, pH 7.5 twice for 5 min, and subsequently blocked in TBS containing 5% dry skimmed milk for 90 min at room temperature (22°C). Membranes were then rinsed in Tween Tris-buffered saline (TTBS) 20 mm Tris, 500 mm sodium chloride, 0.05% Tween 20, pH 7.5 twice for 5 min then incubated overnight at room temperature with primary antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA; MKP-1 (V-15) sc-1199 1 : 500, MKP-3 (C-20) sc-8599 1 : 100 (N-18) sc-8598 1 : 100). They were then rinsed in TTBS buffer twice for 5 min and further incubated with a second antibody (Amersham, anti-rabbit IgG, 1 : 3000 for MKP-1; Santa Cruz Biotechnology, anti-goat IgG, 1: 2000 and 1 : 12000 for MKP-3) for 90 min at room temperature. Finally, the membranes were rinsed in TTBS buffer for 15 min four times, incubated with Enhanced chemiluminescence (ECL) reagent (Amersham) for 60 s then exposed to ECL-Hyperfilm (Amersham) for 60 s. After development, the bands detected by ECL were analyzed quantitatively. To examine the specificity of the detected band for western blotting, a five fold excess of MKP-1 specific blocking peptide, sc-1199P (Santa Cruz Biotechnology) was added to the primary antibodies.

Statistical analysis

The differences among groups were analyzed statistically using a one-way anova followed by Fisher's protected least significant difference procedure as a posthoc test. The level of statistical significance was set at p < 0.05.

Results

Distribution and time course of MKP-1 and MKP-3 mRNAs in brain areas after acute METH administration

The distribution of MKP-1 mRNA in the rat brain in the naive state and after acute METH administration, as detected by in situ hybridization, is shown in Fig. 1, upper panel. MKP-1 mRNA expressed after acute METH treatment distributed most densely in the whole cortices, moderately in the striatum and thalamus, with no signal in the hippocampus. The time course of MKP-1 mRNA levels after acute METH administration is shown in Fig. 1 lower panel. Levels of MKP-1 mRNA increased significantly 0.5–1 h after acute METH administration compared with basal levels. Levels peaked at 0.5 h in the orbital cortex rising by 301.21% (F5,30 = 18.08, p = 0.0001), the prefrontal cortex (194.61%, F5,30 = 10.31, p = 0.0001), the temporal cortex (146.52%, F5,30 = 12.14, p = 0.0001) and thalamus (88.48%, F5,30 = 2.96, p = 0.027); and at 1 h in the parietal cortex (129.23%, F5,30 = 11.88, p = 0.0001) and striatum (63.48%, F5,30 = 3.71, p = 0.01). Levels in each brain region subsided to basal levels by 3 h.

Figure 1.

Distribution and time course of MKP-1 mRNA in the rat brain in the naive state and after acute METH administration. Rats were decapitated before and at 0.5, 1, 3, 6 and 24 h (n = 6 for each time point) after a single i.p. injection of 4 mg/kg METH. The values are expressed as the mean ± SEM. The level of statistical significance was determined by one-way anova followed by Fisher's PSLD. *p < 0.05, **p < 0.01, ***p < 0.001. METH, methamphetamine; N, naive rat.

The distribution of MKP-3 mRNA in the rat brains in the naive state and after acute METH administration, as detected by in situ hybridization, is shown in Fig. 2, upper panel. MKP-3 mRNA expressed after acute METH treatment distributed most densely in the hippocampus CA1 and CA3, moderately in the whole cortex, with no signal in the thalamus. The time course of MKP-3 mRNA levels after acute METH administration is shown in Fig. 2, lower panel. Levels of MKP-3 mRNA increased significantly 0.5–3 h after acute METH administration, compared with basal levels. Levels peaked at 0.5 h in the striatum, rising by 25.38% (F5,30 = 8.10, p = 0.0001); at 1 h in the orbital cortex (50.35%, F5,30 = 12.00, p = 0.0001), prefrontal cortex (45.35%, F5,30 = 5.69, p = 0.0008), parietal cortex (46.44%, F5,30 = 14.92, p = 0.0001), temporal cortex (41.74%, F5,30 = 17.81, p = 0.0001) and hippocampus CA3 (34.26%, F5,30 = 6.56, p = 0.0003); and at 3 h in the hippocampus CA1 (38.16%, F5,30 = 14.48, p = 0.0001). Levels in each brain region subsided to basal levels by 6 h. Adding a 50-fold excess of unlabeled probe of MKP-1 or MKP-3 to the in situ hybridization buffer completely abolished the hybridization signal by the corresponding labeled probe (data not shown).

Figure 2.

Distribution and time course of MKP-3 mRNA in the rat brain in the naive state and after acute METH administration. Rats were decapitated before and at 0.5, 1, 3, 6 and 24 h (n = 6 for each time point) after a single i.p. injection of 4 mg/kg METH. The values are expressed as the mean ± SEM. The level of statistical significance was determined by one-way anova followed by Fisher's PSLD. *p < 0.05, **p < 0.01, ***p < 0.001. METH, methamphetamine; N, naive rat.

Microautoradiography of MKP-1 and MKP-3 mRNAs

An autoradiograph of the parietal cortex superimposed on a Nissl staining of the same section revealed that the MKP-1 mRNA signals in the parietal cortex could be superimposed on layers IV and VI (Fig. 3a). Bright-field photomicrography of MKP-1 mRNA showed increased levels in the parietal cortex 1 h after acute METH administration, compared with 1 h after SAL (Fig. 3b). Bright-field photomicrography of MKP-3 mRNA in the hippocampus 1 h after acute METH administration, compared with 1 h after SAL (Fig. 3c). MKP-1 and MKP-3 mRNAs were expressed in the neuron, but not in the glia.

Figure 3.

(a) An autoradiograph of the parietal cortex superimposed on a Nissl staining of the same section and its magnified image (lower panel). The rat was decapitated 1 h after a single i.p. injection of 4 mg/kg METH. The MKP-1 mRNA signals in the parietal cortex predominantly occur in two distinct layers, layers IV and VI. Bright-field photomicrograph of MKP-1 mRNA in the parietal cortex of layers VI (b) and MKP-3 mRNA in the hippocampus CA1 (c), 1 h after a single intraperitoneal injection of saline (upper panel) or 4 mg/kg methamphetamine (lower panel). The arrow and arrowhead points to a neuron and a glia, respectively.

Effect of pre-treatment with SCH23390 and MK-801 on MKP-1 mRNA

The effect of pre-treatment with SCH23390 on acute METH-induced MKP-1 mRNA expression is shown in Fig. 4(a). There were significant differences among the MKP-1 mRNA levels of the four groups (SAL-SAL, SAL-METH, SCH-SAL, SCH-METH) for each brain region, as follows: orbital cortex (F3,24 = 50.95, p = 0.001); prefrontal cortex (F3,24 = 30.69, p = 0.0001); parietal cortex (F3,24 = 52.59, p = 0.0001); temporal cortex (F3,24 = 50.48, p = 0.0001); striatum (F3,24 = 26.08, p = 0.0001); thalamus (F3,24 = 36.39, p = 0.0001). Treatment with SCH23390 alone (SCH-SAL group) did not alter the MKP-1 mRNA levels in any brain region compared with treatment with SAL (SAL-SAL group). Pre-treatment with SCH23390 inhibited the enhanced MKP-1 mRNA levels induced by acute METH administration. Inhibition rates were 41.2% in the orbital cortex, 45.9% in the pre-frontal cortex, 40.0% in the parietal cortex, 42.8% in the temporal cortex, 91.4% in the striatum, and 45.7% in the thalamus, calculated as:

Figure 4.

Effects of pre-treatment with SCH23390 (0.5 mg/kg) (a) and MK-801 (0.25 mg/kg) (b) on MKP-1 mRNA expression observed after acute METH (4 mg/kg) injection. Rats were pre-treated with saline, SCH23390 or MK-801 30 min before administration of either saline or METH, and killed 1 h later (n = 6 or 8 in each group). The values are expressed as the mean ± SEM. The level of statistical significance was determined by one-way anova followed by Fisher's PSLD. *p < 0.05, **p < 0.01, ***p < 0.001. METH, methamphetamine.

[(SAL-METH)−(SCH-METH)]/[(SAL-METH)−(SAL-SAL)]×100 (%).

The effect of pre-treatment with MK-801 on acute METH-induced MKP-1 mRNA is shown in Fig. 4(b). There were significant differences among the MKP-1 mRNA levels of the four groups (SAL-SAL, SAL-METH, MK-SAL, MK-METH) for each brain region examined, as follows: orbital cortex (F3,24 = 63.68, p = 0.0001); prefrontal cortex (F3,24 = 31.70, p = 0.0001); temporal cortex (F3,24 = 66.83, p = 0.0001); and striatum (F3,24 = 28.82, p = 0.0001) but not in the parietal cortex and thalamus. Treatment with MK-801 alone (MK-SAL group) increased the MKP-1 mRNA levels in the prefrontal and parietal cortices and thalamus compared with treatment with SAL (SAL-SAL group). Pre-treatment with MK-801 attenuated the enhanced MKP-1 mRNA levels induced by acute METH administration in the orbital, prefrontal and temporal cortex and striatum. However, MK-801 did not attenuate those in the parietal cortex and thalamus. Inhibition rates were 32.4% in the orbital cortex, 26.1% in the prefrontal cortex, 54.1% in the temporal cortex, and 76.8% in the striatum, calculated as:

[(SAL-METH)−(MK-METH)]/[(SAL-METH)−(SAL-SAL)]×100 (%).

Effect of pre-treatment with SCH23390 and MK-801 on MKP-3 mRNA

The effect of pre-treatment with SCH23390 on acute METH-induced MKP-3 mRNA expression is shown in Fig. 5(a). There were significant differences among the MKP-3 mRNA levels of the four groups (SAL-SAL, SAL-METH, SCH-SAL, SCH-METH) for each brain region, as follows: orbital cortex (F3,24 = 19.77, p = 0.0001); prefrontal cortex (F3,24 = 15.11, p = 0.0001); parietal cortex (F3,24 = 23.73, p = 0.001); temporal cortex (F3,24 = 19.27, p = 0.001); striatum (F3,24 = 5.93, p = 0.042); hippocampus CA1 (F3,24 = 4.67, p = 0.0104); and hippocampus CA3 (F3,24 = 5.79, p = 0.004). Treatment with SCH23390 alone (SCH-SAL group) did not alter the MKP-3 mRNA levels in any brain region compared with treatment with SAL (SAL-SAL group). Pre-treatment with SCH23390 almost completely attenuated the enhanced MKP-3 mRNA levels induced by acute METH administration. Inhibition rates were 64.7% in the orbital cortex, 74.1% in the prefrontal cortex, 73.9% in the parietal cortex, 66.6% in the temporal cortex, 137.7% in the striatum, 91.2% in the hippocampus CA1, and 132.3% in the hippocampus CA3.

Figure 5.

Effects of pre-treatment with SCH23390 (0.5 mg/kg) (a) and MK-801 (0.25 mg/kg) (b) on MKP-3 mRNA expression observed after acute METH (4 mg/kg) injection. Rats were pre-treated with saline, SCH23390 or MK-801 30 min before administration of either saline or METH and killed 1 h later (n = 6 or 8 in each group). The values are expressed as the mean ± SEM. The level of statistical significance was determined by one-way anova followed by Fisher's PSLD. *p < 0.05, **p < 0.01, ***p < 0.001. METH, methamphetamine.

The effect of pre-treatment with MK-801 on acute METH-induced MKP-3 mRNA is shown in Fig. 5(b). There were significant differences among the MKP-3 mRNA levels of the four groups (SAL-SAL, SAL-METH, MK-SAL, MK-METH) for each brain region examined, as follows: orbital cortex (F3,24 = 21.79, p = 0.0001); prefrontal cortex (F3,24 = 12.43, p = 0.0001); parietal cortex (F3,24 = 21.18, p = 0.0001); temporal cortex (F3,24 = 13.51, p = 0.0001); striatum (F3,24 = 3.46, p = 0.032). Pre-treatment with MK-801 attenuated the enhanced MKP-3 mRNA levels induced by acute METH administration in the striatum and in all cortices except the temporal cortex, partially in the temporal cortex, but not in the hippocampus CA1 or CA3. Treatment with MK-801 alone (MK-SAL group) did not alter the MKP-3 mRNA levels in any brain region compared with treatment with SAL (SAL-SAL group). Inhibition rates were 69.0% in the orbital cortex, 63.3% in the prefrontal cortex, 70.1% in the parietal cortex, 51.6% in the temporal cortex and 109.7% in the striatum.

Effect of chronic METH administration on MKP-1 and MKP-3 mRNAs

The effect of chronic METH treatment for 10 days on MKP-1 and MKP-3 mRNA levels is shown in Fig. 6. There were significant differences among the MKP-1 mRNA levels of the three groups for each brain region examined, as follows: orbital cortex (F2,21 = 28.44, p = 0.0001); prefrontal cortex (F2,22 = 23.16, p = 0.0001); parietal cortex (F2,22 = 20.52, p = 0.0001); temporal cortex (F2,22 = 24.87, p = 0.0001); thalamus (F2,22 = 10.13, p = 0.0008). No significant difference was observed in the striatum.

Figure 6.

Effect of chronic METH administration on MKP-1 and MKP-3 mRNAs levels (n = 8 in each group). Rats were treated with either 4 mg/kg METH or an equal volume of saline once daily for 10 days. The values are expressed as the mean ± SEM. The level of statistical significance was determined by one-way anova followed by Fisher's PSLD. *p < 0.05, **p < 0.01, ***p < 0.001. METH 1 h, 1 h after the last injection of chronic methamphetamine administration; METH 24 h, 24 h after the last injection of chronic methamphetamine administration; SAL 1 h, 1 h after the last injection of chronic saline administration.

On the other hand, there were significant differences among the MKP-3 mRNA levels only in the hippocampus CA1 (F2,22 = 18.99, p = 0.0001).

The increase in levels of expression of MKP-1 and MKP-3 mRNAs 1 h after chronic METH administration compared with those after acute METH administration is shown in Table 1. Change in expression levels 1 h after acute METH administration was calculated against time 0-values, and that 1 h after the last injection of chronic METH administration was calculated against the values 1 h after the last injection of chronic SAL administration. After chronic METH administration the increased expression levels of MKP-1 mRNA were less than after acute administration. However, its increase in the frontal cortices of orbital and prefrontal regions was still over 50%. MKP-3 mRNA levels increased in the all the cortices, the striatum and hippocampus after acute METH administration; however, it increased only in the hippocampus CA1 after chronic METH administration, although its increased levels were similar or more to those observed after acute METH administration.

Table 1.  Summary of MKP-1 and MKP-3 mRNA levels after acute and chronic methamphetamine administration
 acutechronic
 MKP-1MKP-3MKP-1MKP-3
  1. Change in levels 1 h after acute methamphetamine administration was calculated against time 0-values, and that 1 h after the last injection of chronic methamphetamine administration was calculated against the values 1 h after the last injection of chronic saline administration. The values are expressed as percentages. The level of statistical significance was determined by one-way anova followed by Fisher's PSLD. *p < 0.05, **p < 0.01, ***p < 0.001.

orbital cortex301.2%***50.4%***59.9%***12.3%
prefrontal cortex194.6%***45.4%***67.1%***20.5%
parietal cortex129.2%***46.4%***43.6%***−4.8%
temporal cortex131.4%***41.7%***42.6%***41.7%
thalamus88.5%**10.4%27.5%**10.4%
striatum47.2%*18.3%**11.6%11.2%
hippocampus CA137.0%***50.0%***
hippocampus CA334.3%***20.5%

Effect of acute and chronic METH administration on MKP-1 and MKP-3 protein levels

The effect of acute and chronic METH administration on MKP-1 protein levels is shown in Fig. 7. Western blot analysis revealed that the specific band for MKP-1 protein appeared at 38 kDa 3 h after acute METH administration in the prefrontal cortex and striatum, but not in the hippocampus and it appeared 3 h after chronic METH administration in the prefrontal cortex, but not in the striatum and hippocampus. The five fold excess of the blocking peptide for MKP-1 protein completely abolished the specific band (data not shown). The specific band for MKP-3 protein could not be detected in any of the three brain regions 1 h or 3 h after acute and chronic METH or SAL administration, because of the sensitivity of the assay.

Figure 7.

Effect of acute and chronic METH administration on MKP-1 protein levels. Rats were decapitated 3 h after acute administration of 4 mg/kg METH or equal volume of saline (upper panel), and 3 h after the last injection of chronic METH or saline administration for 14 days (lower panel). Arrow indicates a 38-kDa band of MKP-1 and 45 kDa of molecular weight standard band in western blot analysis. Cx, prefrontal cortex; Hippo, hippocampus; Str, striatum.

Discussion

Our findings are the first showing an increase in the levels of MKP-1 mRNA in various cortices, the striatum and thalamus and the levels of MKP-3 mRNA in the cortices, striatum and hippocampus after acute METH administration in the rat brain. There is a very close connection between MAP kinases and MKPs. MAP kinases comprise two major subtypes; extracellular signal-regulated kinase (ERK), which induces cell growth and proliferation and stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK) and p38 kinase, which induce apoptosis, stress reaction and gene transcription. Recent studies showed that ERK is also involved in gene transcription, including in the striatum, in response to psychostimulants or electrical stimulation of corticostriatal pathway in vivo (Sgambato et al. 1998; Valjent et al. 2000). MKPs dephosphorylate and inactive their specific MAP kinases. MKP-1 is selective for inactivating ERK or SAPK/JNK and p38 kinases (Brondello et al. 1997; Franklin and Kraft 1997), while MKP-3 is selective for inactivation of only ERK (Muda et al. 1996a, 1996b; Hafen 1998). MAP kinases in turn induce or activate only specific MKPs. ERK or SAPK directly induce MKP-1 in vitro (Bokemeyer et al. 1996; Brondello et al. 1997), and in vivo (Sgambato et al. 1998). ERK also induces MKP-3 (Muda et al. 1996a; Hafen 1998) and acts as the catalytic activator of MKP-3 (Camps et al. 1998; Muda et al. 1998). Therefore, increased MKP-1 mRNA levels after acute METH administration in several cortices, the striatum and thalamus may indicate the activation of ERK, SAPK/JNK and p38 in those regions, while increased MKP-3 mRNA levels after acute METH administration in the hippocampus CA1 and CA3 may indicate the activation of ERK alone. Several lines of evidences indicate that MAP kinases and MKPs are involved in psychostimulants-induced behavioral sensitization. Chronic administration of morphine or cocaine increases the state of phosphorylation of ERK in the ventral tegmental area (VTA; Berhow et al. 1996). Neurotrophin-3, a nerve growth factor, contributes to the initiation of behavioral sensitization to cocaine by activating the MAP kinase (Pierce et al. 1999). Cocaine induces the expression of hVH-5 mRNA, another MAP kinase phosphatase, in the accumbens, striatum and hippocampus (Thiriet et al. 1998). Acute cocaine administration induces phosphorylation of ERK in the striatum via D1 dopamine and NMDA receptors (Valjent et al. 2000). Our current and previous findings suggest that MKP families and the MAP kinase cascade play important roles in the neural plastic changes involved in psychostimulant-induced behavioral sensitization.

The important findings obtained by our pharmacological blockade experiments can be summarized as follows: (1) the enhanced expression of MKP-1 mRNA induced by acute METH administration in the orbital cortex, prefrontal cortex, temporal cortex and striatum increased via the activation of D1 dopamine and NMDA receptors, whereas that in the parietal cortex and thalamus increased only via the activation of D1 dopamine receptors; (2) the enhanced expression of MKP-3 mRNA in the cortices and striatum induced by acute METH administration increased via the activation of D1 dopamine and NMDA receptors, whereas that in the hippocampus increased only via the activation of D1 dopamine receptors. However, there are few D1 dopamine receptors in the hippocampus, whereas D5 dopamine receptors are relatively abundant (Meador-Woodruff et al. 1992; Ritter and Meador-Woodruff 1997). SCH23390, used as a D1 antagonist in the present study, has an affinity with not only D1 but also D5 dopamine receptors (Sunahara et al. 1991). Accordingly, it is possible that METH-induced MKP-3 mRNA increase in the hippocampus is mediated by the activation of D5 dopamine receptors. After METH administration, dopamine release increases in the striatum and frontal cortex and glutamate release increases in the striatum shown using in vivo dialysis (Maisonneuve et al. 1990; Abekawa et al. 1994; Nishijima et al. 1996). Recently, it was reported that glutamate stimulates SAPK, and ERK in the striatum or ERK in primary cortical neurons (Xia et al. 1996; Schwarzschild et al. 1997; Vanhoutte et al. 1999). Although in vitro study showed that D1 dopamine receptor agonists induce activation of p38 and SAPK in neuroblastoma cells in vitro (Zhen et al. 1998), Schwarzschild et al. (1997) failed to replicate it in vivo study. Recently, D1 dopamine agonists have been shown to stimulate ERK in vivo and vitro (Valjent et al. 2000; Zanassi et al. 2001). Therefore, it was speculated that METH enhanced dopamine release in the striatum and cortices, which activated ERK via D1 dopamine receptors and then activated MKP-1 and MKP-3, and that METH also enhanced glutamate release in the striatum and cortices, which activated ERK and SAPK via NMDA receptors and then activated MKP-1 and MKP-3.

The regional and laminar distributions of D1 dopamine receptors and their mRNAs in the frontal cortex have been well documented. D1 dopamine receptors are known to be preferentially distributed in layer VI, consistent with the known distribution of mesocortical dopaminergic projections (Vincent et al. 1993). Microscopic examination showed that MKP-1 mRNA was enriched in layer IV and VI of the cerebral cortex, which strikingly resembled the previous report of the Arc mRNA expression after METH administration (Kodama et al. 1998). The increase in dopamine release in the prefrontal cortex of at least layer VI after acute METH administration may directly trigger enhancement of cerebral cortical MKP-1 mRNA expression in a manner dependent on D1 dopamine receptor stimulation.

Compared with the comparatively widespread and robust increase in MKP-1 mRNA levels seen in acute METH administration, MKP-1 mRNA levels after chronic METH administration increased moderately and in reduced brain areas. However, over 50% increases were still seen in the frontal cortices such as the orbital and prefrontal cortices. The protein levels of MKP-1 also increased after chronic METH administration in the prefrontal cortex. The increase of MKP-3 mRNA after chronic METH administration was restricted in the hippocampus CA1, but the levels were similar or more to those seen after acute administration. These differences in extent and distribution in the increase of MKP-1 and MKP-3 mRNAs after acute and chronic METH administration indicate distinct roles for these MKPs in different phases of METH-induced behavioral sensitization. The behavioral sensitization phenomenon is considered to consist of two distinct major processes, the early induction process and the later maintenance or expression process. The sensitization phenomenon has been shown to occur after chronic treatment with psychostimulants, and neurochemical changes of increased dopamine efflux and morphological changes in dendrites and spines in the accumbens and frontal cortex were also demonstrated after chronic psychostimulant treatment (Robinson and Kolb 1997). However, recent studies showed that even a single exposure to amphetamine also induced long-term behavioral and neurochemical sensitization (Vanderschuren et al. 1999). Neural adaptation for lasting sensitization such as synaptogenesis and neurite elongation may therefore begin at the first psychostimulant administration. Our findings suggest that the strong activation of MKP-1 in the cortices, striatum and thalamus, and MKP-3 in the cortices, striatum and hippocampus seen after a single METH administration, contribute to the induction process of METH-induced behavioral sensitization. Activation of MKP-1 in the cortices and thalamus and MKP-3 in the hippocampus CA1 seen after chronic METH administration, contributes to the subsequent maintenance process. Taking the substrate specificity of MKP-1 and MKP-3 into consideration, it could be suggested that in the plasticity response for the early induction of sensitization, all three types of MAP kinases, ERK, SAPK/JNK and p38 kinase, may be strongly activated in widespread brain regions, while in the later maintenance phase that results in lasting sensitization, MAP kinases may be activated only in a restricted number of brain regions such as the frontal cortex and hippocampus. However, further studies will be needed to clarify the precise roles of MKP and MAP kinase cascades for the plasticity underlying psychostimulant-induced behavioral sensitization.

Acknowledgement

This study was supported in part by a grant from the Zikei Institute of Psychiatry.

Ancillary