INTRODUCTION

Psychostimulants such as cocaine and amphetamine produce changes in gene regulation in striatal projection neurons (Harlan and Garcia, 1998; Torres and Horowitz, 1999). These molecular alterations are thought to result in functional changes in the affected basal ganglia-cortical circuits and are implicated in drug-induced behavioral changes (Berke and Hyman, 2000; Graybiel et al, 2000). Psychostimulant-induced gene regulation in the striatum is related to excessive stimulation of dopamine receptors (Hyman and Malenka, 2001; Nestler, 2001). Methylphenidate (Ritalin) is a psychostimulant that is widely used in the treatment of attention-deficit hyperactivity disorder in adolescents and adults, and is also diverted for recreational purposes (Kollins et al, 2001; Swanson and Volkow, 2003). Methylphenidate binds and inhibits the dopamine transporter (Schweri et al, 1985; Gatley et al, 1996) and produces dopamine overflow in the striatum (Hurd and Ungerstedt, 1989; Butcher et al, 1991; Kuczenski and Segal, 1997; Gerasimov et al, 2000; Volkow et al, 2001). In addition, this psychostimulant inhibits norepinephrine reuptake (Kuczenski and Segal, 1997) and affects vesicular monoamine transport (Sandoval et al, 2002). Although relatively little is known regarding methylphenidate's propensity to induce neuronal alterations, a number of studies demonstrated that repeated methylphenidate treatment in animal models causes behavioral changes similar to other psychostimulants. These include increased psychomotor activation by a subsequent psychostimulant challenge (behavioral sensitization) (eg Shuster et al, 1982; Gaytan et al, 1997; Crawford et al, 1998; McDougall et al, 1999; Brandon et al, 2001), place preference conditioning (Martin-Iverson et al, 1985; Meririnne et al, 2001; Andersen et al, 2002), enhanced cocaine self-administration (Brandon et al, 2001; Schenk and Izenwasser, 2002), as well as several other behavioral changes (Bolanos et al, 2003; Carlezon et al, 2003). These behavioral effects indicate that methylphenidate can produce enduring neuronal changes.

In a recent study, we investigated whether a methylphenidate treatment regimen that resulted in psychomotor sensitization and enhanced cocaine self-administration in adolescent rats (Brandon et al, 2001) would alter gene regulation in striatal neurons. Our results showed that repeated treatment with 10 mg/kg once daily for 7 days produced increased expression of the neuropeptide dynorphin and attenuated the inducibility of immediate-early genes (c-fos, zif 268) and substance P in the striatum (Brandon and Steiner, 2003). These effects were principally similar to those of repeated cocaine and amphetamine treatments (see Brandon and Steiner, 2003; Willuhn et al, 2003, for discussion). Substance P and dynorphin are contained in neurons of the direct striatal output pathway (striatonigral neurons) (Gerfen and Young, 1988; Graybiel, 1990; Gerfen and Wilson, 1996), and our results thus suggested that repeated methylphenidate treatment produces molecular alterations in the direct pathway.

In the present study, we further characterized molecular changes in the striatum produced by methylphenidate. We investigated effects of acute methylphenidate treatment in adult rats for comparison with the wealth of findings on acute molecular effects of psychostimulants such as cocaine and amphetamine (Harlan and Garcia, 1998; Torres and Horowitz, 1999). Acute effects are often predictive of long-term molecular changes in terms of location and magnitude (eg Brandon and Steiner, 2003; Willuhn et al, 2003). Of the few studies that have examined effects of acute methylphenidate on striatal gene expression, most focused on c-fos induction (Lin et al, 1996; Acheson et al, 2001; Penner et al, 2002; Brandon and Steiner, 2003; Chase et al, 2003). Acute effects on the expression of neuropeptides in the striatum have so far not been investigated. We determined and compared dose–response relationships and time courses for changes in the expression of c-fos and the neuropeptides substance P and dynorphin (direct pathway), and enkephalin (indirect pathway) after acute methylphenidate administration. Moreover, in order to assess whether such molecular alterations are associated with particular cortico-basal ganglia-cortical circuits (functional domains), we mapped the topography of these effects by using 23 sampling areas (sectors) defined mostly by their predominant cortical inputs (Willuhn et al, 2003). These results were compared with those of our recent study on the topography of cocaine-induced gene regulation in the striatum (Willuhn et al, 2003). Preliminary findings of this work were presented in abstract form (Yano and Steiner, 2003).

MATERIALS AND METHODS

Subjects

Male Sprague–Dawley rats (210–300 g) were housed two per cage under standard laboratory conditions (12 h light:12 h dark cycle; lights on at 0700). Food and water were available ad libitum. The rats were allowed 1 week to acclimate, during which they were handled repeatedly. All procedures met the NIH guidelines for the care and use of laboratory animals and were approved by the Rosalind Franklin University Animal Care and Use Committee.

Drugs and Injection Procedures

Drug treatments were performed between 1300 and 1700. Rats were kept in their home cage between drug injection and killing. In the dose–response study, rats received an intraperitoneal injection of 0 (vehicle), 0.5, 2, 5, or 10 mg/kg of methylphenidate hydrochloride (generously provided by the National Institute on Drug Abuse; in 0.02% ascorbic acid, 1 ml/kg; n=6 each) and were killed 1 h later. In the time-course study, 5 mg/kg of methylphenidate was administered, and the rats were killed either immediately (0 min group), or 20, 40 min, 1, 3, or 24 h after the injection (n=5–6 each).

Tissue Preparation and In Situ Hybridization Histochemistry

The rats were killed with CO2. The brain was rapidly removed and frozen in isopentane cooled on dry ice. Brains were stored at −20°C until they were sectioned on a cryostat. Coronal sections (12 μm) were thaw-mounted onto glass slides (Superfrost/Plus, Daigger, Wheeling, IL), dried on a slide warmer, and stored at −20°C. Tissue sections were fixed in 4% paraformaldehyde/0.9% saline for 10 min at room temperature, incubated in a fresh solution of 0.25% acetic anhydride in 0.1 M triethanolamine/0.9% saline (pH 8.0) for 10 min, dehydrated, defatted for 2 × 5 min in chloroform, rehydrated, and air-dried. The slides were then stored at –30°C until hybridization.

Oligonucleotide probes (48-mers; Life Technologies, Rockville, MD) were labeled with [35S]dATP (Steiner and Gerfen, 1993). The probes had the following sequence: c-fos, complementary to bases 207–254, GenBank Accession number X06769; dynorphin, bases 862–909, M10088; enkephalin, bases 436–483, M28263; substance P, bases 128–175, X56306. Labeled probe (∼3 × 106 c.p.m.) in 100 μl of hybridization buffer was added to each slide (Steiner and Kitai, 2000). The sections were coverslipped and incubated at 37°C overnight. After incubation, the slides were first rinsed in four washes of 1 × saline citrate (150 mM sodium chloride, 15 mM sodium citrate). They were then washed 3 × 20 min each in 2 × saline citrate/50% formamide at 40°C, followed by two washes of 30 min each in 1 × saline citrate at room temperature. After a brief water rinse, the sections were air-dried and then apposed to X-ray film (BioMax MR-2, Kodak). The film exposure times were on average 4 (enkephalin, substance P), 9 (dynorphin), and 14 days (c-fos).

Analysis of Autoradiograms

Gene expression in the striatum was assessed on three rostrocaudal levels (one section each) (Figure 1): ‘Rostral’ corresponds to approximately +1.6 mm relative to bregma (Paxinos and Watson, 1998), ‘middle’ to +0.4 mm, and ‘caudal’ to −0.8 mm. The rostral striatum was divided into 10, the middle into six, and the caudal into seven sectors, for a total of 18 sectors representing the caudate-putamen and five the nucleus accumbens (Figure 1). These sectors are mostly defined by their predominant cortical inputs (see Willuhn et al, 2003, for a more detailed discussion). The design of these sectors is based on findings of numerous tract tracing (eg Beckstead, 1979; Donoghue and Herkenham, 1986; McGeorge and Faull, 1987, 1989; Reep et al, 1987; Berendse et al, 1992; Brog et al, 1993; Kincaid and Wilson, 1996; Wright et al, 1999; Reep and Corwin, 1999; Hoffer and Alloway, 2001; Reep et al, 2003) and functional mapping studies (eg Carelli and West, 1991; Brown and Sharp, 1995; Brown et al, 1998). However, their outlines are simplified in many ways in order to facilitate reliable sampling, and thus do not describe exact or exclusive target areas of the connections depicted in Figure 1. Moreover, in Figure 1, some corticostriatal connections are omitted for simplicity, including visual cortical input to the dorsal striatum (eg Reep et al, 2003), projections from the piriform and entorhinal cortex to the ventral and medial striatum (eg McGeorge and Faull, 1989; Brog et al, 1993), converging corticostriatal projections from functionally related cortical areas on different rostrocaudal levels, as well as all interhemispheric projections (eg McGeorge and Faull, 1987, 1989; Wright et al, 2001).

Figure 1
figure 1

Schematic illustration of the 23 striatal sectors used for sampling gene expression. Gene regulation was assessed in coronal sections from three rostrocaudal levels (‘rostral’, approximately at +1.6 mm relative to bregma (Paxinos and Watson, 1998); ‘middle’, +0.4; and ‘caudal’, −0.8) in a total of 18 sectors representing the caudate-putamen and five sectors representing the nucleus accumbens. These sectors were mostly defined by their predominant cortical inputs, as determined in tract tracing and functional mapping studies (see Materials and methods). These inputs are indicated by arrows. Caudate-putamen: m, medial; dm, dorsomedial; d, dorsal; dl, dorsolateral; vl, ventrolateral; v, ventral; c, central; vc, ventral central; dc, dorsal central; nucleus accumbens: mC, medial core; lC, lateral core; mS, medial shell; vS, ventral shell; lS, lateral shell; cortex: AC, anterior cingulate; M2, medial agranular; M1, motor; SS, somatosensory; I, insular; LO, lateral orbital; IL, infralimbic; PL, prelimbic.

Full size image

Gene expression was measured by densitometry on film autoradiograms using a Macintosh-based image analysis system (NIH Image, Wayne Rasband, NIMH). The films were captured using a light table (Northern Light, Imaging Research, St. Catharines, Ontario, Canada) and a Sony CCD camera (Imaging Research). The ‘mean density’ value of a sector was measured by placing a template over the captured image. The mean densities of corresponding sectors in the two hemispheres were averaged after correcting for background by subtracting mean density measured over white matter (corpus callosum). Treatment effects were determined by one-factor ANOVA, followed by two-tailed Dunnett's post hoc tests (Superanova software). For correlation analyses and illustrations of topographies, the increase in gene expression in a given sector was expressed as the percentage of the maximal increase (% of max.) in the 23 striatal sectors, for a particular probe. The illustrations of film autoradiograms are computer-generated images, and are contrast-enhanced when necessary. Maximal hybridization signal is black.

RESULTS

Dose-Dependent Induction of c-fos and Substance P by Acute Methylphenidate

Acute injection of methylphenidate altered striatal gene expression in a dose-dependent and regionally selective manner (Figures 2 and 3, Table 1). At 1 h after drug administration, statistically significant increases in mRNA levels were found for 5 and 10 mg/kg, but not for 0.5–2 mg/kg. After 10 mg/kg, c-fos expression was induced in 14 of the 18 sectors of the caudate-putamen (5 mg/kg: 10/18; Table 1a), but in none of the five sectors of the nucleus accumbens (Figure 3). After 2 mg/kg, there was a minor c-fos response in some of the animals (not statistically significant as a group; Table 1a). Induction of c-fos was most pronounced in the dorsal sectors, especially at middle to caudal striatal levels (Figure 2), and was generally weaker in the medial and lateral sectors, and weakest or absent in the ventral sectors (Table 1a; see also Figures 4 and 7).

Figure 2
figure 2

Dose-dependent effects of methylphenidate on c-fos (a), substance P (b), dynorphin (c), and enkephalin (d) expression in the dorsal caudate-putamen. Mean density values (mean±SEM, arbitrary units) measured in the dorsal sector on rostral, middle, and caudal striatal levels are shown for rats treated with 0–10 mg/kg (i.p.) of methylphenidate and killed 1 h later. **P<0.01, *P<0.05 vs 0 mg/kg.

Full size image
Figure 3
figure 3

Effects of methylphenidate on c-fos (a) and substance P (b) expression in the nucleus accumbens. Mean density values (mean±SEM) measured in the medial and lateral core, and in the medial, ventral, and lateral shell are given for rats treated with 0–10 mg/kg of methylphenidate and killed 1 h later.

Full size image
Table 1 Dose-Dependent Effects of Methylphenidate on the Expression of (a) c-fos and (b) Substance P in the Caudate-Putamen
Full size table
Figure 4
figure 4

Methylphenidate-induced expression of c-fos (a) and substance P (b) in the rostral, middle, and caudal striatum. Illustrations of film autoradiograms depict gene expression at 0 and 40 min (c-fos) or 60 min (substance P) after methylphenidate injection (5 mg/kg). Maximal hybridization signal is black.

Full size image
Figure 7
figure 7

Topography of methylphenidate-induced changes in the expression of c-fos (a), substance P (b), dynorphin (c), and enkephalin (d) in the 23 striatal sectors. The increases in gene expression (P<0.05 vs basal expression, 0 min) on rostral, middle, and caudal striatal levels are depicted for rats that received an injection of methylphenidate (5 mg/kg) and were killed 20, 40 min, 1, 3, or 24 h later. These values are expressed as the percentage of the maximal increase (% of max.) in the 23 sectors for each gene, and are coded as indicated for c-fos and substance P, or are shaded for dynorphin and enkephalin.

Full size image

Changes in substance P expression displayed similar characteristics (Figures 2 and 3, Table 1b). At 1 h after injection of 10 mg/kg, substance P mRNA levels were significantly increased in 13 of 18 caudate-putamen sectors (5 mg/kg: 2/18; Table 1b); no changes were seen in the accumbal sectors (Figure 3). The increase in substance P expression was also most robust in the dorsal sectors of the middle and caudal striatum (Figure 2) and weakest or absent in the ventral sectors (see also Figures 4 and 7).

In contrast to c-fos and substance P, dynorphin mRNA levels were not significantly changed by methylphenidate (Figure 2), with the exception of borderline increases (P<0.05) after 10 mg/kg in the dorsolateral sector on the rostral level and the ventrolateral on the caudal level (data not shown). Similarly, no statistically significant changes in enkephalin expression were found, either in the dorsal sectors (Figure 2) or in any other sector (data not shown).

Time Course of Changes in Gene Expression after Acute Methylphenidate Administration

The time courses for changes in c-fos and neuropeptide expression after injection of 5 mg/kg methylphenidate are depicted in Figures 5, 6 and 7Figure 7 presents regional maps for the different time points and genes. In all, 14 of 18 caudate-putamen sectors (Figure 7, Table 2a) and one of five nucleus accumbens sectors (Figures 6 and 7) displayed a significant c-fos response. C-fos mRNA levels were typically increased by 20 min after drug administration (12 of 15 responding sectors; Figure 5, Table 2a), peaked at 40 min in all 15 sectors (Figure 4), were in most cases still increased at 1 h (12 of 15 sectors), and returned to baseline by 3 h in all sectors. Overall, the c-fos response was again most robust in the dorsal sectors of the middle and caudal striatum (Figures 4, 5 and 7, Table 2a), weaker rostrally, and weakest or absent in ventral sectors. The only significant c-fos response in the nucleus accumbens was found in the lateral part of the shell at 40 min after drug injection (Figures 6 and 7).

Figure 5
figure 5

Time course of methylphenidate-induced expression of c-fos (a), substance P (b), dynorphin (c), and enkephalin (d) in the dorsal caudate-putamen. Mean density values (mean±SEM) measured in the dorsal sector on rostral, middle, and caudal striatal levels are shown for rats that received 5 mg/kg of methylphenidate and were killed either immediately (0 min) or at 20, 40 min, 1, 3, or 24 h after drug administration. **P<0.01, *P<0.05 vs 0 min.

Full size image
Figure 6
figure 6

Time course of methylphenidate-induced expression of c-fos (a) and substance P (b) in the nucleus accumbens. Mean density values (mean±SEM) measured in the medial and lateral core, and in the medial, ventral, and lateral shell are given for rats that received 5 mg/kg of methylphenidate and were killed either immediately (0 min) or at 20, 40 min, 1, 3, or 24 h after drug administration. **P<0.01 vs 0 min.

Full size image
Table 2 Time Course of Methylphenidate-Induced Changes in (a) c-fos and (b) Substance P Expression in the Caudate-Putamen
Full size table

Of the 18, 13 caudate-putamen sectors showed a significant change in substance P expression after administration of 5 mg/kg methylphenidate (Figures 4 and 7, Table 2b). Again, no changes were found in the nucleus accumbens (Figures 6 and 7). Substance P expression changed with a delayed time course compared to c-fos expression (Figures 5 and 7, Table 2b). Substance P mRNA levels were significantly increased by 20 min only in five of the most responsive sectors, among them the three dorsal sectors (Figure 5). Levels peaked at 40–60 min in all 13 sectors (Figure 4), were in most cases still increased at 3 h (eight of 13), and returned to baseline by 24 h in all sectors. Interestingly, the increase in substance P mRNA levels peaked earlier in caudal sectors (40 min) than in rostral sectors (around 1 h) (Figures 5 and 7).

Dynorphin expression was minimally affected by methylphenidate (Figure 7). Levels of mRNA displayed a statistically significant increase only in two of the 23 sectors, in the medial (P<0.05; data not shown) and dorsal (Figure 5) sectors on the middle level at 1 h after drug administration. Similarly, minimal effects were found for enkephalin expression. On the middle level, a borderline increase was seen in the dorsal sector at 20 min (Figure 5), whereas the ventral sector showed a decrease (P<0.05; data not shown) after 24 h. No other effects were found for dynorphin or enkephalin expression (not shown).

Correlations between Changes in c-fos and Substance P Expression

The relationship between the changes in c-fos and substance P expression across the 23 striatal sectors was further characterized by correlation analyses. For these correlations, the difference between a treatment group (dose or time point) and the control (0 mg/kg or 0 min, respectively) was calculated for a given sector and expressed as the percentage of the maximal difference found in the 23 sectors for all doses, or time points (Figure 8). In the dose–response study, the changes in c-fos and substance P expression were compared for each methylphenidate dose (Figure 8a). After injection of 0.5 mg/kg, c-fos and substance P values were not correlated (r=0.216; P>0.05). In contrast, a significant positive correlation between the changes in c-fos and substance P mRNA levels was found for 2 mg/kg (r=0.562; P<0.01), indicating that this dose produced changes in the expression of both genes in at least some of the striatal sectors. Highly significant correlations between c-fos and substance P expression were found for 5 mg/kg (r=0.781; P<0.001) and 10 mg/kg (r=0.905; P<0.001) (Figure 8a).

Figure 8
figure 8

Correlations between the methylphenidate-induced increases in c-fos and substance P expression in the dose–response study (a) and in the time-course study (b). The scatterplots depict the relationship between increases in c-fos and substance P mRNA levels (dose or time point minus 0 mg/kg or 0 min, respectively) in the individual sectors, for each dose (0.5–10 mg/kg) and the time points 20, 40, 60 min and 24 h. The values are expressed as the percentage of the maximal increase (% of max.) in the 23 sectors for each gene. ***P<0.001, **P<0.01.

Full size image

For the time-course study, changes in c-fos and substance P expression were compared for each time point. Highly significant positive correlations were found for 20 min (r=0.879; P<0.001), 40 min (r=0.960; P<0.001), and 60 min (r=0.850; P<0.001) (Figure 8b). In contrast, 24 h after drug administration, a significant negative correlation between changes in c-fos and substance P mRNA levels was observed (r=−0.572; P<0.01) (Figure 8b). Further analysis indicated that this effect was largely due to substance P mRNA levels in maximally responding sectors (40 min) that were below baseline at 24 h (Figure 8b). This was confirmed by significant negative correlations for substance P mRNA levels at 24 h vs substance P mRNA levels at 40 min (r=−0.599, P<0.01), as well as vs c-fos mRNA levels at 40 min (r=−0.592, P<0.01). These results indicate suppressed substance P expression in the most responsive (dorsal) sectors 24 h after acute methylphenidate administration. In contrast, no relationships between acute drug responses and enkephalin mRNA levels at 24 h after injection were seen (enkephalin (24 h) vs c-fos (40 min), r=−0.158, P>0.05; vs substance P (40 min), r=−0.127, P>0.05).

DISCUSSION

Psychostimulants that produce dopamine overflow in the striatum, including cocaine and amphetamine, alter gene regulation in striatal output pathways. Our previous results in adolescent rats (Brandon and Steiner, 2003) and present findings in adults demonstrate that the psychostimulant methylphenidate is no exception, but differs in certain molecular effects from other psychostimulants. First, we show that acute administration of methylphenidate increased the expression of the transcription factor (immediate-early gene) c-fos and the neuropeptide substance P in a dose-dependent and regionally distinct manner. These effects emerged with 2 mg/kg and were very robust after 5–10 mg/kg. For both genes, mRNA levels were elevated within 20 min of drug administration and remained increased for more than 1 (c-fos) or 3 h (substance P). These molecular changes displayed a distinct topography, with maximal increases in dorsal, sensorimotor sectors on middle to caudal striatal levels, and lesser or no effects in rostral and ventral striatal sectors. Second, in contrast to these prominent effects on c-fos and substance P, and unlike the effects of cocaine and amphetamine, acute methylphenidate produced only minimal changes in the expression of the opioid peptides dynorphin and enkephalin. Overall, our results indicate that methylphenidate treatment causes neuroadaptations in neurons of the direct striatal output pathway.

METHYLPHENIDATE AFFECTS TRANSCRIPTION FACTORS

Our present results and those of others demonstrate that methylphenidate alters the expression of transcription factors in the striatum in a manner similar to other psychostimulants (Harlan and Garcia, 1998; Torres and Horowitz, 1999). Several studies reported acute effects of methylphenidate on c-fos expression. Increased Fos protein levels were shown in different species for a range of doses and routes of administration (cat, 2.5 mg/kg, p.o. (Lin et al, 1996); mouse, 5 and 20 mg/kg, s.c. (Penner et al, 2002); 30 mg/kg, i.p. (Trinh et al, 2003); rat, 2–20 mg/kg, s.c. (Chase et al, 2003)). Previously, we demonstrated increased c-fos and zif 268 mRNA levels after 2–10 mg/kg (i.p.) in adolescent rats (Brandon and Steiner, 2003). In our present study, we assessed the effects of 0.5–10 mg/kg (i.p.) in adults and also found increased c-fos expression after 2 mg/kg and higher doses. While 5 and 10 mg/kg produced pronounced effects, increased gene expression after 2 mg/kg was revealed only by a more powerful two-marker correlation analysis (c-fos vs substance P). A similar effect was found for zif 268 and Homer 1a with this dose (Yano and Steiner, 2005). Thus, this threshold dose of 2 mg/kg appeared to produce more robust changes in adolescents (Brandon and Steiner, 2003; Chase et al, 2003) than in adults. A more robust acute molecular response in adolescents (see also Andersen et al, 2001) would be consistent with greater neuroadaptive changes after repeated drug treatment in this age group (Ehrlich et al, 2002). For example, our previous study showed significantly increased dynorphin expression after repeated methylphenidate treatment in adolescents (Brandon and Steiner, 2003). However, such possible age differences will need to be affirmed by direct comparisons.

In addition to the above dose–response relationship, our study is the first to describe the time course of changes in c-fos expression after acute methylphenidate administration. Levels of c-fos mRNA were significantly increased by 20 min, peaked around 40 min, and returned to baseline between 1 and 3 h after drug injection. This time course is very similar to that of c-fos expression after acute cocaine administration (eg Steiner and Gerfen, 1998).

Previous studies also assessed the effects of repeated methylphenidate treatment on immediate-early gene expression. It was found that the inducibility of c-fos and zif 268 by cocaine (Brandon and Steiner, 2003) or by methylphenidate (Chase et al, 2003) was suppressed after 7- or 14-day repeated methylphenidate treatments. These effects are also principally similar to those of repeated cocaine (eg Willuhn et al, 2003) and amphetamine treatments (for reviews, see Harlan and Garcia, 1998; Torres and Horowitz, 1999). Such blunted gene induction is thought to reflect drug-induced neuroadaptations (Steiner and Gerfen, 1998). A number of signaling molecules/mechanisms involved in gene regulation are affected by methylphenidate treatment. These include, for example, the dopamine transporter (Izenwasser et al, 1999; Moll et al, 2001) and the vesicular monoamine transporter (Sandoval et al, 2002); levels of GluR2 glutamate receptors (Andersen et al, 2002) and D1 dopamine receptors (Papa et al, 2002); protein kinase A and dopamine-stimulated adenylyl cyclase activities (Crawford et al, 1998); as well as the transcription factor CREB (Andersen et al, 2002). Several of these play a role in the regulation of dopamine-mediated c-fos and zif 268 expression in the striatum, and such changes could thus underlie altered gene induction after repeated methylphenidate treatment.

Prominent Effects on Substance P, but not Opioid Peptide, Expression by Acute Methylphenidate

Psychostimulants such as cocaine and amphetamine increase substance P expression in the striatum (Harlan and Garcia, 1998; Steiner and Gerfen, 1998). For example, substance P mRNA levels are elevated within 30 min of acute cocaine injection (Steiner and Gerfen, 1993; Drago et al, 1996; Brandon and Steiner, 2003) and remain elevated for more than 2–3 h (eg Hurd and Herkenham, 1992; Adams et al, 2001). Our present study is the first to demonstrate that acute methylphenidate administration also increases the expression of substance P in the striatum. Changes in mRNA levels followed a similar time course, as they were increased within 20 min, peaked at 40–60 min, and returned to baseline only after 3 h. In maximally responding (dorsal) sectors, substance P expression was slightly suppressed 24 h after methylphenidate injection. Interestingly, the time course showed a shift between rostral and caudal sectors. All rostral sectors showed peak increases around 60 min, some (medial) with near-maximal levels even at 3 h. In contrast, increases in caudal, sensorimotor sectors peaked at 40 min, coincident with peak locomotor activation by methylphenidate (Gerasimov et al, 2000). This effect suggests a relationship between substance P expression and behavior-associated neuronal activity in the caudal striatum. Together, our results demonstrate a very robust effect of acute methylphenidate on striatal substance P expression.

These substance P effects contrasted with the minimal effects on the opioid peptides dynorphin and enkephalin. Increases in dynorphin mRNA levels are generally more pronounced after repeated psychostimulant treatments due to accumulation (eg Hurd et al, 1992; Steiner and Gerfen, 1993; Daunais and McGinty, 1994; Spangler et al, 1996; Willuhn et al, 2003), and in fact increased dynorphin expression was seen after repeated methylphenidate treatment (in adolescent rats; Brandon and Steiner, 2003). However, unlike acute methylphenidate administration, both acute cocaine and amphetamine result in significantly enhanced dynorphin expression. Elevated mRNA levels can be seen as early as 30 min after drug injection (Brandon and Steiner, 2003; Willuhn et al, 2003), are prominent at 2–3 h (Hurd and Herkenham, 1992; Smith and McGinty, 1994; Wang et al, 1995; Adams et al, 2003), and have been found to last for 18 (Smith and McGinty, 1994) to 30 h (Wang et al, 1995). In our present study, only borderline increases were observed at 1 h after methylphenidate injection. These results indicate differential molecular changes after treatment with methylphenidate as compared to cocaine and amphetamine.

Similar differential effects were found for enkephalin expression. Perhaps best known are the increases in striatal enkephalin expression (indirect pathway) produced by dopamine depletion or chronic antipsychotic treatment/D2 receptor blockade (cf. Steiner and Gerfen, 1998). For example, enkephalin mRNA levels begin to rise within 1 h and remain elevated for more than 24 h after a D2 receptor antagonist injection (Steiner and Gerfen, 1999). Similarly, increases in striatal enkephalin expression have also been consistently found after acute and repeated cocaine and amphetamine treatment (Steiner and Gerfen, 1993; Wang and McGinty, 1996a; Spangler et al, 1997; Mathieu-Kia and Besson, 1998; Brandon and Steiner, 2003; see Wang and McGinty, 1996b, for review), although these changes were typically smaller than those for substance P and dynorphin expression. In contrast, neither repeated (Brandon and Steiner, 2003) nor acute methylphenidate treatment (present study) produced consistent changes in enkephalin expression.

The basis for this differential regulation of immediate-early genes and substance P vs opioid peptides is not clear, but may be related to the transmitter systems affected by methylphenidate. One important difference between methylphenidate and cocaine/amphetamine is the low affinity of methylphenidate for the serotonin transporter (Schweri et al, 1985; Gatley et al, 1996). Indeed, in contrast to cocaine and amphetamine, methylphenidate does not produce increased extracellular levels of serotonin (Kuczenski and Segal, 1997; Kankaanpaa et al, 2002). Gene expression in striatal neurons is regulated by complex interactions between several transmitter systems, most prominently dopamine, glutamate, and acetylcholine (Hyman et al, 1996; Wang and McGinty, 1996b). Serotonin also affects striatal gene expression (eg Bhat and Baraban, 1993; Torres and Rivier, 1994; Gardier et al, 2000), including that of dynorphin and enkephalin (Morris et al, 1988; Walker et al, 1996; Mijnster et al, 1998; Basura and Walker, 1999; Keefe et al, 2002). It is thus conceivable that the relatively minor effects of methylphenidate on opioid peptide expression, as observed in our present and previous studies (Brandon and Steiner, 2003), reflect a more important role for serotonin in the regulation of these genes, as compared to substance P and the immediate-early genes.

Opioid peptides in striatal output pathways are thought to serve as negative feedback mechanisms to maintain systems homeostasis (Steiner and Gerfen, 1993, 1998; Hyman and Nestler, 1996). Psychostimulant-induced changes in opioid peptide expression are considered neuroadaptations that may underlie certain withdrawal symptoms and other behavioral effects in psychostimulant addiction (Hyman and Nestler, 1996; Nestler, 2001). Our results suggest that methylphenidate is less likely than other psychostimulants to trigger such molecular adaptations.

Functional Domains Affected by Methylphenidate

One purpose of our present study was to map the topography of acute methylphenidate-induced gene regulation in the striatum and to compare it to the topography of cocaine-induced gene regulation (Willuhn et al, 2003). The goal was to determine whether there is an association of methylphenidate-induced molecular changes with particular corticostriatal circuits (functional domains). As in our previous study (Willuhn et al, 2003), we divided the striatum (caudate-putamen and nucleus accumbens) into sectors that were mostly defined by their predominant cortical inputs (Figure 1; see Materials and methods and Willuhn et al, 2003).

Our most important findings can be summarized as follows:

  1. 1)

    Similar to the effects of cocaine (Willuhn et al, 2003), the most pronounced increases in gene expression occurred in sectors receiving inputs from the sensorimotor cortex on middle to caudal striatal levels (Figure 7).

  2. 2)

    In the sensorimotor striatum, the dorsal sectors that receive the densest input from the medial agranular cortex (M2 in Figure 1) (Reep et al, 1987, 2003; Berendse et al, 1992; Kincaid and Wilson, 1996; Reep and Corwin, 1999), in addition to inputs from the somatosensory, visual, and primary motor cortex (eg Reep et al, 1987, 2003; McGeorge and Faull, 1989; Brown and Sharp, 1995; Alloway et al, 1999), displayed maximal changes in gene regulation. Surrounding sectors that are, to some lesser extent, also targets of medial agranular projections (Reep et al, 1987, 2003; Berendse et al, 1992; Kincaid and Wilson, 1996; Reep and Corwin, 1999), also consistently showed changes in gene expression. These findings match cocaine effects (Willuhn et al, 2003). The medial agranular cortex has mixed prefrontal/premotor features (eg connections with the mediodorsal as well as the ventral lateral and ventromedial thalamic nuclei; neurons that project to brain stem and spinal cord; Donoghue and Wise, 1982; Reep et al, 1987, 2003; Passingham et al, 1988; Berendse et al, 1992; Cowan and Wilson, 1994; Reep and Corwin, 1999; see also Preuss, 1995; Uylings et al, 2003), and thus can be considered a prefrontal/motor interface. It remains to be seen whether the observed methylphenidate-induced molecular changes are driven by medial agranular cortical input (Vargo and Marshall, 1995, 1997) or other striatal inputs/mechanisms with a similar regional distribution. However, our findings indicate that sensorimotor circuits under the influence of medial agranular input are particularly prone to psychostimulant-induced neuroplasticity.

  3. 3)

    Consistently also affected, albeit to some lesser degree, were medial and rostral striatal sectors that receive inputs from several prefrontal regions including the anterior cingulate, prelimbic and lateral orbital cortex (Berendse et al, 1992), similar to cocaine effects (Willuhn et al, 2003).

  4. 4)

    Generally, small or no changes in gene regulation were seen in ventral striatal sectors that receive inputs mostly from the dorsal agranular insular cortex (Berendse et al, 1992), on all three rostrocaudal levels.

  5. 5)

    Well appreciated are cocaine-induced molecular changes in the nucleus accumbens (eg Nestler, 2001), although such gene regulation effects are generally modest compared to those in the sensorimotor striatum (see Willuhn et al, 2003). Similar differential effects between caudate-putamen and nucleus accumbens were demonstrated here and before (Brandon and Steiner, 2003), and were also found after doses as high as 30 mg/kg (i.p.) (Trinh et al, 2003). We assessed effects on gene expression in five nucleus accumbens sectors on a level that contains core and shell subdivisions. In contrast to cocaine (Willuhn et al, 2003), acute (present study) or repeated methylphenidate treatment (Brandon and Steiner, 2003) did not induce significant changes in gene regulation in these sectors, with one notable exception. The lateral part of the shell displayed a robust c-fos response at 40 min after methylphenidate injection (a similar effect was seen for zif 268 and Homer 1a expression; Yano and Steiner, 2005). These findings are consistent with our previous findings after repeated methylphenidate treatment. First, the lateral shell was the only accumbal sector with significantly increased dynorphin expression (Brandon and Steiner, 2003). Second, in this subregion only, the c-fos response to a cocaine challenge was enhanced (rather than suppressed, as in the dorsal striatum, see above) after the 7-day repeated methylphenidate treatment (Brandon and Steiner, 2003). The lateral part of the shell, at this rostrocaudal level, is apparently the only subregion of the nucleus accumbens that receives medial agranular (Reep et al, 1987) and ventral agranular insular (Berendse et al, 1992) cortical inputs, in addition to inputs from the piriform and entorhinal cortex, hippocampus, amygdala, and other regions (eg McGeorge and Faull, 1989; Brog et al, 1993; Wright and Groenewegen, 1996). Future work will have to elucidate the functional significance of methylphenidate-induced molecular changes in this accumbal region.

Functional Considerations

Our present and previous (Brandon and Steiner, 2003) studies demonstrate several methylphenidate effects on striatal gene regulation; some of these (c-fos, substance P) are similar to those of other psychostimulants, others (dynorphin, enkephalin) notably differ. Pronounced alterations in gene expression were found for c-fos and substance P, whereas effects on dynorphin and enkephalin expression were modest to minimal. Substance P is a marker for neurons of the direct striatal output pathway (striatonigral neurons), whereas enkephalin is expressed in indirect pathway (striatopallidal) neurons. Psychostimulants such as cocaine and amphetamine induce immediate-early genes predominantly in direct pathway neurons (Cenci et al, 1992; Johansson et al, 1994; Kosofsky et al, 1995), especially under treatment conditions as used in the present study (home cage treatment; eg Badiani et al, 1999; Uslaner et al, 2001). Our findings thus indicate that methylphenidate induces neuronal plasticity in the direct pathway. The lack of enkephalin effects suggests that, at least under the present treatment conditions, methylphenidate is more selective than cocaine and amphetamine in this regard (see above), although our study does not rule out other methylphenidate-induced molecular changes in indirect pathway neurons. Activity in the direct pathway provides facilitatory input back to the cortex (cf. Steiner and Kitai, 2000). The functional consequences of methylphenidate-induced molecular changes in the direct pathway remain to be demonstrated, but are most likely related to the corticostriatal circuits/functional domains affected by this drug. Our findings indicate that methylphenidate preferentially affects sensorimotor domains, somewhat less frontostriatal (associative) domains, and least limbic (ventral striatal) domains.

CONCLUSIONS

Altered gene regulation in basal ganglia-cortical circuits induced by methylphenidate may underlie some of the behavioral changes produced by this psychostimulant (see Introduction). Behavioral changes in animal models have been shown with doses as low as 2 mg/kg (i.p.). Similarly, methylphenidate-induced gene regulation in rats emerged with 2 mg/kg and higher doses (present study; Brandon and Steiner, 2003; Chase et al, 2003). This dose is considered to be within or above the range of therapeutically relevant doses (Gerasimov et al, 2000; Swanson and Volkow, 2003). Although therapeutic doses of methylphenidate can produce dopamine overflow in the striatum in humans (Volkow et al, 2001), it remains to be seen whether or not prolonged clinical use induces molecular changes in the brain. However, recreational users are likely exposed to higher peak levels due to different amounts/routes of intake (Kollins et al, 2001; Swanson and Volkow, 2003) and may thus be more at risk for such gene regulation effects and their behavioral consequences.