Genetic variation in dopamine-related gene expression influences motor skill learning in mice

Authors


Corresponding author: R. Diaz Heijtz, Department of Neuroscience, Karolinska Institutet, Retzius Väg 8, 171 77 Stockholm, Sweden. E-mail: rochellys.heijtz@ki.se

Abstract

Several neurodevelopmental disorders with a strong genetic basis, including attention-deficit/hyperactivity disorder, autism spectrum disorders and developmental coordination disorder, involve deficits in fine motor skills. This phenotype may depend on heritable variation in components of the dopamine (DA) system, which is known to play a critical role in motor skill learning. In this study, we took advantage of two inbred strains of mice (BALB/c and C57BL/6) that differ markedly in the number of midbrain DA neurons in order to investigate the influence of such naturally occurring genetic variation on the acquisition and performance of fine motor skills. Gene expression analysis of midbrain, frontal cortex and striatum showed significant differences in the expression of presynaptic and postsynaptic dopaminergic (DAergic) markers (e.g. tyrosine hydroxylase, DA transporter, DA D4 receptor, DA D5 receptor and DARPP-32) between these two strains. BALB/c mice had lower learning rate and performance scores in a complex skilled reaching task when compared with C57BL/6 mice. A negative correlation was found between the motor learning rate and level of DARPP-32 mRNA expression in the frontal cortex contralateral to the trained forelimb. The rate of motor learning was also negatively correlated with the levels of DARPP-32 and DA D1 receptor mRNAs in the striatum. Our results suggest that genetically driven variation in frontostriatal DAergic neurotransmission is a major contributor to individual differences in motor skill learning. Moreover, these findings implicate the D1R/cAMP/DARPP-32 signaling pathway in those neurodevelopmental disorders that are associated with fine motor skill deficits.

Most of our daily activities require fine motor skills such as the use of skilled hand movements when reaching and grasping small objects. However, acquisition and performance of fine motor skills vary substantially between individuals. Twin studies suggest that heritability is a main factor responsible for interindividual differences in motor skill (Fox et al. 1996; Missitzi et al. 2011). Recent studies have provided evidence that common genetic variations (e.g. single nucleotide polymorphism) account for a certain amount of variance in motor skill learning and motor plasticity (Cheeran et al. 2009). Moreover, several highly heritable neurodevelopmental disorders, including attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorders (ASDs), are often associated with deficits in fine motor skills (Bhat et al. 2011; Eliasson et al. 2004; Meyer & Sagvolden 2006). A core symptom in developmental coordination disorder (DCD) is impaired performance of motor skills, which has not been attributed to general intellectual, sensory or motor neurological impairment (Barnhart et al. 2003). Children with DCD often have co-occurring conditions such as ADHD, ASDs and dyslexia. Molecular genetic studies implicate heritable variation in components of the dopaminergic (DAergic) system in the phenotype of poor motor development (Faraone & Mick 2010), which is known to play a critical role in motor skill learning and corticostriatal plasticity (Costa 2007).

In rodents, skilled forelimb movements can be studied using reaching tasks, including the so-called skilled forelimb reaching task (Klein et al. 2012). This task requires animals to reach for a food pellet through a slot in front of a Plexiglas box, grasp and retrieve the pellet with a single forelimb. Using a similar type of task, Luft and coworkers recently showed that DAergic terminals from the ventral tegmental area (VTA) to primary motor cortex (M1) play a crucial role for motor learning and associated synaptic plasticity in M1 (Hosp et al. 2011; Molina-Luna et al. 2009). Elimination of DAergic terminals and/or blockade of dopamine (DA) receptors in M1 impaired motor skill acquisition, but not execution of a previously acquired motor skill (Hosp et al. 2011; Molina-Luna et al. 2009). These novel findings indicate that M1 requires an optimal level of DA neurotransmission to facilitate motor skill learning. They also raise the possibility that normal genetic variation in the number of midbrain DA neurons and/or DA receptor expression in M1 can explain interindividual differences in motor skill learning.

Substantial genetic variation is observed in the mesotelencephalic DA system of mice, and parallel differences in drug responses and behavior have been identified (Fink & Reis 1981; Reis et al. 1983). Two inbred strains that differ markedly in the number of midbrain DA neurons and activity of tyrosine hydroxylase (TH) are BALB/c and C57BL/6. Mice of the BALB/c strain have higher number of midbrain DA neurons and greater TH content than mice of the C57BL/6 strain (Vadasz et al. 2007). In addition, BALB/c mice display several behavioral traits relevant to autism, including low levels of sociability and high levels of anxiety (see Brodkin 2007). In this study, we took advantage of these two strains of mice in order to investigate the influence of such naturally occurring genetic variation in the DAergic system on the acquisition and performance of fine motor skills. Also, to investigate potential associations with specific components of the DA systems, we studied gene expression of DA receptors and signaling mechanisms in the frontal cortex (including M1) and striatum.

Materials and methods

Animals

All experiments were performed in 10-week-old male C57BL/6 and BALB/c mice (Charles River, Sulzfeld, Germany). The animals arrived in the laboratory 2 weeks before the experiment and were housed in groups in standard plastic cages (Type III Makrolon®, Tecniplast, Buguggiate, Italy) under controlled temperature, humidity and light (12:12 h light–dark cycle; lights on at 0700 h) conditions. Food and water were available ad libitum. Animals involved in the skilled reaching task were housed in pairs. Each set of paired mice was separated by a clear Plexiglas partition (containing small holes; 10 mm diameter) that divided the home cage in half in order to monitor the daily food intake for each individual mouse. All experiments were conducted under an approved protocol from the Ethical Committee on Animal Research, Stockholm North, and in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC).

Gene expression studies

In situ hybridization

Naïve C57BL/6 and BALB/c mice (n = 8 per strain) were killed by cervical dislocation and their brains rapidly removed, frozen on dry ice and kept at −80°C until used. Coronal sections (20 µm) of the ventral midbrain region were prepared on a cryostat and stored at −80°C until used. The DA D2 receptor (D2R) probe was prepared from a BspMII-AflII fragment corresponding to nearly the entire third intracellular domain of the rat D2R, which was subcloned into a pGEM-4Z vector (Bunzow et al. 1988). The DA transporter (DAT) probe was prepared from an EcoRI fragment containing the coding region of the rat DAT, which was subcloned into a pBluescript II SK (−) vector (Shimada et al. 1991). The TH probe was prepared from a Pst1-KpnI fragment corresponding to nearly the entire coding region of the rat TH, which was subcloned into a pBR322 vector (Grima et al. 1985). Linearized plasmids were used to synthesize [35S] UTP-labeled riboprobes. In vitro transcription was carried out using the MAXIscript™ SP6/T7 kit (Applied Biosystems, Uppsala, Sweden) and 35S-UTP (NEG039H; Perkin Elmer, Upplands Väsby, Sweden) according to the manufacturer's instructions. The transcripts were purified using NucAway™ Spin Columns (Applied Biosystems). Fixation, prehybridization, hybridization and washes were performed essentially as previously described (Diaz Heijtz et al. 2010); hybridization was made with 2 × 106 cpm/section 35S-UTP-labeled RNA probe at 55°C for 14–16 h. Non-specific hybridization was determined by incubating sections with the respective 35S-UTP-labeled sense probe under identical conditions to that of the antisense probe. Following RNase A washing, treated slides were hydrated, dried and exposed to ß-Max autoradiographic film (Hyperfilm, ß-film, BIOMAX, Kodak, Sigma-Aldrich, Stockholm, Sweden) for 1–3 days to enable visualization of TH, DAT and D2R hybridization. Calibrated [14C]-labeled standards (Amersham Biosciences, Uppsala, Sweden) were incorporated in all cassettes to allow for quantification. Films were developed in D19 developer for 2 min and in 1:5 dilution of Amfix fixative for 10 min. Films were scanned with an Epson Perfection V700 Photo Scanner as grayscale film, using 800 dpi, and saved as high-quality TIFF files. Optical density values were quantified by using appropriate software (NIH Image J version 1.29, National Institutes of Health). A 14C step standard (GE Healthcare, Uppsala, Sweden) was included to calibrate optical density readings and convert measured values into nCi/g. The substantia nigra (SN) and VTA were identified according to a mouse brain atlas (Franklin & Paxinos 2007). All comparisons between groups were made on sections hybridized together, under identical conditions, and exposed for the same period of time to ß-Max autoradiography film (VWR, Stockholm, Sweden).

Quantitative real-time polymerase chain reaction (qRT-PCR)

The frontal cortex (including M1) and striatum from both naïve (n = 6 per strain) and trained animals (n = 8 per group; see the Skilled forelimb reaching task section below) were rapidly dissected on ice and frozen on dry ice and stored at −80°C until used. Total RNA was isolated using RNeasy® Mini Kits (QIAGEN AB, Sollentuna, Sweden) according to the manufacturer's instructions (including the optional DNase digestion step for 15 min at room temperature) and quantified by spectrophotometry using a NanoDrop® ND-2000 Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). The integrity and purity of the RNA preparations were analyzed by capillary electrophoresis with a Bio-Rad Experion Automated Electrophoresis system (BIO-RAD, Sundbyberg, Sweden) using Experion RNA StdSens Chip kit (BIO-RAD). First-strand cDNAs were synthesized from equal amounts of total RNA (1 µg/reaction) using the iScript cDNA synthesis kit (BIO-RAD) according to the manufacturer's instructions and stored at −20°C until used.

Real-time PCR was carried out using the iCycler iQ5 Real-Time PCR Detection System (BIO-RAD). Briefly, each PCR reaction contained 30 ng cDNA, 0.5 µM of each primer, nuclease-free water and 2× iQ™ SYBR® Green Supermix (SYBR® Green I dye, 50 U/ml iTaq™ DNA polymerase, dNTPs, 6 mM MgCl2, 100 mM KCl, 20 nM fluorescein including stabilizers and 40 mM Tris–HCl, pH 8.4) in a 25 µl reaction. The housekeeping gene, TBP (encoding the TATA-box binding protein, gene ascension number: NM_001004198), was used for normalization. All samples were performed in triplicate. The cycling program was set as: step 1, 95°C for 3 min; step 2, 40 cycles of 95°C for 10 seconds followed by 50–60°C for 30 seconds and step 3, 80 cycles for 10 seconds each, beginning at 55°C and increasing by 0.5°C with each subsequent cycle. Subsequent to the amplification procedure, a melting-curve analysis was performed (set point, 55°C) in order to confirm amplification specificity. The specificity of the gene products was determined via melting-curve analyses.

The Primer-BLAST web-based software (http://www.ncbi.nlm.nih.gov/tools/primer-blast/) was used to design gene-specific primers. Primer sequences and annealing temperatures used for TH, DAT, DA D2 short isoform (D2RS), DA D2 receptor long isoform (D2RL), DA D3 receptor (D3R), DA D1 receptor (D1R), DA D5 receptor (D5R) and DARPP-32 (DA- and cAMP-regulated neuronal phosphoprotein) are listed in Table S1, Supporting Information. The data analysis was based on the 2−ΔΔCt method (Livak & Schmittgen 2001). The normalized ΔCt for each gene of interest (GOI) was calculated by deducting the Ct of the housekeeping gene TBP (as this gene showed no expression difference among the different RNA samples) from the Ct of each GOI. Then, the double delta Ct (ΔΔCt) for each GOI was calculated by deducting the average ΔCt of GOI in the C57BL/6 group from the ΔCt of each GOI in the BALB/c group. The fold changes of each GOI compared with the C57BL/6 group were calculated as 2−ΔΔCt.

Behavioral studies

General behavioral procedure

All behavioral testing took place between 0900 and 1600 h under low illumination in order to reduce stress. Prior to any behavioral procedure, mice were brought in their home cages to a room adjacent to the testing room and allowed to habituate for at least 90 min before testing in order to minimize stress caused by environmental changes. To avoid carry-over effects, independent groups of animals were used in each of the different behavioral tasks (n = 18–20, 8 and 8 per strain for the skilled forelimb reaching task, motor coordination test and open-field activity test, respectively).

Skilled forelimb reaching task

Single pellet reaching box

Single pellet reaching boxes were made of clear Plexiglas (8 cm wide, 21.5 cm long and 20 cm high) as previously described (Farr & Whishaw 2002). In the center of each box's front wall was an open vertical slot 0.8 cm wide that extended from the floor up to a height of 10 cm (see Fig. S1). Outside of the wall, in front of the slot, a shelf (width 4.5 cm and length 8 cm) was mounted 1.1 cm above the floor. Small indentations to hold food pellets were located 1.2 cm from the inside of the front wall, aligned with the edges of the open slot (see Fig. S1). This distance prevents the mouse from retrieving food pellets by use of its tongue.

Feeding and food restriction

Prior to and during skilled reaching training, mice were placed on a restricted diet until they reached 85–90% of their baseline ad libitum body weight. To familiarize the mice with the food pellets used for the skilled reaching task, each mouse received 30 of these pellets (test diet, 20 mg precision-weight, purified rodent tablets, Sandown Scientific, Middlesex, England) 8 h prior to the daily Purina Rodent Chow ration given 1 week prior to training. Once skilled reaching training began, and until the end of the motor training, only rodent chow was provided in the home cage.

Pretraining and training sessions

For 20 min during each of the 2 days prior to skilled reaching training, mice were placed into the reaching box with food pellets available on the shelf (in front of the indentations; see Fig. S1). The purpose was to introduce each animal to the training box and to identify the preferred forelimb for reaching/grasping the pellets through the open slot. By the end of the second day, all animals showed a consistent preference for one forelimb in their attempts to reach/grasp the pellets (>80% of the time). Once the forelimb preference was determined for each individual animal, a single food pellet per trial was placed into the indentation located 1.2 cm from the inside of the front wall contralateral to the preferred forelimb. During the subsequent 10 days, animals in the trained group underwent daily 20-min training sessions consisting of 30 discrete trials, one single pellet per trial. Each food pellet was immediately removed from the shelf when the pellet was displaced too far away from the indentation to prevent additional reaching attempts after an unsuccessful discreet trial. During intertrial intervals, mice were trained to leave the open slot at the front of the box and walk to the rear wall of the cage to wait a few seconds before returning to the front of the cage for the subsequent trial. This was accomplished by occasionally placing a food pellet close to the rear wall of the cage. In addition, food pellets were not placed on the shelf on semirandomly selected trials in order to teach the animals to reach only when a food pellet was present in the front shelf.

Video recording

Reaching performance was video recorded using a SAMSUNG HMX-H100P high definition camcorder. Frame-by-frame analyses of skilled reaches were evaluated using computer-based software (Windows Media Player).

Endpoint analysis of reaching behavior

Reaching behavior was analyzed by measuring:

  1. First attempt success: First attempt success was the percentage of success in which a mouse obtained a food pellet on the first advance of the forelimb toward the food. It was calculated as: Success on first reach % = (number of pellets obtained on first advance/30) × 100.
  2. Total success: A successful reach was defined as one in which an animal grasped a food pellet, transported it in the paw into the cage and placed it into its mouth regardless of the number of forelimb advances toward the food pellet required. Total success was calculated as: Success % = (number of pellets obtained/30) × 100.
  3. Learning rate: The learning rate was calculated by subtracting the average performance of the first 2 days from the average performance of the last 2 days, and divided by the intertrial intervals [(meantrials 9 and 10) − (meantrials 1 and 2)/9 (number of intertrial intervals)].

Two independent cohorts of mice were trained in the skilled reaching task under identical conditions (cohort 1, n = 8 per strain; cohort 2, n = 10–12 per strain). The behavioral data from both cohorts were combined because the results obtained were very similar. Twenty-four hours after the last skilled reaching training session (10th training day), mice in cohort 1 were killed by dislocation and specific brain regions were dissected for gene expression analyses. The brains from mice in cohort 2 were not evaluated in this study.

Motor coordination and balance

Motor coordination and balance were tested using an accelerating Rota-Rod® (UGO Basile, Comerio, Italy). One day before testing, naïve animals (n = 8 per strain) were acclimated to the rotarod by being placed on the cylinder rotating at a fixed speed (i.e. 4 rpm) for two 90-second periods, 2 h apart. The rotarod test was performed by placing an animal on the rotating cylinder and measuring the time each animal was able to walk on top of the rod while maintaining its balance, for 5 min maximum. The speed of the rotarod accelerated from 4 to 40 rpm over 5 min. Mice received 10 training trials with 5-min intertrial interval per day.

Open-field test

Naïve BALB/c and C57BL/6 mice (n = 8 per strain) were individually placed into an open-field activity box (48 × 48 cm; TSE ActiMot System, TSE, Bad Homburg, Germany) and their spontaneous motor activity was measured for 120 min. After their initial exposure to the open-field activity box on day 1, the mice were re-exposed to the same box for two additional consecutive days. The computer program automatically recorded the following parameters: distance traveled (in meters) and number of rears.

Statistical analysis

Statistical analyses were performed using the STATVIEW computer software (version 4.0). All behavioral experiments were analyzed using either repeated-measures analysis of variance (anova) or factorial anova when appropriate. All post hoc comparisons were made using a Bonferroni/Dunn test when significant anova effects were found. Data from gene expression studies were analyzed using an unpaired t-test (two-tailed). The threshold for statistical significance was set as P ≤ 0.05. All data are presented as the mean (M) ± standard error of the mean (SEM).

Results

Gene expression studies

We first compared the expression of key presynaptic DAergic markers in the VTA and SN of naïve BALB/c and C57BL/6 mice by means of in situ hybridization technique. The mRNA expression levels of TH, DAT and D2R were significantly higher in both the VTA (TH: t(14) = −2.438, P < 0.05; DAT: t(14) = −2.185, P < 0.05; D2R: t(14) = −2.905, P < 0.05) and SN (TH: t(14) = −5.437, P < 0.0001; DAT: t(14) = −3.410, P < 0.01; D2R: t(14) = −5.440, P < 0.0001) of BALB/c mice compared with C57BL/6 mice (Fig. 1). In another set of naïve BALB/c and C57BL/6 mice, mRNA expression levels of the two D2R receptor splicing variants, D2RS and D2RL, as well as D3R, TH and DAT were evaluated by means of qPCR. Each of these markers were more highly and significantly (D2RS: t(14) = −5.729, P < 0.001; D2RL: t(14) = −4.961, P < 0.001; D3R: t(14) = −2.841, P < 0.05; TH: t(14) = −2.489, P < 0.05; DAT: t(14) = −2.563, P < 0.05) expressed in the ventral midbrain of BALB/c mice than in C57BL/6 mice (see Fig. S2).

Figure 1.

BALB/c mice show altered expression of presynaptic dopaminergic markers in the ventral midbrain region. Representative autoradiographs showing the mRNA expression levels of (a) TH, (b) DAT and (c) D2R at the level of the midbrain region of naïve adult C57BL/6 and BALB/c mice. The pseudo-coloring indicates signal intensity from low (black/purple) to high (yellow/white). Data in bar graphs show mean optical density (OD) values for (a′) TH, (b′) DAT and (c′) D2R mRNAs in the VTA and SN. The results (a′–c′) are presented as means ± SEM; n = 8 per group. Asterisks denote where BALB/c mice differ significantly (*P < 0.05, **P < 0.01, ***P < 0.001) from C57BL/6 mice.

The mRNA expression levels of DA receptors (D1R, D5R, D2RS, D2RL and D3R) and the signaling mechanism (i.e. DARPP-32) were also evaluated in both the frontal cortex and striatum of naïve BALB/c and C57BL/6 mice by means of qPCR technique. In BALB/c mice, D5R mRNA levels were significantly lower in both the frontal cortex and striatum (t(10) = 11.874, P < 0.0001 and t(10) = 2.417, P < 0.05, respectively) compared with C57BL/6 mice (Fig. 2b). In the striatum, D3R mRNA levels were significantly lower in BALB/c mice (t(10) = 4.596, P < 0.01; Fig. 2d). In contrast, D3R, D4R and DARPP-32 mRNA levels were significantly (D3R: t(10) = −3.487, P < 0.01; D4R: t(10) = −5.481, P < 0.001; DARPP-32: t(10) = −7.811, P < 0.0001) higher in the frontal cortex of BALB/c mice compared with C57BL/6 mice (Fig. 2d–f).

Figure 2.

BALB/c mice show altered expression of dopamine-related markers in the frontal cortex and striatum. Quantitative RT-PCR was used to examine mRNA expression levels of (a) D1R, (b) D5R, (c) D2RL, (d) D3R, (e) D4R and (f) DARPP-32 in the frontal cortex and striatum of naïve adult C57BL/6 (black bars) and BALB/c (white bars) mice. Expression level of each gene examined was normalized to TATA-binding protein levels and expressed relative to the C57BL/6 group. Data in bar graphs represent means ± SEM; n = 6 per group. Asterisks denote where BALB/c mice differ significantly (*P < 0.05, **P < 0.01, ***P < 0.001) from C57BL/6 mice.

Behavioral studies

Skilled forelimb reaching task

The performance of BALB/c and C57BL/6 mice in the skilled reaching task is illustrated in Fig. 3. Repeated-measures anova for success on the first reach attempt (%) showed significant effects of strain (F1,36 = 17.91, P < 0.001), training day (F9,324 = 13.57, P < 0.0001) and a strain by training day interaction (F9,324 = 2.97, P < 0.01). The BALB/c mice displayed little improvement over the 10-day testing period (Fig. 3a). Post hoc Bonferroni/Dunn analyses confirmed that BALB/c mice had significantly (P < 0.05) lower scores in success on the first reach attempt (%) throughout the testing period compared with C57BL/6 mice. Moreover, BALB/c mice had a significantly (P < 0.05) lower learning rate for success on the first reach attempt (%) than C57BL/6 mice (Fig. 3a′).

Figure 3.

Deficits in skilled reaching behavior in BALB/c mice. Endpoint measurements of reaching behavior are presented. (a) Success on the first reach attempt, in percentage. (b) Total success, in percentage. (a′ and b′) Bar graphs show the learning rate for each endpoint. The results are presented as means ± SEM; n = 18–20 per group. Asterisks denote where BALB/c mice differ significantly (*P < 0.05, **P < 0.01, ***P < 0.001) from C57BL/6 mice.

Analysis of total success (%) by repeated-measures anova showed a significant effect of training day (F9,324 = 5.89, P < 0.0001) and a strain by training day interaction (F9,324 = 3.11, P < 0.01). Post hoc Bonferroni/Dunn analyses showed that BALB/c mice had significantly lower scores in total success on days 5 and 7 compared with C57BL/6 mice (P < 0.05; Fig. 3b). The learning rate for total success (%) between the two strains differed significantly (P < 0.05; Fig. 3b′), with BALB/c mice exhibiting almost no improvement between day 1 and day 10 of testing (day 1 vs. day 10; 32.4 ± 4.3 vs. 38.0 ± 4.5) compared with C57BL/6 mice (day 1 vs. day 10; 25.2 ± 4.4 vs. 51.2 ± 5.1).

Correlation between learning rate of skilled reaching behavior and expression of DA-related markers

Given the crucial role of DA in motor learning and corticostriatal synaptic plasticity, we tested the hypothesis that variation in skilled reaching behavior would be associated with expression levels of DA receptors and/or DARPP-32 (for details about the skilled reaching behavior of mice included in this analysis, see Fig. S3).

A highly significant negative correlation was found between the learning rate for total success (%) and level of DARPP-32 mRNA expression in the frontal cortex and striatum contralateral to the trained forelimb (r = −0.65, P = 0.006 and r = −0.81, P < 0.001, respectively; see Fig. 4a,b), but not ipsilaterally (see Fig. 4a′,b′). Similarly, a significant negative correlation was obtained between the learning rate for success of the first reach attempt (%) and level of DARPP-32 mRNA expression in the frontal cortex contralateral to the trained forelimb (r = −0.56, P = 0.026; data not shown). In addition, a significant negative correlation was found between the learning rate for total success (%) and the level of D1R in the striata, both contralateral and ipsilateral to the trained forelimb (r = −0.51, P = 0.044 and r = −0.55, P = 0.029, respectively), but not in the frontal cortex region.

Figure 4.

Correlations between learning rate for total success and expression of dopamine-related markers in the frontal cortex and striatum. Circles represent C57BL/6 (black) and BALB/c (white) mice.

Motor coordination and balance test

The performance of BALB/c and C57BL/6 mice on the rotarod test is illustrated in Fig. 5. Repeated-measures anova for latency at which mice fell off the rotating cylinder showed significant effects of strain (F1,14 = 23.68, P < 0.001) and trial (F9,126 = 20.25, P < 0.0001). However, no significant strain by trial interaction was found (F9,126 = 1.34, P > 0.1). Post hoc Bonferroni/Dunn analyses showed that both BALB/c and C57BL/6 mice improved significantly (P < 0.0001) from trial 1 to trial 10. However, BALB/c mice had significantly shorter latencies to fall than C57BL/6 mice (P < 0.05; Fig. 5).

Figure 5.

Deficits in gross motor coordination in BALB/c mice. The latency to fall off from an accelerating rotarod is presented. The speed of the rotarod accelerated from 4 to 40 rpm over 300 seconds. Each mouse had 10 trials per test with an intertrial interval of 5 min. The results are presented as means ± SEM; n = 8 per group. Asterisks denote where BALB/c mice differ significantly (*P < 0.05, **P < 0.01, ***P < 0.001) from C57BL/6 mice.

Open-field activity test

Naïve BALB/c and C57BL/6 mice were placed into an open-field box and their spontaneous motor activity was recorded for 2 h. Repeated-measures anova showed a significant strain by time interaction for both distance traveled and number of rears (F11,154 = 15.67, P < 0.0001 and F11,154 = 26.03, P < 0.0001, respectively). Post hoc Bonferroni/Dunn analyses showed that BALB/c mice traveled significantly shorter distances and exhibited lower numbers of rears during the first 30 min of testing compared with C57BL/6 mice (Fig. 6a,b). However, BALB/c mice showed poor habituation response to the testing box. While the locomotor and rearing activity of C57BL/6 mice gradually decreased over time, reaching almost zero by the end of testing, the locomotor activity of BALB/c mice remained nearly constant throughout the testing period (Fig. 6a). Moreover, BALB/c mice showed a significant (P < 0.01; Fig. 6b) increase in the number of rears during the 90- to 120-min interval of testing.

Figure 6.

BALB/c mice display impaired habituation to a novel environment. (a) Average distance traveled (meters) and (b) number of rears measured in 10-min time bins across a 120-min session in an open-field box. (c) Representative traces of movement patterns of C57BL/6 and BALB/c mice at the 0–10, 60–70 and 110–120 min time intervals of the 120-min open-field test session; distance traveled and rearing activity are shown in dark red and blue colors, respectively. The results are presented as means ± SEM; n = 8 per group. Asterisks denote where BALB/c mice differ significantly (*P < 0.05, **P < 0.01, ***P < 0.001) from C57BL/6 mice.

After their initial exposure to the open-field box, the mice were re-exposed to the same box for two consecutive days. BALB/c mice displayed a similar pattern of locomotor and rearing activities on days 2 and 3 to that observed on day 1 (see Fig. S4).

Discussion

This study shows significant variation in skilled reaching behavior in two inbred mouse strains (BALB/c and C57BL/6 mice) known to show substantial genetic variation in the mesencephalic DAergic system. Specifically, variations in the rate of motor learning correlated with divergent DA-related gene expression in frontal cortex and striatum. These results implicate genetically driven variation in frontostriatal DAergic neurotransmission as a key contributor to individual differences in fine motor skills.

Strain differences in gene expression

Previous studies have reported considerable natural variation in gene expression among different inbred mouse strains in various brain regions, including hippocampus, amygdala and cerebellum (Nadler et al. 2006). In this study, we show that the expression of various DAergic markers in midbrain, frontal cortex and striatum differs substantially between BALB/c and C57BL/6 mice. In the midbrain, we found that naïve BALB/c mice express higher levels of various DA-related genes (e.g. TH, DAT and D2R) in both VTA and SN compared with C57BL/6 mice. This finding is consistent with previous studies showing that BALB/c mice have more midbrain DA neurons than mice of the C57BL/6 strain (Vadasz et al. 2007). Moreover, these two strains of mice also exhibit substantial variation in the expression of various postsynaptic DAergic-related markers in the frontal cortex (i.e. DARPP-32, D4R and D5R) and striatum (i.e. DARPP-32, D3R and D5R). At the biochemical level, there is evidence that BALB/c mice have higher DA turnover in various cortical and subcortical brain regions, including prefrontal cortex and striatum, compared with C57BL/6 mice (Browne et al. 2011). Taken together, these findings suggest an alteration in DA neurotransmission within the frontostriatal circuitry of BALB/c mice involved in motor control and cognitive functions.

Strain differences in skilled reaching behavior

Recent studies have shown that DAergic projections originating from the VTA to M1 are necessary for successful motor skill learning (Hosp et al. 2011; Molina-Luna et al. 2009). In this study, we show that BALB/c and C57BL/6 mice, which differ in the number of VTA DA neurons (Vadasz et al. 2007), also exhibit differences in the acquisition and performance of motor skills. Although BALB/c mice were able to learn the skilled forelimb reaching task, their performance was significantly poorer than that of C57BL/6 mice, especially in the most sensitive measure of skilled performance, success on the first reach attempt. Moreover, BALB/c mice exhibited a lower learning rate than C57BL/6 mice. These results suggest that natural genetic variation in frontostriatal DAergic neurotransmission is an important neurobiological mechanism underlying individual differences in skilled reaching behavior.

The above suggestion is supported by our gene expression studies showing alterations in the expression of several DA markers in the frontal cortex and striatum of BALB/c mice. Thus, BALB/c mice had higher expression levels of DARPP-32, D3R and D4R in the frontal cortex (which also includes M1) compared with C57BL/6 mice, but lower levels of D5R. More interestingly, we found that the improvement of skilled reaching behavior (as reflected by the rate of learning) was negatively correlated with the expression levels of DARPP-32 in the frontal cortex and striatum contralateral to the trained forelimb.

DARPP-32 is a signaling hub that plays a critical role in the integration of numerous neurotransmitter and neuromodulator signals arriving at dopaminoceptive neurons (for review, see Svenningsson et al. 2004). In particular, it is a main target for both DA and glutamate signaling (Girault 2012; Walaas et al. 2011). DARPP-32 is highly enriched in medium spiny neurons of the striatum and glutamatergic pyramidal neurons of the frontal cortex (albeit at lower levels) (Kuroiwa et al. 2012; Ouimet et al. 1984, 1992; Perez & Lewis 1992). The function of DARPP-32 depends on its relative state of phosphorylation at multiple regulatory sites (e.g. Thr-34, Thr-75 and Ser-97) that are regulated by calcium and cAMP signaling pathways (Girault 2012; Walaas et al. 2011). For example, when DARPP-32 is phosphorylated at the Thr-34 site by cAMP-dependent protein kinase (e.g. by DA acting via D1R), it is converted into a potent inhibitor of the multifunctional serine/threonine protein phosphatase-1 (PP1). The PP1 controls the phosphorylation state and hence the physiological activity of a number of downstream targets such as α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid and N-methyl-d-aspartate glutamate receptors, kinases (e.g. calcium/calmodulin-dependent protein kinase II) and transcription factors (e.g. CREB) that are necessary for synaptic plasticity and learning. Our findings suggest that DARPP-32 plays an important role in the acquisition of new fine motor skills and that its overexpression in frontostriatal circuitry could lead to deficits in fine motor skills. It is therefore tempting to speculate that genetic variation in the expression of DARPP-32 could affect the activity of PP1, and thereby synaptic plasticity and motor skill learning.

In the striatum both contralateral and ipsilateral to the trained forelimb, we found a negative correlation between the rate of motor learning and the expression levels of D1R, but not D5R. Similarly, we found a negative correlation with the expression levels of DARPP-32 in the contralateral striatum. However, in naïve animals, the expression of neither D1R nor DARPP-32 mRNAs in the striatum differs significantly between BALB/c and C57BL/6 mice. One possible explanation for such discrepancy is that extensive motor training may alter the expression of DA markers (i.e. downregulation of D1R and DARPP-32 in the striatonigral medium-size neurons) and that the mechanism(s) mediating this type of plasticity may be aberrant in poor learners. This suggestion is consistent with results of previous mouse studies showing substantial reorganization of the striatum during the acquisition and automatization of a motor skill, in which performance is independent of D1R-expressing, striatonigral medium-size neurons after extensive training (Yin et al. 2009). Moreover, there is evidence that the expression of DARPP-32 in the striatum is sensitive to training (i.e. in instrumental training involving manipulation of forelimbs during level pressing) (Segovia et al. 2012). Overall, our findings indicate that D1R, in addition to DARPP-32 intracellular signaling pathway, is a key DA receptor involved in the acquisition of motor skills in the striatum.

Strain differences in motor coordination and balance

This study shows that BALB/c mice also show poorer motor coordination and balance on the rotarod compared with C57BL/6 mice. These findings are consistent with earlier studies reporting differences in eight inbred strains of mice, including mice of the BALB/cByJ and C57BL/6J substrains, in both motor adaptability and motor learning on the rotarod (Mcfadyen et al. 2003). It is also tempting to relate these motor deficits to alterations in DA neurotransmission within the frontostriatal circuitry. However, in this study, we did not investigate potential correlations between rotarod performance and gene expression. Nevertheless, there is evidence that DA differentially modulates motor skill learning and motor adaptability on the rotarod paradigm. Recently, Chagniel et al. showed that general motor abilities (e.g. coordination, motion speed and muscular strength) remain relatively intact after DAergic depletion in mice, whereas acquisition of skilled behavior is impaired after severe DAergic lesions (Chagniel et al. 2012). Thus, it is likely that other neurotransmitter systems play a more important role in the observed differences in motor coordination between BALB/c and C57BL/6 mice.

Strain differences in spontaneous motor behavior

In this study, we also compared the exploratory activity and habituation profile of BALB/c and C57BL/6 mice after repeated exposure to a novel environment. Mice exposed to a novel environment (i.e. an open-field box) typically display high levels of exploratory behavior. However, when mice are repeatedly placed into the same environment, as well as after a prolonged exposure to an open field within a session, a progressive reduction occurs in exploratory behavior as the novel environment becomes familiar. Our results showed that BALB/c mice, in contrast to C57BL/6, failed to habituate to the open-field box. In fact, BALB/c mice had similar levels of locomotor activity throughout the entire testing period. In contrast, other authors have found BALB/c mice to be more anxious and hypoactive in the open-field test compared with C57BL/6 mice (Moy et al. 2007), and even more active compared with several strains, including C57BL/6 mice (Isles et al. 2004). These discrepancies are likely due to the fact that some authors used shorter testing periods than ours (Moy et al. 2007), whereas others analyzed only cumulative data (Isles et al. 2004). Instead, our data indicate that mice of the BALB/c strain display impaired habituation response to a novel environment, which leads to an increased level of locomotor activity over time.

In the open-field test, BALB/c mice also showed decreased levels of rearing activity during the initial novelty phase, but increased levels during the habituation period. Their initial response to the open-field exposure is consistent with the anxiety-like trait of the BALB/c strain (Moy et al. 2007). It is worth noting that reduced response habituation has been previously associated with impaired attention in both humans and animals (Tipper et al. 1989; Zhuang et al. 2001). Interestingly, we also found that BALB/c mice express higher levels of D4R in the frontal cortex than C57BL/6 mice. The D4R, which is expressed mainly in the prefrontal cortex, is of particular interest because genetic variants of the D4R gene in human have been associated with ADHD (Faraone & Mick 2010). Thus, future studies using specific cognitive paradigms that assess attentional processes (e.g. five-choice serial reaction time task) are warranted to determine whether there is an attentional component to the open-field behavior of the BALB/c strain.

Implications for neurodevelopmental disorders

Children with neurodevelopmental disorders such as ADHD and ASDs often experience motor problems. The wide range of motor problems include difficulties in both learning and performing a wide variety of motor skills (e.g. tying shoes and playing sports) and deficits in fine motor skills (e.g. poor handwriting) (Sergeant et al. 2006). The biological mechanisms underlying fine motor skill problems in these children are poorly understood. This study provides evidence supporting the notion that normal genetic variation in the DAergic system might contribute substantially to variability in the acquisition of motor skills in humans. More specifically, the results suggest the involvement of the D1R/cAMP/DARPP-32 signaling pathway in abnormal development of fine motor skills.

Acknowledgments

This work was supported by the Foundation Olle Engkvist, Swedish Brain Foundation, Freemasons Children's House fund, Swedish Research Council (5925), Swedish Foundation for Strategic Research, VINNOVA, Strategic Neuroscience Program at Karolinska Institutet and the Knut and Alice Wallenberg Foundation. The authors declare no conflicts of interest.

Ancillary