Defective response inhibition and collicular noradrenaline enrichment in mice with duplicated retinotopic map in the superior colliculus
The superior colliculus is a hub for multisensory integration necessary for visuo-spatial orientation, control of gaze movements and attention. The multiple functions of the superior colliculus have prompted hypotheses about its involvement in neuropsychiatric conditions, but to date, this topic has not been addressed experimentally. We describe experiments on genetically modified mice, the Isl2-EphA3 knock-in line, that show a well-characterized duplication of the retino-collicular and cortico-collicular axonal projections leading to hyperstimulation of the superior colliculus. To explore the functional impact of collicular hyperstimulation, we compared the performance of homozygous knock-in, heterozygous knock-in and wild-type mice in several behavioral tasks requiring collicular activity. The light/dark box test and Go/No-Go conditioning task revealed that homozygous mutant mice exhibit defective response inhibition, a form of impulsivity. This defect was specific to attention as other tests showed no differences in visually driven behavior, motivation, visuo-spatial learning and sensorimotor abilities among the different groups of mice. Monoamine quantification and gene expression profiling demonstrated a specific enrichment of noradrenaline only in the superficial layers of the superior colliculus of Isl2-EphA3 knock-in mice, where the retinotopy is duplicated, whereas transcript levels of receptors, transporters and metabolic enzymes of the monoaminergic pathway were not affected. We demonstrate that the defect in response inhibition is a consequence of noradrenaline imbalance in the superficial layers of the superior colliculus caused by retinotopic map duplication. Our results suggest that structural abnormalities in the superior colliculus can cause defective response inhibition, a key feature of attention-deficit disorders.
KeywordsRetinotopy Visual system EphA signaling Superior colliculus Noradrenaline Response inhibition Attention-deficit disorders
The superior colliculus (SC) is a midbrain structure that integrates sensory inputs from multiple modalities (Wallace et al. 1993; Holmes and Spence 2005; May 2006) and plays a central role in visuo-spatial orientation, attention and sensorimotor processing (Stein 1984; May 2006; Gandhi and Katnani 2011). Defects in SC function have been associated with a number of neuropathological and neuropsychiatric disorders including epilepsy (Ross and Coleman 2000), schizophrenia (Fuentes 2001) and autism spectrum disorder (ASD) (Kleinhans et al. 2011). Recently, collicular hyperstimulation has been proposed to underlie attention-deficit/hyperactivity disorder (ADHD) symptoms, especially the impulsivity and distractibility associated with the disorder (Overton 2008; Miller 2009; Dommett et al. 2009). However, direct experimental evidence for such a link remains elusive.
The SC presents a particular feature, namely the topographic organization of its sensory inputs (Sperry 1963; Lemke and Reber 2005; May 2006). Axons of retinal ganglion cells (RGCs) project to the superficial layers of the SC along spatial axes reflecting their position along corresponding axes in the retina (the retino-collicular map). Layer V neurons of the V1 cortex also project in a topographic manner to the superficial layers of the SC, the cortico-collicular map, which is in register with the retino-collicular map (May 2006; Triplett et al. 2012). This creates a topographic representation of the visual field in the superficial layers of the SC, also called retinotopy. Auditory and somatosensory afferents projecting to deep layers of the SC are also aligned with the visual maps (Meredith and Stein 1985; King et al. 1998; May 2006) enhancing perception of salient stimuli and influencing decision and overt behavior (Stein et al. 2009).
To determine if hyperstimulation of the SC, due to duplication of the retinotopic projections, influences collicular-related behavior, wild-type (WT), heterozygous (EphA3KI/+) and homozygous (EphA3KI/KI) Isl2-EphA3KI mice were subjected to a series of well-established behavioral tests. As a first approach, we tested general visual ability (cliff test, optokinetic reflex, Morris water maze with visible platform) as the effects of disrupted EphAs gradients in the RGCs and duplicated retinotopy in the SC on visual perception have never been described before. We then focused on general sensorimotor (locomotor activity, circadian rhythmicity, light/dark box test) and integrative features (beam walking test) and on collicular-related behavior, especially visuo-spatial orientation and memory (Morris water maze with hidden platform) and response inhibition (Go/No-Go task). Our results show that EphA3KI/KI mutant mice exhibit defective response inhibition when compared to WT or EphA3KI/+ littermates. Visual acuity, sensorimotor activity, visuo-spatial learning, motivation and memory were similar in the different genotypes. Molecular characterization demonstrated elevated noradrenaline levels in the superficial layers of the SC in EphA3KI/KI animals where the retinotopy is duplicated. Expression levels of receptors, transporters and enzymes of the monoaminergic signaling pathway were similar to WT littermates. Interestingly, these changes resemble specific symptoms of the adult and predominantly inattentive-type of ADHD patients (Diamond 2005; Biederman and Faraone 2005).
Materials and methods
Mice were bred and housed in our mouse facility (Chronobiotron, UMS 3415, CNRS, Strasbourg) and tested during the light phase (ZT2–ZT10) of their light/dark cycle except for indicated experiments. All procedures were in accordance with national (council directive 87/848, October 1987) and European community (2010/63/EU) guidelines. Official agreement numbers for animal experimentation were 67-292 for CM, 67-215 for J-CC and 67-358 for KG, AG was under their responsibility. Mice were genotyped by PCR of genomic DNA from tail biopsies as described previously (Reber et al. 2004). Four- to seven-month-old male littermates of each genotype (EphA3KI/KI, EphA3KI/+ and WT) on a mixed genetic background (C57/Bl6 × 129Sv/J) were subjected to behavioral tests and molecular analyses. Standard laboratory rodent food and water were available ad libitum throughout all experiments, except for the Go/No-Go task, for which all mice were kept at 85 % of their free-feeding weight.
Three distinct cohorts of 4- to 7-month-old WT, EphA3KI/+ and EphA3KI/KI males littermates were characterized using fixed sequences of test ranging as much as possible from the least to the most invasive test. Inter-test intervals (ITI) varied along the sequences to limit order effect. The first cohort of 4- to 7-month-old males littermates (n = 6–9 per group) was first tested in the light/dark box test (Boeuf et al. 2009) (ITI 5 days) and then only in the Go/No-Go task (Meziane et al. 1993). The second cohort of 4- to 7-month-old males littermates (n = 7 per group) was dedicated to sensorimotor evaluations. They were first tested for circadian wheel running activity (Mendoza et al. 2008) and general locomotor activity (Yassine et al. 2013) (ITI 15 days) followed by the Morris water maze paradigm (Moreau et al. 2008) (ITI 15 days), the beam walking test (Moreau et al. 2008) (ITI 3 days) and the visual cliff test (Gibson and Walk 1960) (ITI 21 days). The optokinetic reflex (Douglas et al. 2005) was studied on a third cohort of 4-month-old (n = 7–10) male littermates. Detailed descriptions can be found in Online resource 1.
Transcript levels were analyzed by semi-quantitative PCR and monoamine levels were measured by high-pressure liquid chromatography as described in the Online resource 1.
Unless otherwise indicated, data were analyzed by analysis of variance with repeated measure factors to study interactions between genotype and side, trial, day, 15-min block, quadrant, runway (rANOVA). All statistical outcomes were confirmed by a Kruskal–Wallis test applied on the light–dark single factors or within each repeated measure, as group sizes in behavioral studies were relatively small. When required, post hoc analyses were performed with the Newman–Keuls (NK) multi-comparison test (Statistica 8.0; Statsoft, Inc., Tulsa, OK). The time spent in the goal quadrant of the water maze was compared to the 15-s chance value by means of a t test. The 15-s chance value corresponds to the time spent for random search in four quadrants during the 60 s probe test. All behavioral data are expressed as mean ± standard error of the mean (SEM). HPLC and qPCR data were analyzed using the non-parametric Kruskal–Wallis (KW) test. All expression data are represented using boxplots (min, q1, median, q3, max).
The functional contribution of the SC in specific behavior has been investigated in a variety of experiments, including electrophysiological recording, inactivation and lesion approaches (Binns 1999; Huberman and Niell 2011) but little has been done at a more integrated level in animal models with congenital defects.
General locomotor activity, sensory motor coordination and circadian rhythm
Visuo-spatial orientation and memory
Anxiety, response inhibition
In principle, this defective behavior could be caused by impaired attention or increased distraction (Barkley 2004). To test this possibility, we repeated the reinforced Go task, but added visual (flashing light) and auditory (tone) distractors. Mice of all genotypes showed significantly increased running times by reducing their speed in trials with tones (70 dB tone: F1,18 = 5.48, p < 0.05; 90 dB tone: F1,18 = 9.18, p < 0.01; Fig. 7c) and flash lights (F1,18 = 92.06, p < 0.0001; Fig. 7c) compared to non-distracted trials. Notably, all EphA3KI/KI mice increased their running times when exposed to a flashing light, (one mouse stopped to look toward the origin of the stimulus) although the difference between EphA3KI/KI and WT littermates did not reach statistical significance (Flash latency: F2,18 = 1.17, p = 0.33; Fig. 7c). These data indicate that a flashing light and loud tones are effective distractors during the Go task.
Analysis of regional monoamine levels
Our study provides first evidence for specific behavioral and molecular changes in mice with genetically altered retinotopy in the superior colliculus and consequently enhanced visual inputs. In the Go/No-Go task, EphA3KI/KI mice performed normally on Go trials by increasing their running speed, but they were completely unable to inhibit their running response on No-Go trials.
In the light/dark box test, EphA3KI/KI mice entered the aversive light box more readily than control mice. Altogether, our behavioral tests revealed that EphA3KI/KI mice exhibit defective response inhibition, a form of impulsivity. The observation that heterozygous EphA3KI/+ mice behave like WT littermates in the Go/No-Go task suggests that a partial duplication of the retino-collicular map (Brown et al. 2000) is not sufficient to trigger defective response inhibition. The observed behavioral changes were remarkably specific, as all other paradigms tested, namely vision, visuo-spatial orientation, sensorimotor function, motivation, learning and memory as well as exploratory behavior and anxiety were similar in WT, EphA3KI/+ and EphA3KI/KI mice. Defective response inhibition could be the consequence of enhanced levels of noradrenaline that we detected in the superficial layers of the SC of EphA3KI/KI mice. Enhanced noradrenaline levels in the SC could alter the behavior of the EphA3KI/KI mice by modulating the signal-to-noise ratio in this structure (Mooney et al. 1990; Tan et al. 1999) and thereby changing its level of activation (Dommett et al. 2009). In hamsters, in vivo and in vitro studies demonstrated a suppression of collicular neuron response upon noradrenaline application (Mooney et al. 1990; Tan et al. 1999). In rats, Sato and Kayama reported that iontophoretically applied noradrenaline exerts an excitatory action, indicating an increase of the signal-to-noise ratio, in accordance with our hypothesis (Sato and Kayama 1983). Whether noradrenaline increases or decreases the signal-to-noise ratio in the superficial layers of the SC is still debated. However, it clearly affects the processing of salient stimuli in a context-specific manner (Sato and Kayama 1983; Mooney et al. 1990; Tan et al. 1999).
The increase in noradrenaline was specific to the superficial layers of the SC, where the retinotopy is duplicated. Moreover, the increase only concerned noradrenaline, whereas other monoamines including dopamine, serotonin and adrenaline showed similar concentrations for all genotypes and brain regions. The increase in noradrenaline was not accompanied by changes in transcript levels of genes involved in monoamine metabolism. Therefore, we hypothesize that the increase of noradrenaline in the superficial layers of the SC may be the consequence of the duplication of the RGCs projections, which are functional, as shown by optical intrinsic imaging (Triplett et al. 2009). Previous studies revealed that RGC axons release noradrenaline upon activation (Osborne and Patel 1985). Alternatively, the increase may come from a duplication of projections from the locus coeruleus (LC), the major source of noradrenaline in the brain, to the superficial layers of the SC (Takemoto et al. 1978; Fritschy et al. 1990). Whether LC projections to the SC are duplicated is unknown as the mapping of the LC to the SC is hindered by the small size and specific sub-nuclei organization of the LC. However, it appears possible given that cortico-collicular projections are also duplicated in the EphA3KI/KI animals although projecting V1 neurons do not express ectopic EphA3 (Triplett et al. 2009). RGCs project to different brains areas, including lateral geniculate nucleus (LGN) and non-image forming structures such as the suprachiasmatic nucleus (SCN), the medial tegmental nucleus (MTN) or the olivary pretectal nucleus (OPN). Triplett and colleagues show no targeting defects in the LGN of Isl2-EphA3 animals (Triplett et al. 2009). The same group recently demonstrated that among 1 % of RGCs projecting to the SCN (the intrinsically photoreceptive RGCs—ipRGCs), 3 % are Isl2-positive and that these SCN-targeting Isl2-positive RGCs only transiently innervate the SCN during the development (Triplett et al. 2014). MTN and OPN also show innervation by Isl2-positive RGCs at early postnatal stages which is pruned by P6 (Triplett et al. 2014). The behavioral and molecular changes in EphA3KI/KI mice including defective response inhibition and noradrenaline enrichment in the superficial layers of the SC phenocopy some of the symptoms observed in ADHD patients, specifically the adult and predominantly inattentive-type (Barkley 1997; Aron and Poldrack 2005; Biederman and Faraone 2005; Bekker et al. 2005; Fisher et al. 2011; American Psychiatric Association 2013). These symptoms are also main features of Autism Spectrum Disorder (ASD) (Murray 2010). Our findings support the hypothesis that adult ADHD patients present collicular hyperstimulation leading to the appearance of impulsivity and attentional impairments (Overton 2008; Miller 2009; Dommett et al. 2009). Moreover, they are in line with the idea that dysregulation of the central noradrenergic systems contributes to the pathophysiology of ADHD (Biederman and Spencer 1999). Currently, progress on the etiology, diagnosis and treatment of ADHD is hindered by the limited number of experimental models. Most of the available rodent models are based on impaired monoaminergic transmission (van der Kooij and Glennon 2007; Sontag et al. 2010) and present some of the phenotypic features of ADHD patients. Our findings suggest that EphA3KI/KI animals may serve as a new model to study ADHD pathology and complement the limited arsenal of ADHD/ADD-related experimental approaches to understand and treat these neuropsychologic diseases.
The authors thank Dr. Sophie Reibel-Foisset, Nicolas Lethenet and Laurence Huck (Chronobiotron, Unité Mixte de Service 3415, Centre National de la Recherche Scientifique, Strasbourg) for animal care and Pedwin Pallet for help with recordings of the optokinetic reflex. This work was supported by Partner University Fund (M.R.), Centre National de la Recherche Scientifique (CNRS) and Université de Strasbourg (UdS). Publication costs are supported by the Neurex network (TriNeuron – Program Interreg IV Upper Rhine) http://www.neurex.org.
Conflict of interest
The authors report no biomedical financial interests or potential conflicts of interest.
- American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders, 5th edn. doi: 10.1176/appi.books.9780890423349
- Diamond A (2005) Attention-deficit disorder (attention-deficit/hyperactivity disorder without hyperactivity): a neurobiologically and behaviorally distinct disorder from attention-deficit/hyperactivity disorder (with hyperactivity). Dev Psychopathol 17:807–825. doi: 10.1017/S0954579405050388 CrossRefPubMedCentralPubMedGoogle Scholar
- King AJ, Schnupp JW, Thompson ID (1998) Signals from the superficial layers of the superior colliculus enable the development of the auditory space map in the deeper layers. J Neurosci Off J Soc Neurosci 18:9394–9408Google Scholar
- Loos M, Staal J, Schoffelmeer ANM et al (2010) Inhibitory control and response latency differences between C57BL/6 J and DBA/2 J mice in a Go/No-Go and 5-choice serial reaction time task and strain-specific responsivity to amphetamine. Behav Brain Res 214:216–224. doi: 10.1016/j.bbr.2010.05.027 CrossRefPubMedGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.