Department of Pharmacology, University of Minnesota, Minneapolis, MN, USA

Correspondence to: F H Burton, PhD, Department of Pharmacology, University of Minnesota, 6-120 Jackson Hall, 321 Church Street SE, Minneapolis, MN 55455-0217, USA. E-mail: burto006@umn.edu

The tic disorder Tourette's Syndrome (TS) and obsessive-compulsive disorder (OCD) are comorbid behavioral disorders, suggesting a shared but still unknown neuronal basis. To 'circuit-test' such behaviors, we previously engineered transgenic mice expressing a neuropotentiating protein (cholera toxin A1 subunit) within a cortical-limbic subset of dopamine D1-receptor expressing (D1+) neurons known to trigger glutamatergic excitation of orbitofrontal, sensorimotor, limbic and efferent striatal circuits thought to be hyperactive or affected in OCD and TS. These mice exhibited OCD-like behaviors including generalized behavioral perseveration and compulsion-like leaping and grooming-associated pulling and biting of skin and hair. We now report that these OCD-like mice, like humans, also exhibit comorbid TS-like behaviors, including juvenile-onset tics; increased tic number, complexity and flurries; increased tic severity in males; voluntary tic suppression; and tic responsiveness to a non-cataleptic TS+OCD drug therapy (clonidine, 0.01 mg kg^-1). These data suggest that hormonal gender differences, apart from the influence of genetic or autoimmune etiologic factors, may be sufficient to aggravate tic severity in human TS males compared to TS females. These data also proffer a precise neuronal basis for TS+OCD, wherein tics and primary compulsions or obsessions are evoked by hyperactivity of various cortical-limbic projection neurons' glutamatergic output to efferent targets like the striatum. The 'Cortical-limbic Glutamatergic Neuron' (CGN) neuronal circuit model merges formerly opposed neurotransmitter models of TS and OCD, and is consistent with new clinical reports of increased cortical hyperactivity, striatal glutamate and striatal inhibitory D2 receptors, and reduced striatal responsiveness, in these disorders.

Molecular Psychiatry (2002) 7, 617-625. doi:10.1038/sj.mp.4001144

glutamate; serotonin; norepinephrine; dopamine; Tourette; OCD; ADD; ADHD; trichotillomania; gender; D1 receptor; somatosensory

Introduction

The tic disorder Tourette's Syndrome (TS) and Obsessive- Compulsive Disorder (OCD) are genetic and/or autoimmune neurological syndromes^1,2,3 which are comorbid in 40-75% of patients initially diagnosed with either disorder, with comorbidity likeliest to occur in their childhood-onset forms.^4,5 While OCD is a heterogeneous disorder with varying presentations, usually including primary obsessions and secondary (obsession-dependent) compulsions, comorbid TS+OCD often also features primary (obsession-independent) compulsions. The spectrum of TS+OCD symptoms includes generalized perseveration or repetition of otherwise normal behaviors, as in OCD; biting, plucking or pulling at the skin or hair of others or themselves, as in OCD or the related compulsive disorder trichotillomania (TTM);⁶ and repetitive or ritualistic abnormal movements and motor or vocal tics, as in TS.^7,8 Although the molecular causes of these genetic and/or autoimmune neurological syndromes remain undetermined, the high comorbidity of OCD and TS implies a shared or overlapping neuronal basis,^4,5,9 one associated with either hyperactivity/ hyper-responsiveness or lack of inhibition of cortical orbitofrontal, somatosensory-sensorimotor and amygdalar regions and their efferent striatal circuits.^{10,11,12,13,14,15} Exactly which neurons in these regions actually evoke TS+OCD symptoms is also unknown. However, cortical glutamatergic output neurons are known to co-express dopamine D1 and serotonin 5-HT2 receptors, while their presynaptic cortical inhibitory interneurons and postsynaptic striatal neurons express dopamine D2 receptors,^16,17,18 possibly accounting for the therapeutic effects of manipulating these three neurotransmitters.^10,19

Animal models of cortical-limbic hyperactivity-induced OCD-like or TS-like behavior may be useful to predict the specific neuronal circuitry of such disorders, as well as to screen new potential therapeutic drugs for TS, OCD and TTM. We recently created such a model of OCD-like compulsive symptoms, the D1CT-7 transgenic mice,²⁰ using a neuropotentiating transgene that expresses the cAMP-elevating, stimulus-evoked neurotransmission-increasing intracellular A1 subunit of cholera toxin²¹ in only two small subsets of neurons in the adult CNS.²⁰ The first subset is dopamine D1 and serotonin 5-HT2a,c receptor co-expressing (D1/5-HT2+) glutamatergic pyramidal projection neurons in somatosensory/insular cortex layers II-III and piriform cortex layer II,^16,20,22 which are known to send lateral and descending projections that glutamatergically excite sensorimotor and orbitofrontal plus deeper-layer corticostriatal neurons' glutamate output.²⁰ The second subset is D1 + GABAergic interneurons in the intercalated nucleus (ICN) of the amygdala, which indirectly triggers amygdalar glutamate output and anxiety.²⁰ This chronic hyperactivation of some of the same orbitofrontal, somatosensory-sensorimotor, and limbic glutamatergic output neurons proposed to be hyperactive in OCD was shown to cause the D1CT-7 mice to engage in unique, OCD-like behaviors²⁰ that are distinct from drug-induced stereotypies or limbic seizure behaviors²³ and that, as in human OCD (and TS), are aggravated by stress.^24,25 These OCD compulsion-like abnormal behaviors of the D1CT-7 transgenic mice include episodes of generalized perseveration or repetition of all normal behaviors as in OCD, and non-aggressive hair and skin-pulling and biting of others or themselves during grooming, as in OCD or OCD-related TTM.²⁰ The mice also exhibit abnormal repeated climbing/leaping,²⁰ a behavior that falls within the spectrum of abnormal motor compulsions in both OCD and TS.^7,8

This study is an observational study of the behaviors of this unique mouse model to determine whether it models behaviors seen in human TS+OCD. We hypothesized that: (1) tics would be significantly more frequent in transgenic mice with the D1CT-7 genotype than in control mice without this genotype; (2) the D1CT-7 mice would have a higher percentage of non-shaking tics (indicating greater complexity) than control mice; (3) the D1CT-7 mice would have more flurry-associated tics than control mice; (4) male D1CT-7 mice would exhibit greater severity in one or more of these measures than female D1CT-7 mice; (5) ticcing would be present in juvenile, not just adult, D1CT-7 mice; (6) ticcing would be temporarily suppressed during more concentration-demanding activities. In addition, we report the results of an interventional study where we gave clonidine, an alpha-2 adrenergic agonist medication commonly used for tic suppression, to these transgenic mice. We hypothesized that clonidine would reduce tic frequency in D1CT-7 mice, independent of its effect on locomotion.

In this study we report findings that confirm these hypotheses, indicating that the D1CT-7 mice will be useful for understanding the neuronal basis of the overlapping compulsions and tics symptomatic of comorbid TS+OCD. We discuss how our findings imply that hyperactive output of cortical-limbic glutamatergic neurons is sufficient to evoke TS+OCD symptoms, supporting a 'Cortical-limbic Glutamatergic Neuron' (CGN) hyperactivity model of these disorders. We discuss how the CGN model of TS+OCD circuitry is consistent with the standard model of parallel cortico-striatal-thalamo-cortical circuits,²⁶ and incorporates older disparate glutamate, dopamine, norepinephrine and serotonin 'neurotransmitter' models of TS and OCD.^{2,10,15,27,28} We discuss how the CGN hyperactivity model of TS+OCD is consistent with recent transcranial magnetic stimulation (TMS) studies showing cortical hyperactivity, hyper-responsiveness or lack of inhibition in both TS and OCD,^13,14 and is also supported by three recent clinical neuroimaging studies: Magnetic resonance spectroscopy (MRS) showing increased striatal excitatory glutamate deposition in OCD and its normalization after drug therapy;¹⁵ positron emission tomography (PET) showing increased striatal inhibitory D2 receptor activity in TS;²⁹ and functional magnetic resonance imaging (fMRI) showing decreased striatal responsiveness in TS.³⁰ Finally, we discuss how comorbid TS, OCD and ADD/ADHD symptoms may be explained by a spectrum between hyperactivity and inactivity of cortical-limbic glutamatergic neurons.

Materials and methods

Subjects

To measure tic incidence, tic complexity, tic-flurry incidence and the influence on these parameters of gender in D1CT-7 transgenic (Tg) and nontransgenic control (C) mice, 68 adult males or females (n = 6 Tg males; 6 C males; 27 Tg females; 29 C females) were used (after ticcing was established in Tg males it was necessary to exclude them from follow-up drug-response studies, for continual breeding instead with C females, due to Tg females OCD/TTM-like compulsive biting of offspring and subsequent unsuitability for breeding). To measure juvenile tic incidence, nine 16-day-old mice (n = 5 Tg; 4 C) were used. To measure tic responsiveness to clonidine, 14 adult females (n = 8 Tg, 6 C) were used. Animals were naive to behavioral or drug assays prior to testing, and experiments were carried out with the investigators blinded to the animals' transgenic or control genotype status. Other than their behavioral characteristics, no phenotypic differences clearly differentiate Tg from C mice that could unblind the raters. Rating accuracy was confirmed by establishment of exact concordance in tic number and time-index between two independent blind raters, and by subsequent confirmation of tic detection by reanalysis of the videotapes at each time index. Animals were housed in groups of 2-5, allowed unrestricted access to food and water, and maintained in a constant temperature, 12 h day-night cycle. Experiments were conducted during the light phase of the cycle. Care was taken to ensure that the mice used in this study received no unnecessary discomfort. Animals were maintained and studies were carried out in accordance with the Animal Welfare Act and the NIH Guide for the Care and Use of Laboratory Animals, under the approval of the University of Minnesota Institutional Animal Care and Use Committee. The University animal facility is fully accredited by the American Association for the Accreditation of Laboratory Animal Care.

Quantification of mouse tics

Ticcing was analyzed from videotapes of transgenic (Tg) vs control non-transgenic (C) littermate male and female mice. Videotapes were made of a 15-min period beginning after 15 min of habituation to a new cage. Because no significant difference in any tic measure was exhibited between females that were placebo (i.p. saline)-injected (45-30 min prior to videotaping) and females that were uninjected (not shown), data were combined from both groups, while all males were placebo-injected. Tics were defined as any very brief (0.05-0.1 s, as determined by a duration of 1.5-3 frames in 30 fps videotape recordings) isolated head and/or body jerk or shake, other than those associated with acoustic startle or obvious shedding of litter visible on the coat. Tic incidence in transgenic vs control mice was quantified as the mean number of tics observed in individual mice during the 15-min videotape. Observers were blinded to genotype status. Using the logged times of each tic in the videotapes, the tics were subsequently replayed in slow motion to categorize their type of movement. The possible types of tic movement included shaking of the head or whole body (rotational vector); and jerking of the head, body or part of the body in one or more directions (linear vectors: up, down, left, right, forward, backward). The percentage of non-shaking tics was used as an indicator of tic complexity (with higher percentages of non-shaking tics reflecting more exhibition of other types of jerking tics). Tics were also defined as occurring within a 'tic flurry' if their logged time of occurrence was within 5 s of another tic.

Quantification of tics in juvenile mice

Tic incidence in juveniles was determined from videotapes of transgenic vs control non-transgenic littermate mice of both genders aged 16 days. Tics were defined as described above. The mice were videotaped for 15 min total as a group, with each animal separately identified by color tail markings. Tic incidence in transgenic vs control mice was quantified as the mean number of tics observed in individual mice over a 15 min interval, beginning after 15 min of habituation of the group to a new cage. Observers were blinded to genotype status.

Quantification of effect of clonidine on tics and locomotion

Clonidine (Sigma-RBI, Natick, MA, USA) was prepared by dissolving the drug in 0.9% saline. All administrations of clonidine or saline (vehicle) were delivered intraperitoneally (i.p.) at an experimentally determined (see Results) non-sedating dose of 0.01 mg kg^-1, a dose also reported to be effective in suppressing serotonergic drug-induced head-twitches.³¹ Clonidine was administered in an injection volume of 10 ml kg^-1 body weight. Mice were individually videotaped over 15 min, after 15 min of habituation to a new cage and 30 min after clonidine or placebo (i.p. saline) injection. The videotapes were viewed for assessment of tic incidence as described above. The videotapes were then reexamined to quantify the extent of locomotion (measured as the mean number of cage midline crossings during the 15 min observation), to confirm that clonidine was reducing tics by a mechanism other than non-selective locomotor reduction (ie, by a non- cataleptic, non-sedating action).

Statistics

Overall significance of transgenic genotype or gender on adult tic incidence, expressed as mean tic counts + standard error of the mean (SEM), the mean + SEM percentage of non-shaking tics (as a measure of tic complexity), or the mean + SEM percentage of flurry-associated tics, was established by factorial analysis of variance (ANOVA). When factors of interest were shown significant by ANOVA, unpaired 2-tailed Student's t-tests were performed. Significance of transgenic genotype on tic incidence in juvenile mice was established by factorial ANOVA. Significance of genotype, vehicle or 0.01 mg kg^-1 (i.p.) clonidine treatment, and genotype ´ treatment interaction on extent of adult tics and locomotor activity was established by repeated measures ANOVA. This revealed a significant effect of transgenic genotype, treatment and genotype ´ treatment interaction on tic incidence (but not on locomotion), after which individual comparisons were made of the effects of genotype and treatment by unpaired and paired 2-tailed Student's t-tests, respectively. For the study of multiple tic measures and of tic aggravation by male gender, significance was assumed at P < 0.0167 for either ANOVA across genotype or ANOVA across gender, after Bonferroni correction for three tic measure comparisons; at P < 0.025 for two t-tests (or equivalent Mann-Whitney test if non-parametrically distributed data) across genotype (TgM vs CM; TgF vs CF) for any tic measure where genotype was significant by ANOVA; and at P < 0.05 for one t-test across gender (TgM vs TgF) for any tic measure where gender was significant by ANOVA. Significance was assumed at P < 0.05 for single comparisons of genotype or drug effect and at P < 0.025 for two unpaired t-tests in the clonidine drug study.

Results

OCD-like transgenic mice exhibit comorbid increased ticcing

Figure 1(a) shows that, in contrast to normal, non-transgenic control (C) littermate mice (which, like normal humans, twitch only infrequently, exhibiting on average three twitches every 15 min), D1CT-7 transgenic (Tg) mice exhibit a five-fold elevated incidence of twitching (F(1,66) = 26.159; P < 0.0001; n = 33 Tg, 35 C; significant effect of transgenic vs control genotype by ANOVA after Bonferroni correction for multiple comparisons). This repetitive twitching (ie, ticcing) was observed in both Tg males (P = 0.0242; n = 6 Tg, 6 C; Mann-Whitney U-test) and Tg females (P < 0.0001; n = 27 Tg, 29 C; unpaired Student's t-test).

Transgenic mice show increased tic complexity

Figure 1(b) shows that D1CT-7 transgenic (Tg) mice also exhibit a significant, TS-like increase in tic complexity, shown by their two-fold or greater increase in the percentage of non-shaking tics, compared to the percentage of non-shaking twitches in control (C) mice (F(1,52) = 10.330; P = 0.0022; n = 25 Tg, 29 C; significant effect of genotype by factorial ANOVA after Bonferroni correction for multiple comparisons). This increase in tic complexity was observed in both Tg females (P = 0.0162; n = 19 Tg, 23 C; unpaired Student's t-test) and Tg males, although due to lower subject number the effect in Tg males only approached significance (P = 0.0714; n = 6 Tg, 6 C; unpaired Student's t-test). To illustrate one example of a non-shaking tic, a Tg female's head 'jerking' tic, exhibiting a downward-left linear vector of motion, is shown in three 33-millisecond videotape freeze-frames (Figure 2).

Transgenic and male transgenic mice show more tic flurries

Figure 1(c) shows that D1CT-7 transgenic (Tg) mice exhibit a significant TS-like increase in the percentage of tics that occur within tic flurries, compared to the percentage of flurry-associated twitches in control (C) mice (F(1,66) = 15.064, P = 0.0002; n = 33 Tg, 35 C; significant effect of genotype by factorial ANOVA after Bonferroni correction for multiple comparisons). This increase in tic flurries was observed in both Tg males (P = 0.0009; n = 6 Tg, 6 C; unpaired Student's t-test) and Tg females (P = 0.0128; n = 27 Tg, 29 C; unpaired Student's t-test). In addition, Figure 1(c) also shows that Tg males further exhibit a significant aggravation in tic flurries compared to Tg females (F(1,31) = 6.852, P = 0.0136; n = 6 TgM, 27 TgF; significant effect of gender by factorial ANOVA after Bonferroni correction for multiple comparisons, and unpaired Student's t-test).

Transgenic mouse tics in juveniles

Figure 3 shows that, similar to the pediatric presentation of TS, juvenile D1CT-7 transgenic (Tg) mice aged 16 days, not just adult mice, exhibit an increased incidence of tics when compared to control (C) mice. Factorial ANOVA showed a significant overall effect of genotype (F(1,7) = 8.177; P = 0.0244, n = 5 Tg, 4 C). At earlier postnatal times (<10 days), all mouse pups typically exhibit choreiform movements, until fine motor control gradually develops. During this earlier period, D1CT-7 Tg and control non-transgenic mouse movements appear indistinguishable (not shown), suggesting that the TS-like ticcing in Tg mice is either a juvenile-onset behavior or a juvenile-onset retention of immature neuronal activity.

Transgenic mouse tics are alleviated by a drug for TS+OCD

Figure 4(a) shows that clonidine (0.01 mg kg^-1, i.p.), a non-cataleptic drug that reduces ticcing in human TS+OCD, similarly reduces tics in D1CT-7 transgenic (Tg) mice. Placebo (saline-vehicle) treated (Veh) Tg mice show elevated tic counts compared to C mice (F(1,12) = 7.458; P = 0.0182 for genotype effect by repeated measures ANOVA; P = 0.0183 for Tg-Veh ´ C-Veh comparison by unpaired t-test; n = 8 Tg, 6 C), but upon clonidine (Clon) treatment show reduced tic counts (F(1,12) = 6.972; P = 0.0216 for genotype ´ clonidine interaction by repeated measures ANOVA; n = 8 Tg, 6 C; P = 0.0148 for Tg-Veh x Tg-Clon comparison by paired t-test; n = 8). In contrast to (a), Figure 4(b) shows that, while a stimulatory effect of Tg genotype on locomotor activity was confirmed (F(1,12) = 5.516; P = 0.0368, n = 8 Tg, 6 C; significant genotype effect by repeated measures ANOVA), which is consistent with prior observations of increased OCD-like repetitive locomotor and non-locomotor behaviors by these mice,^20,23,32 there was no observed suppressant effect of clonidine on Tg or control (C) mouse locomotion, ie, no genotype ´ clonidine interaction (F(1,12) = 1.206; P = 0.2937, n = 8 Tg, 6 C; non-significant by repeated measures ANOVA). This indicates that, as in human TS and TS+OCD, the tic-suppressing effect of clonidine in TS+OCD-like D1CT-7 Tg mice occurs in the absence of general locomotor inhibition (catalepsy) or sedation.

Discussion

D1CT-7 transgenic mice as a symptomatic, circuit model of human TS+OCD

Humans with comorbid TS+OCD exhibit, along with abnormally hyperactive/hyper-responsive cortical-limbic neuronal excitation of the striatum, several noteworthy symptoms that we hypothesized should thus also be present in our cortical-limbic neuropotentiated D1CT-7 transgenic mouse model¾symptoms that were indeed detected in this study, as discussed below. As such, these transgenic mice with engineered cortical-limbic neuron hyperactivity/hyper-responsiveness may represent, if not a genetic or autoimmune model, a symptomatic and drug response-predictive 'neuronal' or circuit model of TS+OCD symptoms:

First, while normal humans twitch only infrequently, humans with TS exhibit an increased incidence of twitching, ie, tics.^7,8 Similarly, unlike normal mice, which twitch infrequently, the transgenic D1CT-7 mice tic repeatedly, with five-fold more frequent twitches.

Second, in human TS, tics are frequently complex or variable, rather than a repetition of only one type of twitch.^7,8 The transgenic D1CT-7 mice's tics are likewise variable, consisting of multiple types of head and body jerks or shakes.

Third, in human TS, tics often also occur in flurries, with the severity of presentation also greater in TS males than TS females.^7,8 Similarly, tics in the D1CT-7 mice show a parallel gender difference in one measure of severity¾tic flurries. Tics in male D1CT-7 mice more often occur in flurries than do tics in female D1CT-7 mice.

Fourth, in humans, TS usually begins when the individual is still a juvenile, during or soon after completion of CNS maturation, rather than beginning later on in adulthood.^7,8 What triggers the early, pediatric onset of tic symptoms is unknown, but has been postulated to involve either genetically-programmed neurodevelopmental or neuronal signal transduction changes,^1,33 childhood autoimmune reactions³⁴ or prepubertal to pubertal hormonal changes.³⁵ Similarly, the onset of increased ticcing in D1CT-7 mice does not begin in adult or old transgenic mice, but in juvenile mice.

Fifth, in humans, tics in TS+OCD usually arise as sudden, out-of-context interruptions of other ongoing activities, but are also usually temporarily suppressible by engaging in a highly concentration-intensive activity.^7,8 Videotapes of the D1CT-7 mice's tics or tic flurries indicate that they likewise usually represent sudden, out-of-context interruptions of other ongoing activities (data not shown). Moreover, the activities interrupted by ticcing are usually those like locomotion or rearing, rather than more attention- or concentration-dependent activities like grooming or hanging from cage bars¾implying that the D1CT-7 mice, like humans with TS or TS+OCD, can suppress their ticcing during activities that are more difficult or demanding of concentration.

Last, in humans, tics in TS+OCD are reduced by D2 antagonist neuroleptics like haloperidol or sulpiride.¹⁹ We have previously shown that D2 antagonists similarly suppress symptomatic behavior in the D1CT-7 transgenic mice.³² However, such drugs act in part by tending to suppress all movement (ie, catalepsy), while other TS+OCD drug therapies can reduce ticcing without triggering catalepsy or sedation. One such non-cataleptic, tic-reducing drug is the alpha-2 agonist clonidine,^19,28,36 which, by reducing presynaptic norepinephrine (NE) influx to limbic and cortical targets, is thought to reduce cortical-limbic excitatory glutamatergic output to the striatum. Similar to clonidine's ability to reduce ticcing in human TS+OCD in the absence of any cataleptic effects, the TS-like ticcing in D1CT-7 transgenic mice is likewise reduced by clonidine, without concomitant suppression of the mice's locomotion. Although our finding is consistent with the partial effectiveness of clonidine in tic alleviation in TS and TS+OCD, one limitation of this study is that it has no implications for clonidine treatment of OCD. Indeed, there is no convincing clinical evidence for the usefulness of clonidine in alleviating compulsions in either OCD alone or when comorbid with TS. As to why clonidine alleviates tics but not compulsions in TS+OCD, any difference in how the parallel hyperactive circuits mediating tics and compulsions are connected to afferent neurons with different receptors for a drug (like clonidine) should cause each circuit to be differentially responsive to amelioration by that drug.

Gender-specific hormones may aggravate tic severity in male TS compared to female TS

The influence of autoimmune and genetic etiologic factors is well established in the origin of TS, and each may certainly also play a role in tic severity. However, our data suggest that the hormonal differences between genders may by itself be sufficient to aggravate tic severity in males and/or reduce tic severity in females, once tics are triggered by other causes.

For example, in human TS, ticcing severity and penetrance are slightly greater in males than in females.^7,8 This has been postulated to be due variously to a TS gene-by-gender interaction,^1,33 an autoimmunity-by-gender interaction,³⁴ or a hormone-by-gender interaction³⁵ (ie, tic facilitation in males, or tic suppression in females, by increasing levels of a gender-associated hormone during childhood, when TS symptoms usually begin). We've observed that male transgenic mice tic significantly more severely than do female transgenics in one measure¾tic flurries¾while likewise showing trends toward more tic complexity. This may help resolve the question of whether gender-associated hormones alone can be sufficient to increase the prevalence of TS symptoms in human males or decrease them in TS females. Because the D1CT-7 mice's tics are directly engineered by neuropotentiation, not by a TS-gene mutation or autoimmunity, evidence of increased tic severity in male transgenics supports the hypothesis that gender-specific hormones are all that is required to aggravate ticcing in TS males or reduce ticcing in TS females, once ticcing is caused by other etiologic factors.

Neuronal basis of comorbid TS+OCD-like symptoms

The transgenic D1CT-7 mice exhibit selective potentiation of a small subset of neurons in certain cortical-limbic CNS regions shown to be hyperactive in OCD and TS. Because they not only exhibit consequent juvenile-onset OCD-like symptoms,^{20,23,24,25,32} but, as shown here, also exhibit TS-like tics and therapeutic TS+OCD drug responses, this suggests that these cortical-limbic neuropotentiated transgenic mice are a neuronal or circuit model of comorbid TS+OCD. But what is this circuit?

These potentiated neurons are intermediate-layer somatosensory cortical D1/5-HT2+ pyramidal glutamatergic projection neurons whose descending and horizontal projections excite deeper-layer corticostriatal glutamatergic output neurons and lateral sensorimotor and orbitofrontal glutamatergic output neurons; and amygdalar D1+ interneurons likewise thought to indirectly trigger amygdalar glutamatergic neuron output to the striatum.²⁰ Based on this prior information, as well as more recent studies suggesting the TS+OCD-like mice have hyperactive forebrain glutamate output^37,38 and consequent glutamate-induced up- regulation of efferent striatal D2 receptors or receptor signaling,³² we propose a specific neuronal mechanism underlying both TS+OCD symptoms and their therapeutic drug responsiveness (Figure 5).

Our findings imply that, in both human TS+OCD and its D1CT-7 mouse neuronal model, TS and OCD symptoms are evoked by abnormal potentiation of various cortical-limbic D1/5-HT2+ neurons and consequent glutamatergic output, leading to excessive cortical-limbic glutamate excitation of efferent striatal as well as cortical neurons. This is proposed to both cause TS and OCD symptoms in humans and our mice, and also cause compensatory up-regulation of efferent striatal neurons' inhibitory D2 receptors and down-regulation of their cAMP levels, with consequent reduced striatal responsiveness to cortical stimulation. Regarding the wide spectrum of TS+OCD symptoms, we presume that hyperactivity or hyper-responsiveness of the same type of glutamatergic output neuron in the parallel circuits of different topographic regions of the cortex and amygdala²⁶ selectively mediates the somatosensory premonitions and tics, primary compulsions, and primary obsessions of TS+OCD.

This cortical-limbic glutamatergic neuron (CGN) model of TS+OCD is consistent with five new clinical findings. First, cortical transcranial magnetic stimulation (TMS) studies have shown evidence of cortical hyperactivity, hyper-responsiveness and reduced cortical inhibition in both TS and OCD.^13,14,39

Second, an MRS glutamate imaging study showed that OCD patients' symptoms were associated with abnormally increased levels of striatal glutamate, and that symptom alleviation by SSRI treatment was associated with normalization of their striatal glutamate concentration.¹⁵

Third, the CGN model also explains a PET study's finding in human TS+OCD patients (and a related study on our TS+OCD-like mice) which found evidence of elevated striatal D2 receptor function in the disorder.^29,32 Inhibitory striatal D2 receptors or signaling would likely become compensatorily up- regulated, or at minimum must remain fully active, to counteract ongoing glutamatergic hyperexcitation of the striatal D2+ neurons (Figure 5). The CGN model thus explains the apparent paradox of how TS+OCD patients and TS+OCD-like mice could have elevated striatal D2 receptor function but still remain highly responsive (rather than becoming insensitive) to D2 antagonists like haloperidol and sulpiride, or to striatal dopamine reducers like bromocriptine and pergolide.^19,40,41,42 This is because even modest antagonism of the inhibitory D2 receptors would free the motion-reducing D2+ striatal neurons to respond to the excessive glutamate, allowing them to suppress tics and mobility.

Fourth, the CGN model's proposed oppositional up-regulation of inhibitory striatal D2 receptors in response to glutamatergic hyperexcitation also explains findings of a third clinical fMRI study, showing reduced striatal responsiveness to cortical input in TS patients.³⁰

Fifth, the CGN neural model of TS+OCD also explains the reported therapeutic efficacy of 5-HT2 antagonists⁴³ and of the NE release-inhibiting drug clonidine¹⁹ in TS+OCD, because these drugs are thought to in turn reduce corticostriatal and amygdalostriatal glutamate output.^18,36

Relationship of ADD/ADHD and TS+OCD symptoms

The OCD-like generalized behavioral perseverance of these D1+ neuron-potentiated D1CT-7 transgenic mice²⁰ is opposite to a reported attention deficit disorder (ADD)-like behavior of D1 receptor-knockout mice, generalized behavioral truncation.⁴⁴ This suggests that fluctuation between hyperactivity/hyper-responsiveness and hypo-activity/hypo-responsiveness of cortical-limbic D1+ glutamatergic output neurons may respectively cause comorbid TS+OCD and ADD or attention-deficit hyperactivity disorder (ADHD) symptoms. Such a model also accounts for the reported worsening of tics by Ritalin,⁴⁵ as well as the reported efficacy in ADD/ADHD of NE-releasing drugs like (a)tomoxetine.^46,47 An antipodal glutamate model of non-TS-associated OCD and ADHD has also recently been proposed.⁴⁸.

Conclusion

Our present findings indicate that mice which have been engineered, by transgenic potentiation of cortical-limbic D1+ neurons, to mimic TS+OCD-associated cortical-limbic hyperactivity and excessive glutamate output to the striatum, exhibit not only OCD-like behaviors but also TS-like behavior. Their TS-like symptoms include increased tic number, complexity and flurries, more severe tic flurries in males, juvenile onset of ticcing, ability to temporarily suppress ticcing, and tic responsiveness to therapeutic TS+OCD drugs. These findings imply that excessive glutamate release from cortical-limbic output neurons triggers comorbid TS+OCD symptoms. Moreover, in light of recent findings of antipodal ADD-like behavior in D1 receptor-deficient mice, our data also imply that the opposite state¾insufficient glutamate release from these same cortical-limbic output neurons¾may trigger antipodal ADD/ADHD symptoms, analogous to the positive vs negative symptoms of schizophrenia. These findings support a cortical-limbic glutamatergic neuron (CGN) 'hyperactivity' model of TS+OCD vs 'hypoactivity' model of ADD/ADHD, pointing toward the development of future anti-glutamatergic TS+OCD pharmaceuticals and glutamate-enhancing ADD/ ADHD pharmaceuticals.

Acknowledgements

This work was supported by the TSA and NARSAD.

1 Pauls DL, Leckman JF. The inheritance of Gilles de la Tourette's syndrome and associated behaviors. Evidence for autosomal dominant transmission. N Engl J Med 1986; 315: 993-997. MEDLINE

2 Singer HS, Walkup JT. Tourette syndrome and other tic disorders. Diagnosis, pathophysiology, and treatment. Medicine (Baltimore) 1991; 70: 15-32. MEDLINE

3 Garvey MA, Giedd J, Swedo SE. PANDAS: the search for environmental triggers of pediatric neuropsychiatric disorders. Lessons from rheumatic fever. J Child Neurol 1998; 13: 413-923. MEDLINE

4 Leonard HL, Lenane MC, Swedo SE, Rettew DC, Gershon ES, Rapoport JL. Tics and Tourette's disorder: a 2- to 7-year follow-up of 54 obsessive-compulsive children. Am J Psychiatry 1992; 149: 1244-1251. MEDLINE

5 Frankel M, Cummings JL, Robertson MM, Trimble MR, Hill MA, Benson DF. Obsessions and compulsions in Gilles de la Tourette's syndrome. Neurology 1986; 36: 378-382. MEDLINE

6 Van Ameringen M, Mancini C, Oakman JM, Farvolden P. The potential role of haloperidol in the treatment of trichotillomania. J Affect Disord 1999; 56: 219-226. MEDLINE

7 The Tourette Syndrome Classification Study Group. Definitions and classification of tic disorders. Arch Neurol 1993; 50: 1013-1016.

8 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), 4th edn American Psychiatric Press: Washington, DC, 1994, p xxvii, 78-105, 886.

9 Sheppard DM, Bradshaw JL, Purcell R, Pantelis C. Tourette's and comorbid syndromes: obsessive compulsive and attention deficit hyperactivity disorder. A common etiology? Clin Psychol Rev 1999; 19: 531-552.

10 Kurlan R, Kersun J, Ballantine HT Jr, Caine ED. Neurosurgical treatment of severe obsessive-compulsive disorder associated with Tourette's syndrome. Mov Disord 1990; 5: 152-155. MEDLINE

11 Swedo SE, Pietrini P, Leonard HL, Schapiro MB, Rettew DC, Goldberger EL et al. Cerebral glucose metabolism in childhood-onset obsessive-compulsive disorder. Revisualization during pharmacotherapy. Arch Gen Psychiatry 1992; 49: 690-694. MEDLINE

12 Breiter HC, Rauch SL, Kwong KK, Baker JR, Weisskoff RM, Kennedy DN et al. Functional magnetic resonance imaging of symptom provocation in obsessive-compulsive disorder. Arch Gen Psychiatry 1996; 53: 595-606. MEDLINE

13 Greenberg BD, Ziemann U, Cora-Locatelli G, Harmon A, Murphy DL, Keel JC et al. Altered cortical excitability in obsessive-compulsive disorder. Neurology 2000; 54: 142-147. MEDLINE

14 Ziemann U, Paulus W, Rothenberger A. Decreased motor inhibition in Tourette's disorder: evidence from transcranial magnetic stimulation. Am J Psychiatry 1997; 154: 1277-1284. MEDLINE

15 Rosenberg DR, MacMaster FP, Keshavan MS, Fitzgerald KD, Stewart CM, Moore GJ. Decrease in caudate glutamatergic concentrations in pediatric obsessive-compulsive disorder patients taking paroxetine. J Am Acad Child Adolesc Psychiatry 2000; 39: 1096-1103. MEDLINE

16 Sheldon PW, Aghajanian GK. Excitatory responses to serotonin (5-HT) in neurons of the rat piriform cortex: evidence for mediation by 5-HT1C receptors in pyramidal cells and 5-HT2 receptors in interneurons. Synapse 1991; 9: 208-218. MEDLINE

17 Jakab RL, Goldman-Rakic PS. 5-Hydroxytryptamine2A serotonin receptors in the primate cerebral cortex: possible site of action of hallucinogenic and antipsychotic drugs in pyramidal cell apical dendrites. Proc Natl Acad Sci USA 1998; 95: 735-740. MEDLINE

18 Aghajanian GK, Marek GJ. Serotonin, via 5-HT2A receptors, increases EPSCs in layer V pyramidal cells of prefrontal cortex by an asynchronous mode of glutamate release. Brain Res 1999; 825: 161-171. MEDLINE

19 Cohen DJ, Riddle MA, Leckman JF. Pharmacotherapy of Tourette's syndrome and associated disorders. Psychiatr Clin North Am 1992; 15: 109-129. MEDLINE

20 Campbell KM, de Lecea L, Severynse DM, Caron MG, McGrath MJ, Sparber SB et al. OCD-Like behaviors caused by a neuropotentiating transgene targeted to cortical and limbic D1+ neurons. J Neurosci 1999; 19: 5044-5053. MEDLINE

21 Burton FH, Hasel KW, Bloom FE, Sutcliffe JG. Pituitary hyperplasia and gigantism in mice caused by a cholera toxin transgene. Nature 1991; 350: 74-77. MEDLINE

22 Wright DE, Seroogy KB, Lundgren KH, Davis BM, Jennes L. Comparative localization of serotonin1A, 1C, and 2 receptor subtype mRNAs in rat brain. J Comp Neurol 1995; 351: 357-373. MEDLINE

23 Campbell KM, McGrath MJ, Burton FH. Behavioral effects of cocaine on a transgenic mouse model of cortical-limbic compulsion. Brain Res 1999; 833: 216-224. MEDLINE

24 McGrath MJ, Campbell KM, Veldman MB, Burton FH. Anxiety in a transgenic mouse model of cortical-limbic neuropotentiated compulsive behavior. Behav Pharmacol 1999; 10: 435-443. MEDLINE

25 McGrath MJ, Campbell KM, Burton FH. The role of cognitive and affective processing in a transgenic mouse model of cortical-limbic neuropotentiated compulsive behavior. Behav Neurosci 1999; 113: 1249-1256. MEDLINE

26 Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986; 9: 357-381. MEDLINE

27 Cohen DJ, Shaywitz BA, Caparulo B, Young JG, Bowers MB Jr. Chronic, multiple tics of Gilles de la Tourette's disease. CSF acid monoamine metabolites after probenecid administration. Arch Gen Psychiatry 1978; 35: 245-250. MEDLINE

28 Hollander E, Fay M, Cohen B, Campeas R, Gorman JM, Liebowitz MR. Serotonergic and noradrenergic sensitivity in obsessive-compulsive disorder: behavioral findings. Am J Psychiatry 1988; 145: 1015-1017. MEDLINE

29 Wolf SS, Jones DW, Knable MB, Gorey JG, Lee KS, Hyde TM et al. Tourette syndrome: prediction of phenotypic variation in monozygotic twins by caudate nucleus D2 receptor binding. Science 1996; 273: 1225-1227. MEDLINE

30 Peterson BS, Skudlarski P, Anderson AW, Zhang H, Gatenby JC, Lacadie CM et al. A functional magnetic resonance imaging study of tic suppression in Tourette syndrome. Arch Gen Psychiatry 1998; 55: 326-333. MEDLINE

31 Matsumoto K, Mizowaki M, Thongpraditchote S, Murakami Y, Watanabe H. alpha2-Adrenoceptor antagonists reverse the 5-HT2 receptor antagonist suppression of head-twitch behavior in mice. Pharmacol Biochem Behav 1997; 56: 417-422. MEDLINE

32 Campbell KM, McGrath MJ, Burton FH. Differential response of cortical-limbic neuropotentiated compulsive mice to dopamine D1 and D2 receptor antagonists. Eur J Pharmacol 1999; 371: 103-111. MEDLINE

33 Cohen DJ, Leckman JF, Pauls D. Neuropsychiatric disorders of childhood: Tourette's syndrome as a model. Acta Paediatr Suppl 1997; 422: 106-111. MEDLINE

34 Swedo SE, Leonard HL, Mittleman BB, Allen AJ, Rapoport JL, Dow SP et al. Identification of children with pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections by a marker associated with rheumatic fever. Am J Psychiatry 1997; 154: 110-112. MEDLINE

35 Peterson BS, Leckman JF, Scahill L, Naftolin F, Keefe D, Charest NJ et al. Steroid hormones and CNS sexual dimorphisms modulate symptom expression in Tourette's syndrome. Psychoneuroendocrinology 1992; 17: 553-563. MEDLINE

36 Lichter DG, Jackson LA. Predictors of clonidine response in Tourette syndrome: implications and inferences. J Child Neurol 1996; 11: 93-97. MEDLINE

37 Campbell KM, Veldman MB, McGrath MJ, Burton FH. TS+OCD-like neuropotentiated mice are supersensitive to seizure induction. NeuroReport 2000; 11: 2335-2338. MEDLINE

38 McGrath MJ, Parks CRI, Burton FH. Glutamatergic drugs exacerbate symptomatic behavior in a transgenic model of comorbid Tourette's Syndrome and obsessive-compulsive disorder. Brain Res 2000; 877: 23-30. MEDLINE

39 Edgley SA, Lemon RN. Experiments using transcranial magnetic brain stimulation in man could reveal important new mechanisms in motor control. J Physiol (Lond) 1999; 521: 565. MEDLINE

40 Robertson MM, Schnieden V, Lees AJ. Management of Gilles de la Tourette syndrome using sulpiride. Clin Neuropharmacol 1990; 13: 229-235. MEDLINE

41 Ceccherini-Nelli A, Guazzelli M. Treatment of refractory OCD with the dopamine agonist bromocriptine. J Clin Psychiatry 1994; 55: 415-416. MEDLINE

42 Gilbert DL, Sethuraman G, Sine L, Peters S, Sallee FR. Tourette's syndrome improvement with pergolide in a randomized, double-blind, crossover trial. Neurology 2000; 54: 1310-1315. MEDLINE

43 Bonnier C, Nassogne MC, Evrard P. Ketanserin treatment of Tourette's syndrome in children. Am J Psychiatry 1999; 156: 1122-1123. MEDLINE

44 Cromwell HC, Berridge KC, Drago J, Levine MS. Action sequencing is impaired in D1A-deficient mutant mice. Eur J Neurosci 1998; 10: 2426-2432. MEDLINE

45 Castellanos FX, Giedd JN, Elia J, Marsh WL, Ritchie GF, Hamburger SD et al. Controlled stimulant treatment of ADHD and comorbid Tourette's syndrome: effects of stimulant and dose. J Am Acad Child Adolesc Psychiatry 1997; 36: 589-596. MEDLINE

46 Kratochvil CJ, Bohac D, Harrington M, Baker N, May D, Burke WJ. An open-label trial of tomoxetine in pediatric attention deficit hyperactivity disorder. J Child Adolesc Psychopharmacol 2001; 11: 167-170. Article MEDLINE

47 Michelson D, Faries D, Wernicke J, Kelsey D, Kendrick K, Sallee FR et al. Atomoxetine in the treatment of children and adolescents with attention-deficit/hyperactivity disorder: a randomized, placebo-controlled, dose-response study. Pediatrics 2001; 108: E83. MEDLINE

48 Carlsson ML. On the role of cortical glutamate in obsessive-compulsive disorder and attention-deficit hyperactivity disorder, two phenomenologically antithetical conditions. Acta Psychiatr Scand 2000; 102: 401-413. Article MEDLINE

Figure 1 Comorbid TS-like complex tics and tic flurries in cortical-limbic neuropotentiated OCD-like mice. Filled bars, D1CT-7 transgenic (Tg) mice; Open bars, control (C) non-transgenic littermates. (a) Increased incidence of tics in D1CT-7 transgenic mice. Data are shown as a bar graph of the mean number (+ SEM) of head or body tics occurring during videotaped 15 min windows of observation. *P < 0.05, ***P < 0.001 for between-group (Tg vs C genotype) comparisons in each gender (unpaired 2-tailed Student's t-test or Mann-Whitney U-test on non-parametrically distributed data), n = 6 Tg male, 6 C male, 27 Tg female, 29 C female. (b) Increased complexity of tics in D1CT-7 transgenic mice. Data are shown as a bar graph of the mean percentage (+ SEM) of non-shaking tics (ie, a tic-complexity measure based on the percentage of additional types of jerking tics). **P < 0.01 for between-group (Tg vs C genotype) ANOVA collapsed across gender (signified by underlining beneath asterisk), n = 25 Tg, 29 C; *P < 0.05 for between-group (Tg vs C) comparisons in each gender (unpaired 2-tailed Student's t-test), n = 6 Tg male, 6 C male, 19 Tg female, 23 C female. (c) Increased percentage of tic flurries in D1CT-7 transgenic mice and further elevation in male transgenic mice. Data are shown as a bar graph of the mean percentage (+ SEM) of flurry-associated tics (percentage of tics occurring within 5 s of another tic). *P < 0.05, ***P < 0.001 for between-group (Tg vs C genotype) comparisons in each gender (unpaired 2-tailed Student's t-test), n = 6 Tg male, 6 C male, 27 Tg female, 29 C female; ⁺P < 0.05 for between-group (Tg male vs Tg female) comparison across gender (unpaired 2-tailed Student's t-test), n = 6 Tg male, 27 Tg female.

Figure 2 Illustration of ticcing in TS+OCD-like mice. Shown are four sequential 1/30th sec (33 ms) video frames of a typical tic (duration 0.1 s) exhibited by TS+OCD-like D1CT-7 transgenic mice. Tic shown here is a 'head jerking' tic, whose downward-left linear vector evident in the sequential 33 ms frames is distinguishable from a forward-pointing rotational vector typical of head shaking tics.

Figure 3 Tics occur in juvenile TS+OCD-like mice. Data are shown as a bar graph of the mean number (+ SEM) of head or body twitches occurring during videotaped 15 min windows of observation, revealing elevated tics in transgenic mice. Abbreviations: Transgenic (D1CT-7 transgenic mice); Control (non-transgenic mice); *P < 0.05 for between-group (Tg vs C) comparison (unpaired Student's t-test; n = 5 Tg; 4 C).

Figure 4 TS+OCD-like mice's tics are reduced by a non- cataleptic TS+OCD drug treatment. (a) Clonidine (0.01 mg kg^-1) normalizes tics in D1CT-7 transgenic mice. Data are shown as a bar graph of the mean number (+ SEM) of head or body twitches occurring during videotaped 15 min windows of observation, revealing elevated tics in transgenic mice and reduction of their tics by clonidine treatment. (b) Tic reduction by clonidine is not associated with reduced locomotion. Data are shown as described above, revealing that tic reduction by clonidine does not involve reduced locomotor function. Abbreviations: Tg (D1CT-7 transgenic female mice); C (non-transgenic female mice); Veh (saline vehicle i.p. injection); Clon (0.01 mg kg^-1 clonidine i.p. injection); *P < 0.05 for between-group (Tg vs C) comparisons (unpaired Student's t-test); ⁺P < 0.05 for between-treatment (Veh vs Clon) comparison (paired Student's t-test); n = 8 Tg, 6 C.

Figure 5 Cortical-limbic glutamatergic neuron (CGN) hyperactivity model of TS+OCD symptoms. (a) Simplified schematic of normal corticostriatal/amygdalostriatal and indirect (D2) vs direct (D1) striatal motor output circuit. (b) Predicted changes in TS+OCD and TS+OCD-like D1CT-7 mice, wherein hyperactivity or hyper-responsiveness of the same type of glutamatergic output neuron in different topographic regions of cortex and amygdala mediates tics, primary compulsions, and primary obsessions. (c) Predicted subsequent changes in response to classical D2 antagonist neuroleptics (like sulpiride or haloperidol). (d) Predicted subsequent changes in response to either cortical-limbic norepinephrine release-inhibiting, stress-reducing alpha-2 agonists like clonidine, or serotonin receptor-desensitizing drugs like serotonin-selective reuptake inhibitors (SSRIs) or serotonin receptor antagonists like ketanserin. This cortical-limbic glutamatergic neuron (CGN) model of TS+OCD symptoms incorporates prior glutamate, dopamine, norepinephrine and serotonin neurotransmitter models of TS and OCD. Abbreviations: GLU, glutamate; CTX, cortex; AMY, amygdala; STR, striatum; D1, dopamine (DA) D1 receptors; D2, DA D2 receptors; NE, norepinephrine; 5-HT, serotonin. Symbols: triangles, excitatory glutamatergic cortical-limbic output neurons; filled triangle, hyperactivated neuron (due to D1CT-7 transgenic effect or human disorder); circles, striatal GABAergic neurons of the direct (D1) and indirect (D2) pathway; , neuron-exciting (glutamatergic) or motion-exciting neurotransmission; ---|, motion-inhibiting neurotransmission.