Homozygous BH.7A/Ztm-ci3/ci3 rats show several abnormal behaviors when observed in the home cage or after transfer to a new environment (new cage or rotometer). These abnormal behaviors occur in phases or bursts either spontaneously or in response to stress, e.g., new environment and acoustic stimuli, and are characterized by circling behavior and locomotor hyperactivity (24). Circling consists of tight 360° rotations. A detailed analysis of the circling behavior in a new environment showed that the majority of ci3 rats (36/38  =  95%) exhibit a side-preference of at least 70% in their rotational behavior during a 5-min test trial (24). Repeated testing of the same rats on separate days demonstrated that the side-preference was consistent in almost 70% of the animals. Two subgroups were observed, i.e., animals with right-sided (38%) and left-sided (62%) rotational preferences. The average number of rotations of 38 ci3 rats in three subsequent 5-min trials was 24.7  ±  2.6 without any difference between genders (Fig. 3.1b). The behavioral abnormalities of ci3 rats persisted over the whole lifespan. The background strain (BH.7A(LEW)/Won) did not show any circling or other behavioral abnormalities in the new cage or rotometer (24).

Apart from bursts of hyperactivity and circling, no other behavioral abnormalities were observed in the ci3 mutants. They showed no obvious gait disturbances, e.g., ataxia, and no stereotypic head-movements, e.g., opisthotonus, which are typical for LEW/Ztm-ci2/ci2 rats. Only in some rare instances, opisthotonus was observed in a few ci3 rats, so that head movement is certainly not a typical abnormality in this rat mutant.

In additional experiments, ci3 rats were video-recorded in their home cage without disturbance by an investigator over two periods of 4 h each, one period during the dark and one during the light period (24)(Fig. 3.1c). Seventy-one percent of the ci3 rats exhibited spontaneous circling when recorded during the light phase. Average turning rate in these rats was 31 (range 4-208) full rotations in 4 h. During the dark phase, all ci3 rats exhibited intense circling with an average rate of 189 (range 7-561) full rotations in 4 h, the difference between day and night being highly significant (P  =  0.0071; Fig. 3.1c). The relatively low circling rate during the day could be markedly enhanced by transferring the rats to a new environment, resulting in rotational behavior of all ci3 rats with an average turning rate of 22 full rotations over a 5-min observation period in three subsequent trials. This indicates, that stress such as transfer to a new environment markedly enhances turning rates during the light (inactive) period in these rats. During the dark phase, rotational behavior always occurred when the rats were active, i.e., the behavior occurred in bursts during normal forward locomotion. Before and after such a salvo of circling, rats behaved normal. The average ­side-preference of turning during the night was 76% (24).

As described above, both mutants show bursts of lateralized circling behavior, but the locomotor hyperactivity is much more intense in the ci2 mutant (32). Circling rates are comparable in both mutants. Figure 3.1c shows a comparison of the spontaneous rotational behavior of the ci2 and ci3 mutants during the light and dark phase. Ci2 rats tend to show more rotations during the night than ci3 rats, but the difference is not statistically significant (P  =  0.0810). In addition to circling and hyperactivity, ci2 rats exhibit gait disturbances, e.g., ataxia, and stereotypic head-movements, e.g., opisthotonus, which are not seen in ci3 rats (Table 3.1). Furthermore, ci2 rats exhibited marked impairment in the wire hang and rotarod/accelerod tests of sensorimotor functions, which was not determined in ci3 rats (32). However, the most striking behavioral difference was disclosed during swimming tests (Table 3.1). Whereas heterozygous ci2/+ rats and LEW/Ztm rats displayed normal swimming behavior resulting in proper orientation of the rats with respect to water surface, ci2/ci2 mutant rats showed abnormal swimming behavior associated with lack of orientation: when placed in water, the mutants spiraled underwater (in a corkscrew fashion) unable to maintain their noses above the water surface (19). Mutants needed to be rescued promptly to prevent drowning. In analogy to other behavioral abnormalities (lateralized circling, hyperactivity, ataxia, opisthotonus), the swimming inability of the ci2/ci2 mutant rats did not change with age as indicated by tests in rats of different age up to 1.5 years. The swimming inability and ataxia of the ci2/ci2 rats suggested abnormal vestibular function, which was subsequently confirmed (see below). In contrast, ci3/ci3 mutant rats displayed normal swimming behavior resulting in proper orientation of the rats with respect to water surface. However, the ­normal swimming behavior along the perimeter of the basin, which was seen in both ci3 rats and unaffected controls of the BH.7A(LEW)/Won background strain, was interrupted in ci3 rats by phases during which the rats swam in tight circles, which was not observed in controls (24).

In view of the swimming inability of the ci2 mutant and the similarities in behavioral alterations with the deaf stargazer (stg) rat mutant described by Truett et al. (15) in the Zucker strain, the auditory and vestibular systems of the ci2 rat were investigated by an integrative behavioral, neurophysiological, and neuroanatomical approach (19). Recording of brainstem auditory evoked potentials demonstrated a complete hearing loss in the adult ci2/ci2 mutant rat, whereas heterozygous (ci2/+) littermates exhibited normal auditory evoked potentials. In tests for vestibular dysfunction in rats with inner ear defects, i.e., the air-righting reflex test (33) and the tail-hanging test (34), ci2/ci2 exhibited marked deficits in both tests, whereas both heterozygous ci2/+ rats and normal LEW rats showed a normal air-righting reflex and landing response in the tail-hanging test (35). Histologic analysis of the inner ear of adult mutants revealed virtually complete loss of the cochlear neuroepithelium, whereas no such hair cell degeneration was seen in the vestibular parts of the inner ear (19). However, at least some hair cells of the epithelia of the utriculus, sacculus, and the cristae ampullaris of the vestibular apparatus of the inner ear of adult ci2/ci2 rats showed a protrusion of the cytoplasm into the endolymphatic space. A similar defect has been described for the shaker-2 mouse, in which, besides other ultrastructural alterations, vestibular hair cells were observed protruding into the endolymphatic space (36). Such alterations in ci2/ci2 rats would be a possible explanation for the disturbed balance of mutant rats as exemplified by the ataxia and the inability to swim. However, because such alterations were observed bilaterally, it is difficult to explain lateralized circling by these vestibular defects. The histological findings in mutant circling rats strongly indicate that the hearing loss of the mutants is of the sensory neural type, the most prevalent type of hearing loss. In the cochlear nuclei of the brainstem of mutant rats, neurons exhibited an abnormal shape, reduced size, and increased density compared to controls. In contrast, no abnormal neuronal morphology was seen in the vestibular nuclei, but a significantly reduced neuronal density was found in the medial vestibular nucleus (19).

In ci3 rats, brainstem auditory evoked potential testing and different tests of vestibular function did not disclose any auditory or vestibular defects (24). Although some ci3 rats showed abnormalities in the air-righting reflex and tail-hanging tests, most ci3 rats behaved normally in these tests, and all ci3 rats could swim, excluding any marked defect of the vestibular system (24). Furthermore, no morphological abnormalities were seen during histological examination of the cochlear and vestibular nuclei in the brainstem (Table 3.1).

Ontogenetic studies in LEW/Ztm-ci2/ci2 indicated vestibular hair cell loss in young (about 4 weeks of age) mutant rats, but hair cell regeneration during further development, so that vestibular hair cells appeared morphologically normal in adult rat mutants (19). However, as described above, part of the vestibular hair cells of adult ci2/ci2 showed protrusions into the endolymphic space, suggesting alterations in the cytoskeletal architecture (19). Furthermore, reduced neuronal density was determined in the medial vestibular nucleus of mutant rats. In view of the various efferent connections of the vestibular nuclei (37) and the finding that vestibular information is transmitted not only to the cerebellum, cortex, and spinal cord but also, via the thalamus, to the striatum (38), it is conceivable that abnormalities in vestibular nuclei during brain development of ci2/ci2 rats lead to secondary changes in brain regions such as the striatum thought to be critically involved in circling behavior (5, 7). This view is in line with previous observations of Alleva and Balazs (39, 40) using the ototoxic drug streptomycin in rats. Whereas this drug induced head tremor and difficulty with the righting reflex in adult rats, administration of streptomycin to postnatal rats induced circling, hyperactivity, repetitive head movements, and backward locomotion. Alleva and Balazs (39) concluded from their experiments with streptomycin-induced auditory and vestibular defects in rats that the neurological syndrome developing in postnatal rats involves both the vestibular apparatus and higher motor centers and that particularly the hyperactivity and circling are suggestive of a central site of action, presumably involving a dopaminergic mechanism. In a subsequent study, this group found increased dopamine D2 receptor binding in the striatum of dyskinetic rats that were treated subacutely with streptomycin as neonates (41). In ci2/ci2 mutant rats, we also found increased dopamine receptor binding in the striatum (42). Thus, the data from these different models seem to suggest that vestibular defects during postnatal development, independent of whether induced or inherited, lead to secondary changes in the dopaminergic system within the basal ganglia, which would be a likely explanation for the typical behavioral phenotype seen in these models.

For further evaluation of this interesting hypothesis, we adapted the protocol that Alleva and Balazs (39) used in neonatal Sprague-Dawley rats to neonatal LEW rats in order to directly compare the phenotype induced by streptomycin in LEW rats with that of the ci2 LEW rat mutant. For this purpose, we treated neonatal LEW rats over 3 weeks by streptomycin, which induced bilateral degeneration of cochlear and vestibular hair cells (35). Following this treatment period, the behavioral syndrome of the streptomycin-treated animals, including the lateralized rotational behavior, was almost indistinguishable from that of ci2 mutant rats. However, in contrast to the ci2 mutant rat, all alterations, except the hearing loss, were only transient, disappearing between 7 and 24 weeks following treatment. In conclusion, in line with our hypothesis, vestibular defects induced in normal LEW rats led to the same phenotypic behavior as the inherited vestibular defect of ci2 mutant rats. However, with increasing time for recovery, adaptation to the vestibular impairment developed in streptomycin-treated rats, whereas all deficits persisted in the mutant animals.

Based on the comparison between streptomycin-treated and ci2 mutant rats, we proposed a hypothesis of circling behavior in rodents with inner ear defects that integrates the role of abnormal striatal dopaminergic activity in circling behavior. As shown by several other groups, vestibular defects are associated with increased dopamine levels or turnover and increased dopamine receptor density in the striatum (12, 41–44), most likely as a result of alterations of the ascending vestibular pathway to the basal ganglia. Whether the alterations in striatal dopaminergic activity are asymmetric after aminoglycoside treatment is not yet known. The behavioral phenotypes developing after streptomycin or by the ci2 mutation are almost indistinguishable, consisting of both vestibular (e.g., abnormal swimming pattern, opisthotonus) and hyperdopaminergic (circling, hyperactivity) symptoms. Interestingly, these behaviors disappear over time following streptomycin but not in the ci2 mutant, the reasons of which need to be further explored. At least in part, the transient nature of the abnormal behaviors resulting from treatment with streptomycin could be explained by adaptation to the vestibular impairment by the use of visual cues, which is not possible in ci2 rats because of progressive retinal degeneration in these mutants (see below). Although further experiments are needed to prove this hypothesis, our experiments show that direct comparisons between these two models serve to understand the mechanisms underlying the complex behavioral phenotype in rodents with vestibular defects and how these defects are compensated. Our data and previous observations in other circling rat mutants (21, 24) suggest that the concept of circling behavior in deaf rodent mutants being simply a direct consequence of peripheral vestibular defects needs to be re-evaluated.

During behavioral testing of ci2/ci2 rats, including the visual cliff test (45) the impression arose that these rats could have visual defects. Therefore, electroretinograms (ERGs) were recorded from homozygous ci2/ci2 and heterozygous ci2/+ rats as well as normal LEW/Ztm rats of the background strain (46). The ERG protocols included flash and flicker stimuli under scotopic and phototopic conditions. In mutant rats, a functional disturbance of photoreceptors was observed, which progressed with age. In adult homozygotes, light-evoked retinal responses under rod-isolated (dark-adapted, dim stimulus), and cone-isolated conditions (light-adapted, flicker stimulus) were not present at all or only observed as small remaining responses. Age-matched LEW/Ztm control animals showed normal scotopic and phototopic ERG responses. Interestingly, significantly reduced amplitudes of the scotopic a- and b-waves were also observed in heterozygous ci2/+ rats. Histological examination of the retinal circumference from both eyes showed complete loss of the outer segments and the outer nuclear layer of the peripheral retina in ci2 mutants with visual defects. Furthermore, the thickness of the outer segments and the outer nuclear layer of the central retina was decreased, whereas retinal pigment epithelium and chorioid did not show any abnormalities (47). The complete loss of receptor outer segments and the progressive loss of the outer nuclear layer of the retina suggest that the retinal phenotype of ci2 rats results from a primary rod-cone defect, resembling retinitis pigmentosa. However, not all adult ci2/ci2 or ci2/+ rats exhibited such marked functional and morphological retina alterations, but the genetic expressivity of the retinal dystrophy was about 60%, compared to the 100% expressivity of circling and other behavioral alterations in homozygous ci2/ci2 rats. In addition, as yet unknown environmental factors seemed to affect the occurrence of visual defects in ci2/ci2 rats (91). Furthermore, animals with marked functional and morphological retina alterations were not completely blind, but could discriminate between light and dark as indicated by experiments on circadian rhythms in such rats. In ci3 rats, we have no indication of any visual defect.

Most of the behavioral differences between LEW/Ztm-ci2/ci2 and BH.7A/Ztm-ci3/ci3 (Table 3.1), particularly the ataxia and swimming inability that was observed in ci2 but not ci3 rats, could be explained by the vestibular defects of ci2 mutants. However, the fact that the lateralized circling behavior of ci2/ci2 and ci3/ci3 rats was indistinguishable seemed to indicate that a common denominator causes the lateralized rotational behavior in both rat mutants. The most likely explanation for such lateralized circling is an imbalance of forebrain dopamine systems, particularly an imbalance of nigrostriatal function (5, 10). Therefore, various morphological, neurochemical, and functional experiments were performed to characterize the dopamine system in the two mutants.

Ci2 mutant rats of both genders had a significantly higher tissue content of dopamine in the striatum contralateral to the preferred direction of rotation (22, 42). Bilateral striatal microdialysis in freely moving ci2 rats showed that circling was associated with a significant increase in dopamine release in the contralateral striatum, substantiating that the rats turn away from the brain hemisphere with higher striatal dopaminergic activity (48). Histological examination of the striatum and substantia nigra pars compacta failed to disclose any morphological abnormality in ci2 rats, whereas a significant asymmetry in the density of dopaminergic neurons was determined in the ventral tegmental area (VTA) (42). Electrophysiological experiments demonstrated a significantly increased discharge rate and altered discharge pattern with burst-like firing in the pars reticulata of the substantia nigra (SNR) of ci2 mutants, indicating an abnormal basal ganglia output activity in these rats (49). Treatment of ci2 rats with dopamine receptor antagonists inhibited circling (50). All these data strongly indicate that dopaminergic abnormalities contribute to the behavioral phenotype of the ci2 mutant.

In ci3 rats, the dopamine level was lower in the striatum contralateral to the preferred direction of turning, which is in contrast to the ci2 rat, in which dopamine levels are higher in the contralateral striatum (22). The dopaminergic input to the striatum originates from the dopaminergic neurons of the VTA (A9), substantia nigra pars compacta (SNC; A10), and substantia nigra pars lateralis (SNL) (51). Thus, we examined whether there is any laterality in density of dopaminergic neurons in these brain regions of the mutant ci3 rat (24). No asymmetry was found in VTA, SNL or the lateral part of the SNC. However, in the medial SNC, which contains many more dopaminergic cells than the lateral part, a significant asymmetry was determined in the more caudal parts of this subregion of the SNC. Here, the density of dopaminergic neurons was lower in the side contralateral to the preferred direction of circling. Thus, this asymmetry in the medial SNC corresponded to the asymmetry in dopamine levels in the striatum of ci3 mutant rats. Hence, the ci3 rats circled toward the hemisphere with higher dopaminergic cell density in medial SNC and higher dopamine level in the striatum, i.e., these rats exhibited a contraversive circling preference.

It is often stated in the literature that rats generally circle away from the striatum with higher dopaminergic activity, i.e., exhibit an ipsiversive circling bias (5, 7, 8). However, this mainly stems from observations in the Ungerstedt model, in which an almost complete unilateral degeneration of nigro-striatal dopamine neurons is produced in rats by 6-hydroxydopamine (6-OHDA; for review see refs. (5, 52, 53)). In such rats, amphetamine, by increasing dopamine release predominantly in the intact striatum, induces ipsiversive circling (i.e., toward the lesioned side), whereas the dopamine receptor agonist apomorphine produces contraversive turning, which is explained by denervation-induced dopamine receptor supersensitivity in the lesioned striatum (54). With respect to spontaneous behaviors in 6-OHDA lesioned rats, rats initially exhibit rotational behavior toward the lesioned side, particularly when placed in a new environment, but occasionally 2 weeks after surgery, rats may burst into intense “paradoxical” rotation contralateral to the lesioned side as a response to stress (5). After subtotal lesions of the nigrostriatal system, which do not lead to denervation supersensitivity in the striatum, rats do not always circle as expected (54–56). Some rats fail to circle upon administration of apomorphine or amphetamine and ­others turn in the opposite direction, e.g., contraversively with amphetamine.

The direction of circling following subtotal lesions of the SNC appears to depend on the precise location of the lesion. Thus, it has been found that while lesions restricted to the medial part of the SNC cause rats to circle ipsiversively, laterally placed lesions induce contraversive circling (55–59). The mechanisms which determine the direction of circling produced by small lesions of medial and lateral SNC are not known, but may relate to the different projections arising from these sites (51, 60). In apparent contrast to the findings from lesions of medial SNC in normal rats, the ci3 mutant rats circle away from the hemisphere with reduced dopaminergic cell density in part of the medial SNC. This could indicate that functional differences exist across the rostro-caudal extension of the medial SNC (24).

For a number of normal, intact rat strains, spontaneous circling has been observed during the dark phase (1, 5, 7), but at much lower rates than found in mutant ci2 or ci3 rats. The neurobiological basis for nocturnal rotational behavior in unlesioned rats is incompletely understood. In several rat strains, small asymmetries in striatal dopamine levels or uptake have been described and these differences are thought to underlie nocturnal circling observed in such strains (1, 7). Shapiro et al. (61) have reported two kinds, or populations, of rats: those with their turning biases directed away from and those with their turning biases directed toward the hemisphere containing the striatum with the higher dopamine uptake. These data from unlesioned rats clearly argue against the prevailing view of the relationship between striatal dopamine and turning behavior, i.e., that a dopaminergic predominance in one striatum always “pushes” the rat toward the other side (61). Similar to the rat subpopulation turning toward the striatum with the higher dopamine uptake described by Shapiro et al. (61), the ci3 mutant rat circles toward the striatum with the higher dopamine level, but at much higher intensities than ever described for normal, unlesioned rats. Interestingly, some of the ci3 rats changed the direction of their lateral preference for circling from one test to another or between trials with and without stress, which has been previously reported for amphetamine-induced circling in normal rats (1).

The lower density of dopaminergic neurons in a part of the medial SNC in one hemisphere of ci3 mutant rats was not only significant vs. the other hemisphere in the same rats, but also compared to cell density in this part of the medial SNC in normal rats of the background strain (24). No such difference was found when the same subregions of the SNC were compared in ci2 mutant rats, but a significant asymmetry was found in the VTA of these mutants (42). The reason for the lowered density of dopaminergic neurons in the medial SNC contralateral to the preferred direction of circling in ci3 rats is not clear, but the difference to BH.7A(LEW)/Won controls may indicate a localized, discrete cell loss in this area. A topographically organized cell loss in the medial SNC has been previously reported in a rat mutant (AS/AGU) from an Albino-Swiss strain, but the cell loss was bilateral, not resulting in any circling behavior, but staggered gait, hind-limb rigidity, whole body tremor, and difficulty in initiating movements (62). The cell loss in the SNC of AS/AGU rat mutants was associated with a bilateral reduction of striatal dopamine levels of about 30% (63). Hypokinesia, such as observed in the AS/AGU rat mutant, is also observed when a normal rat receives bilateral lesions of the SNC (9), whereas the stereotypic circling induced by unilateral lesions or other unilateral manipulations of the SNC or occurring in the ci3 mutant rat resembles involuntary rapid movements as observed in hyperkinetic disorders.

Quantitative autoradiographic determination of binding densities of dopamine transporter and D1 and D2 receptors in several parts of the striatum and substantia nigra indicated that ci2 rats have a significantly higher binding density of dopamine transporter and D1 receptors in the striatum and D2 receptors in the substantia nigra than LEW/Ztm controls (42). In subsequent studies, we measured the densities of 12 neurotransmitter receptors in the basal ganglia and vestibular nuclei of adult circling mutants (ci2/ci2), non-circling littermates (ci2/+), and controls from the background strain (LEW/Ztm) and found that ci2/ci2 mutants, in addition to higher expression of dopamine transporter and receptors, show lower expression of GABA_A and higher expression of nicotinic cholinergic receptors in a number of regions compared to controls (64). Furthermore, several interhemispheric differences in receptor binding densities were determined in normal LEW/Ztm and homozygous ci2/ci2, but, surprisingly, not in heterozygous ci2/+ rats. In ci2/ci2 rats, we found a correlation between intensity of circling and adenosine A_2A receptor densities in nucleus accumbens and thalamic regions (64). However, the fact that circling and hyperactivity in ci2 mutants can be blocked by dopamine receptor antagonists indicates that these behaviors are predominantly due to the alterations in the dopamine system of these rats.

In a further study in homozygous ci2/ci2 and heterozygous ci/+ rats, brain uptake of [¹⁴C]-labeled deoxyglucose (2-DG) was used to screen for altered neural activity in various brain regions (65). 2-DG uptake was significantly decreased in the primary auditory cortex and superior colliculus of homozygous ci2 rats compared with heterozygous controls. These changes are obviously related to the sensory dysfunctions of the mutant rats.

In ci3 rats, quantitative autoradiography was used to examine the binding of [³H]SCH 23390, [³H]raclopride, and [³H]7-OH-DPAT (7-hydroxy-N,N-di-n-propyl-2-aminotetralin) to dopamine D1, D2, and D3 receptors, respectively, in various brain regions of ci3 rats and unaffected rats of the background strain (BH.7A(LEW)/Won)(66). In contrast to ci2 rats, no significant differences between ci3 rats and controls were obtained for D1 and D2 receptor binding in any region, but mutant rats differed from controls in dopamine D3 binding in several regions. The dopamine D3 receptor is expressed primarily in regions of the brain that are thought to influence motivation and motor functions (67–69). The D3 receptor has a 70-fold greater affinity for dopamine than D1 or D2 dopamine receptors, and is expressed on post- and presynaptic sites, where it can function as autoreceptor (70). The functional role of the D3 receptor is difficult to study because of its low abundance (approximately 1% of D2 receptors) and the lack of selective agonists and antagonists. Behavioral analysis of mutant mice lacking functional D3 receptors showed that such mice exhibit an increase in basal and ­novelty-induced locomotor activity, an effect not associated with anxiety state (71, 72). The hyperactivity in these D3 knockout mice could be enhanced by low doses of cocaine, which increases synaptic dopamine levels by preventing dopamine reuptake into dopaminergic nerve terminals (72). D3 knockout mice have extracellular levels of dopamine in the striatum twice as high as their wild-type littermates (73, 74), suggesting an important role of the D3 receptor in the control of basal extracellular dopamine levels (75). In ci3 rats, a significant decrease in D3 binding was seen in the shell of the nucleus accumbens, the islands of Calleja, and the subependymal zone. Furthermore, a significant laterality in D3 binding was determined in ci3 rats in that, binding was lower in the contralateral hemisphere in the shell of the nucleus accumbens and the islands of Calleja. These data indicate that alterations of dopamine D3 receptors may be involved in the behavioral phenotype of the ci3 rat, thus substantiating the findings from a recent genetic linkage analysis that indicated the D3 receptor gene as a candidate gene in this rat mutant (see below).

In ci2/ci2 rats, d-amphetamine significantly increased the number of full body turns in an open field or rotometer, whereas no full body turns were observed in LEW/Ztm controls (22). Apomorphine did not increase circling in ci2 rats, whereas circling behavior was enhanced by the N-methyl-d-aspartate (NMDA) receptor antagonist MK-801 (22), which is known to increase dopamine turnover in the striatum (76). Circling (and hyperactivity) was blocked by the dopamine D2 receptor antagonist raclopride in ci2 rats (50). Interestingly, ci2 rats were less susceptible than unaffected littermates to the cataleptogenic effects of haloperidol and raclopride (50). Catalepsy is often used as a generic term for “active” immobility responses that are characterized by a behavioral immobility, which is also associated with varying degrees of enhanced muscular rigidity and/or “waxy flexibility”; a most conspicuous behavioral sign is the sustaining of an awkward or unusual posture, such as maintaining paws on an elevated bar (77). The best-known drugs that cause catalepsy also block dopaminergic receptors. Indeed, a common assumption is that catalepsy is a reliable functional index of nigrostriatal dopaminergic activity (77). It is typically induced in rats by dopamine antagonists such as haloperidol or raclopride and can be prevented or dramatically reduced by bilateral lesions of the substantia nigra or striatum (77, 78). The findings with haloperidol and raclopride in ci2 rats suggest that these animals possess an alteration in the function or density of D2-like receptors. Indeed, a previous autoradiographic determination of [³H]raclopride binding to dopamine D2 receptors of homozygous ci2 rats showed that binding density was not different from controls in the striatum, but significantly higher in the substantia nigra (42). The enhanced density of D2 receptors in homozygous ci2 rats could explain that higher doses of haloperidol are needed to induce catalepsy in homozygous ci2 mutants compared to unaffected heterozygous littermates.

In analogy to the observations in ci2/ci2 rats, d-amphetamine, and MK-801 intensified circling in ci3/ci3 rats (24, 50). Furthermore, haloperidol antagonized circling in ci3 mutants. However, in contrast to the findings in ci2 rats, the cataleptogenic efects of haloperidol and raclopride were not reduced in ci3 rats, which can be explained by the normal expression of dopamine D2 receptors in these animals (66).

Initial breeding experiments in LEW/Ztm-ci2/ci2 rats indicated that circling in these animals is determined monogenetically (i.e., mediated by a single gene) by a recessive autosomal gene termed circling, ci (22). Chwalisz et al. (79) performed a genome-wide scan of a (LEW/Ztm-ci2 × BN/Ztm) F1 × LEW/Ztm-ci2 backcross population with anonymous microsatellite markers to analyze the genetics of this mutant rat. This linkage analysis resulted in a region of interest on chromosome 10 containing Myo15, which encodes for the unconventional myosin XVa (79). A mutation of this gene is also responsible for the circling behavior of the shaker-2 mouse (27) and autosomal recessive nonsyndromic hearing impairment (ARNSHI; DFNB3) in humans (80). Like other unconventional myosins, including myosin VIIa, myosin XVa mRNA, and protein are expressed in cochlear and vestibular hair cells and have a role in the formation and/or maintenance of the actin-rich structures of these cells (81). Furthermore, myosin XVa is essential for the graded elongation of stereocilia during their functional maturation (81), explaining the cochlear and vestibular hair cell degeneration observed in both the ci2 and shaker-2 mutants. Myosin VIIa is required for aminoglycoside accumulation in cochlear hair cells (82), but the role of myosin XVa, if any, in this respect is not known. Unconventional myosins are not only expressed by sensory hair cells but play also a role in brain development and cell motility (83–85), which could explain that gene mutations affecting these molecular motor proteins may induce a complex phenotype involving both deafness and neurological symptoms (80, 86). Furthermore, several unconventional myosins, including VIIa, are required for normal retinal function (87–89).