Cerebellar contribution to higher and lower order rule learning and cognitive flexibility in mice.

Cognitive flexibility has traditionally been considered a frontal lobe function. However, converging evidence suggests involvement of a larger brain circuit which includes the cerebellum. Reciprocal pathways connecting the cerebellum to the prefrontal cortex provide a biological substrate through which the cerebellum may modulate higher cognitive functions, and it has been observed that cognitive inflexibility and cerebellar pathology co-occur in psychiatric disorders (e.g., autism, schizophrenia, addiction). However, the degree to which the cerebellum contributes to distinct forms of cognitive flexibility and rule learning is unknown. We tested lurcher↔wildtype aggregation chimeras which lose 0-100% of cerebellar Purkinje cells during development on a touchscreen-mediated attentional set-shifting task to assess the contribution of the cerebellum to higher and lower order rule learning and cognitive flexibility. Purkinje cells, the sole output of the cerebellar cortex, ranged from 0 to 108,390 in tested mice. Reversal learning and extradimensional set-shifting were impaired in mice with⩾95% Purkinje cell loss. Cognitive deficits were unrelated to motor deficits in ataxic mice. Acquisition of a simple visual discrimination and an attentional-set were unrelated to Purkinje cells. A positive relationship was observed between Purkinje cells and errors when exemplars from a novel, non-relevant dimension were introduced. Collectively, these data suggest that the cerebellum contributes to higher order cognitive flexibility, lower order cognitive flexibility, and attention to novel stimuli, but not the acquisition of higher and lower order rules. These data indicate that the cerebellar pathology observed in psychiatric disorders may underlie deficits involving cognitive flexibility and attention to novel stimuli.

Impaired hypercarbic and hypoxic responses from developmental loss of cerebellar Purkinje neurons: implications for sudden infant death syndrome.

Impaired responsivity to hypercapnia or hypoxia is commonly considered a mechanism of failure in sudden infant death syndrome (SIDS). The search for deficient brain structures mediating flawed chemosensitivity typically focuses on medullary regions; however, a network that includes Purkinje cells of the cerebellar cortex and its associated cerebellar nuclei also helps mediate responses to carbon dioxide (CO2) and oxygen (O2) challenges and assists integration of cardiovascular and respiratory interactions. Although cerebellar nuclei contributions to chemoreceptor challenges in adult models are well described, Purkinje cell roles in developing models are unclear. We used a model of developmental cerebellar Purkinje cell loss to determine if such loss influenced compensatory ventilatory responses to hypercapnic and hypoxic challenges. Twenty-four Lurcher mutant mice and wild-type controls were sequentially exposed to 2% increases in CO2 (0-8%) or 2% reductions in O2 (21-13%) over 4 min, with return to room air (21% O2/79% N2/0% CO2) between each exposure. Whole body plethysmography was used to continuously monitor tidal volume (TV) and breath frequency (f). Increased f to hypercapnia was significantly lower in mutants, slower to initiate, and markedly lower in compensatory periods, except for very high (8%) CO2 levels. The magnitude of TV changes to increasing CO2 appeared smaller in mutants but only approached significance. Smaller but significant differences emerged in response to hypoxia, with mutants showing smaller TV when initially exposed to reduced O2 and lower f following exposure to 17% O2. Since cerebellar neuropathology appears in SIDS victims, developmental cerebellar neuropathology may contribute to SIDS vulnerability.

19p13.2 microduplication causes a Sotos syndrome-like phenotype and alters gene expression.

Up to 90% of individuals affected by Sotos syndrome have a pathogenic alteration of NSD1 (encodes nuclear receptor-binding Su-var, enhancer of zeste, and trithorax domain protein 1), a histone methyltransferase that functions as both a transcriptional activator and a repressor. Genomic copy number variations may also cause a Sotos-like phenotype. We evaluated a three-generation family segregating a Sotos-like disorder characterized by typical facial features, overgrowth, learning disabilities, and advanced bone age. Affected individuals did not have a detectable NSD1 mutation, but rather were found to have a 1.9 Mb microduplication of 19p13.2 with breakpoints in two highly homologous Alu elements. Because the duplication included the DNA methyltransferase gene (DNMT1), we assessed DNA methylation of peripheral blood and buccal cell DNA and detected no alterations. We also examined peripheral blood gene expression and found evidence for increased expression of genes within the duplicated region. We conclude that microduplication of 19p13.2 is a novel genomic disorder characterized by variable neurocognitive disability, overgrowth, and facial dysmorphism similar to Sotos syndrome. Failed compensation of gene duplication at the transcriptional level, as seen in peripheral blood, supports gene dosage as the cause of this disorder.

High-throughput behavioral phenotyping in the expanded panel of BXD recombinant inbred strains.

Genetic reference populations, particularly the BXD recombinant inbred (BXD RI) strains derived from C57BL/6J and DBA/2J mice, are a valuable resource for the discovery of the bio-molecular substrates and genetic drivers responsible for trait variation and covariation. This approach can be profitably applied in the analysis of susceptibility and mechanisms of drug and alcohol use disorders for which many predisposing behaviors may predict the occurrence and manifestation of increased preference for these substances. Many of these traits are modeled by common mouse behavioral assays, facilitating the detection of patterns and sources of genetic coregulation of predisposing phenotypes and substance consumption. Members of the Tennessee Mouse Genome Consortium (TMGC) have obtained phenotype data from over 250 measures related to multiple behavioral assays across several batteries: response to, and withdrawal from cocaine, 3,4-methylenedioxymethamphetamine; "ecstasy" (MDMA), morphine and alcohol; novelty seeking; behavioral despair and related neurological phenomena; pain sensitivity; stress sensitivity; anxiety; hyperactivity and sleep/wake cycles. All traits have been measured in both sexes in approximately 70 strains of the recently expanded panel of BXD RI strains. Sex differences and heritability estimates were obtained for each trait, and a comparison of early (N = 32) and recent (N = 37) BXD RI lines was performed. Primary data are publicly available for heritability, sex difference and genetic analyses using the MouseTrack database, and are also available in GeneNetwork.org for quantitative trait locus (QTL) detection and genetic analysis of gene expression. Together with the results of related studies, these data form a public resource for integrative systems genetic analysis of neurobehavioral traits.

Next generation tools for high-throughput promoter and expression analysis employing single-copy knock-ins at the Hprt1 locus.

We have engineered a set of useful tools that facilitate targeted single copy knock-in (KI) at the hypoxanthine guanine phosphoribosyl transferase 1 (Hprt1) locus. We employed fine scale mapping to delineate the precise breakpoint location at the Hprt1(b-m3) locus allowing allele specific PCR assays to be established. Our suite of tools contains four targeting expression vectors and a complementing series of embryonic stem cell lines. Two of these vectors encode enhanced green fluorescent protein (EGFP) driven by the human cytomegalovirus immediate-early enhancer/modified chicken beta-actin (CAG) promoter, whereas the other two permit flexible combinations of a chosen promoter combined with a reporter and/or gene of choice. We have validated our tools as part of the Pleiades Promoter Project (http://www.pleiades.org), with the generation of brain-specific EGFP positive germline mouse strains.

CAG repeat lengths > or =335 attenuate the phenotype in the R6/2 Huntington's disease transgenic mouse.

With spontaneous elongation of the CAG repeat in the R6/2 transgene to > or =335, resulting in a transgene protein too large for passive entry into nuclei via the nuclear pore, we observed an abrupt increase in lifespan to >20 weeks, compared to the 12 weeks common in R6/2 mice with 150 repeats. In the > or =335 CAG mice, large ubiquitinated aggregates of mutant protein were common in neuronal dendrites and perikaryal cytoplasm, but intranuclear aggregates were small and infrequent. Message and protein for the > or =335 CAG transgene were reduced to one-third that in 150 CAG R6/2 mice. Neurological and neurochemical abnormalities were delayed in onset and less severe than in 150 CAG R6/2 mice. These findings suggest that polyQ length and pathogenicity in Huntington's disease may not be linearly related, and pathogenicity may be less severe with extreme repeats. Both diminished mutant protein and reduced nuclear entry may contribute to phenotype attenuation.

Increased calbindin-D28k immunoreactivity in striatal projection neurons of R6/2 Huntington's disease transgenic mice.

Striatal degeneration in Huntington's disease (HD) is associated with increases in perikaryal calbindin immunolabeling in yet-surviving striatal projection neurons. Since similar increases have also been observed in surviving striatal projection neurons after intrastriatal injection of the excitotoxin quinolinic acid, the increased calbindin in HD striatum has been interpreted to suggest an excitotoxic process in HD. We used immunolabeling to assess if calbindin is elevated in striatal projection neurons of R6/2 HD transgenic mice. These mice bear exon 1 of the human huntingtin gene with 144 CAG repeats and show some of the neuropathological signs (e.g., neuronal intranuclear inclusions) and clinical traits (e.g., wasting prior to early death) of HD. We found an increased frequency of calbindin-immunoreactive neuronal perikarya in the striatum of 6- and 12-week-old R6/2 mice compared to wild-type controls. This increase was most notable in the normally calbindin-poor dorsolateral striatum. We found no significant changes in the total area of striatum occupied by the calbindin-negative striosomes and no consistent changes in striatal calbindin mRNA. The increase in calbindin in R6/2 striatal neurons was thus limited to the matrix compartment, and it may be triggered by increased Ca2+ entry due to the demonstrated heightened NMDA sensitivity of these neurons. The data further support the similarity of R6/2 mice to HD, and are consistent with the occurrence of an excitotoxic process in striatum in both.

Compartmentation of the reeler cerebellum: segregation and overlap of spinocerebellar and secondary vestibulocerebellar fibers and their target cells.

The cerebellum of the reeler mutant mouse has an abnormal organization; its single lobule is composed of a severely hypogranular cortex and a central cerebellar mass (CCM) consisting of Purkinje cell clusters intermixing with the cerebellar nuclei. As such the reeler represents an excellent model in which to examine the effect of the abnormal distribution of cerebellar cells on afferent-target relationships. To this effect we studied the organization of the spinocerebellar and secondary vestibulocerebellar afferent projections in homozygous reeler mice (rl/rl) using anterograde tracing techniques. Spinal cord injections resulted in labeled spinocerebellar mossy fiber rosettes in specific anterior and posterior regions of the cerebellar cortex. Some vestiges of parasagittal organization may be present in the anterior projection area. Within the CCM, labeled fibers appeared to terminate on distinct groups of Purkinje cells. Thus, the spinocerebellar mossy fibers seem to form both normal and heterologous synapses in the reeler cerebellum. Secondary vestibular injections resulted in both retrograde and anterograde labeling. Retrograde labeling was seen in clusters of Purkinje cells and cerebellar nuclear cells; anterograde labeling was distributed in the white matter and in specific regions of the anterior and posterior cortex of the cerebellum. The labeled spinocerebellar and secondary vestibulocerebellar afferents overlapped in the anterior region but in the posterior region the vestibulocerebellar termination area was ventral to the spinocerebellar area. An area devoid of labeled terminals was also observed ventral to the posterior secondary vestibulocerebellar termination field. Using calretinin immunostaining it was determined that this area contains unipolar brush cells, a cell type found primarily in the vestibulocerebellum of normal mice. Our data indicate that despite of the lack of known landmarks (fissures, lobules) the spinocerebellar and vestibulocerebellar afferent projections in the reeler cerebellum do not distribute randomly but have specific target regions, and the position of these regions, relative to each other, appears to be conserved. Two caveats to this were the finding of overlapping terminal fields of these afferents in the anterior region, and a posteroventral region that contains unipolar brush cells yet is devoid of secondary vestibulocerebellar afferents. The distribution of Purkinje cells and cerebellar nuclear cells is not random either; those that give rise to cerebellovestibular efferents form distinct groups within the central cerebellar mass.

A relationship between cerebellar Purkinje cells and spatial working memory demonstrated in a lurcher/chimera mouse model system.

New emphasis has been placed upon cerebellar research because of recent reports demonstrating involvement of the cerebellum in non-motor cognitive behaviors. Included in the growing list of cognitive functions associated with cerebellar activation is working memory. In this study, we explore the potential role of the cerebellum in spatial working memory using a mouse model of Purkinje cell loss. Specifically, we make aggregation chimeras between heterozygous lurcher (Lc/+) mutant embryos and +/+ (wildtype) embryos and tested them in the delayed matching-to-position (DMTP) task. Lc/+ mice lose 100% of their Purkinje cells postnatally due to a cell-intrinsic gain-of-function mutation. Lc/+<->+/+ chimeras therefore have Purkinje cells ranging from 0 to normal numbers. Through histological examination of chimeric mice and observations of motor ability, we showed that ataxia is dependent upon both the number and distribution of Purkinje cells in the cerebellum. In addition, we found that Lc/+ mice, with a complete loss of Purkinje cells, have a generalized deficit in DMTP performance that is probably associated with their motor impairment. Finally, we found that Lc/+<->+/+ chimeric mice, as a group, did not differ from control mice in this task. Rather, surprisingly, analysis of their total Purkinje cells and performance in the DMTP task revealed a significant negative relationship between these two variables. Together, these findings indicate that the cerebellum plays a minor or indirect role in spatial working memory.

Differential changes in striatal projection neurons in R6/2 transgenic mice for Huntington's disease.

In early adult-onset Huntington's disease (HD), enkephalinergic striatopallidal projection neurons show preferential loss, reduced preproenkephalin (PPE) expression in surviving striatopallidal neurons, and loss of fibers in their projection target area. We examined PPE and PPT (preprotachykinin) gene expression in striatal projection neurons and in striatal projection fibers immunoreactive for the PPE product enkephalin (ENK) and the PPT product substance P (SP) in a transgenic HD model, the R6/2 mouse, to see if changes occur in these neuron types similar to those seen in early adult-onset HD. The results show that PPE mRNA level, the number of striatal neurons expressing PPE, and the staining intensity of fibers immunoreactive for ENK in the pallidum were all decreased. By contrast, the SP-containing striatal projection systems to the pallidum and substantia nigra were relatively normal in R6/2 mice. The selective reduction in striatal PPE in R6/2 mice is reminiscent of adult-onset HD, but the preservation of the striatonigral projection system is not. Thus, R6/2 mice do not strictly mimic adult-onset HD in their striatal pathology.

Neurons lacking huntingtin differentially colonize brain and survive in chimeric mice.

To determine whether neurons lacking huntingtin can participate in development and survive in postnatal brain, we used two approaches in an effort to create mice consisting of wild-type cells and cells without huntingtin. In one approach, chimeras were created by aggregating the 4-8 cell embryos from matings of Hdh (+/-) mice with wild-type 4-8 cell embryos. No chimeric offspring that possessed homozygous Hdh (-/-) cells were obtained thereby, although statistical considerations suggest that such chimeras should have been created. By contrast, Hdh (-/-) ES cells injected into blastocysts yielded offspring that were born and in adulthood were found to have Hdh (-/-) neurons throughout brain. The Hdh (-/-) cells were, however, 5-10 times more common in hypothalamus, midbrain, and hindbrain than in telencephalon and thalamus. Chimeric animals tended to be smaller than wild-type littermates, and chimeric mice rich in Hdh (-/-) cells tended to show motor abnormalities. Nonetheless, no brain malformations or pathologies were evident. The apparent failure of aggregation chimeras possessing Hdh (-/-) cells to survive to birth is likely attributable to the previously demonstrated critical role of huntingtin in extraembryonic membranes. That Hdh (-/-) cells in chimeric mice created by blastocyst injection are under-represented in adult telencephalon and thalamus implies a role for huntingtin in the development of these regions, whereas the neurological dysfunction in brains enriched in Hdh (-/-) cells suggests a role for huntingtin in adult brain. Nonetheless, the lengthy survival of Hdh (-/-) cells in adult chimeric mice indicates that individual neurons in many brain regions do not require huntingtin to participate in normal brain development and to survive.

Abnormal dispersion of a purkinje cell subset in the mouse mutant cerebellar deficient folia (cdf).

Purkinje cells of different molecular phenotypes subdivide the cortex of the cerebellum both rostrocaudally into parasagittal bands and mediolaterally into transverse zones. Superimposed on the Purkinje cell compartmentation, the cerebellar cortex is pleated into a reproducible array of lobes and lobules. During cerebellar development, Purkinje cell bands are formed through the rostrocaudal dispersal of embryonic clusters, triggered primarily by a Reelin-dependent signaling pathway. In the naturally occurring mouse mutant cerebellar deficient folia (cdf), there is a failure of Purkinje cell dispersion that results in widespread Purkinje cell ectopia in the adult. The ectopia is restricted primarily to that subset of Purkinje cells that does not express zebrin II/aldolase C and that forms ectopic clusters in among the cerebellar nuclei. Most Purkinje cells that express zebrin II are located normally in a monolayer. Thus, the cerebellum of cdf mutants has a failure of Purkinje cell dispersion that is confined primarily to a zebrin II-negative (zebrin II(-)) subpopulation. Despite the Purkinje cell ectopia, the parasagittal band organization of the cerebellum is still clear. The shortening of the cortex is distributed evenly over all lobules, with the result that transverse expression boundaries are relocated with respect to the lobules and fissures. The number of Purkinje cells in the cdf/cdf cerebellum is similar to the number in littermate controls. Therefore, it appears that the lesion in cdf results in the failure of a zebrin II(-) Purkinje cell subset to disperse either due to a cell intrinsic defect or due to an abnormal interaction between the Purkinje cells and either granule cells or afferent inputs.

Abnormalities in the functioning of adipocytes from R6/2 mice that are transgenic for the Huntington's disease mutation.

In an effort to characterize the basis of abnormalities in body weight regulation (i.e. wasting) in Huntington's disease (HD), we examined adipocytes in a transgenic model of HD, the R6/2 mouse. These mice typically show severe wasting beginning at approximately 12 weeks of age and die between 12 and 15 weeks. Despite an overall growth retardation compared with wild-type littermates, we observed an enhanced accumulation of body fat at 8-9 weeks of age in R6/2 mice fed laboratory chow or a synthetic high fat, high sugar diet. The obesity was not accompanied by symptoms associated with diabetes, as there were no abnormalities in serum glucose, serum insulin or the ability of insulin to stimulate glucose metabolism in epididymal adipose tissue. As expected, the obesity in the high fat, high sugar-fed R6/2 mice was accompanied by increased serum leptin. The ability of insulin to stimulate leptin release from isolated epididymal adipose tissue was also enhanced in R6/2 mice. In contrast, the ability of isoproterenol to inhibit leptin release was reduced in adipose tissue from R6/2 mice, as was the lipolytic effect of isoproterenol. These data suggest that the obesity observed at 8-9 weeks in R6/2 mice may stem from a defect in fat breakdown by adipocytes.

Granule cells and cerebellar boundaries: analysis of Unc5h3 mutant chimeras.

Mutations in the Unc5h3 gene, a receptor for the netrin 1 ligand, result in abnormal migrations of both Purkinje and granule cells to regions outside the cerebellum and of granule cells to regions within the cerebellum. Because both Purkinje and granule cells express this molecule, we sought to determine whether one or both of these cell types are the primary target of the mutation. Chimeric mice were made between wild-type ROSA26 transgenic mouse embryos (whose cells express beta-galactosidase) and Unc5h3 mutant embryos. The resulting chimeric brains exhibited a range of phenotypes. Chimeras that had a limited expression of the extracerebellar phenotype (movement of cerebellar cells into the colliculus and midbrain tegmentum) and the intracerebellar phenotype (migration of granule cells into white matter) had a normal-appearing cerebellum, whereas chimeras that had more ectopic cells had attenuated anterior cerebellar lobules. Furthermore, the colonization of colliculus and midbrain tegmentum by cerebellar cells was not equivalent in all chimeras, suggesting different origins for extracerebellar ectopias in these regions. The granule cells of the extracerebellar ectopias were almost entirely derived from Unc5h3/Unc5h3 mutant embryos, whereas the ectopic Purkinje cells were a mixture of both mutant and wild-type cells. Intracerebellar ectopias in the chimera were composed exclusively of mutant granule cells. These findings demonstrate that both inside and outside the cerebellum, the granule cell is the key cell type to demarcate the boundaries of the cerebellum.

Extrinsic modulation of retinal ganglion cell projections: analysis of the albino mutation in pigmentation mosaic mice.

Tyrosinase is a key enzyme involved in the synthesis of melanin in the retinal pigment epithelium (RPE). Mice that are homozygous for the albino allele at the tyrosinase locus have fewer retinal ganglion cells with uncrossed projections at the optic chiasm. To determine the site of the albino gene action we studied the projections of retinal ganglion cells in two types of pigmentation mosaic mice. First, we generated mosaic mice that contain a translocated allele of the wild-type tyrosinase on one X chromosome but that also have the lacZ reporter transgene on the opposite X chromosome. In these lacZ/tyrosinase mice, which are homozygous for the albino allele on chromosome 7, X-inactivation ensures that tyrosinase cannot be functional within 50% of the retinal ganglion cells and that these individual cells can be identified by their expression of the lacZ reporter gene product, beta-galactosidase. The proportion of uncrossed retinal ganglion cells expressing beta-galactosidase was found to be identical to the proportion that did not express it, indicating that the albino mutation associated with axonal behavior at the optic chiasm must affect ganglion cells in a cell-extrinsic manner. Second, to determine whether the RPE is the source of the extrinsic signal, we generated aggregation chimeras between pigmented and albino mice. In these mosaic mice, the extent of the uncrossed projection corresponded with the amount of pigmented cells within the RPE, but did not correspond with the genotypes of neural retinal cells. These studies demonstrate that the albino mutation acts indirectly upon retinal ganglion cells, which in turn respond by making axonal guidance errors at the optic chiasm.

Rescue of cerebellar granule cells from death in weaver NR1 double mutants.

The weaver mutation results in the extensive death of midline cerebellar granule cells. The mutation consists of a single base pair substitution of the gene encoding the G-protein-activated inwardly rectifying potassium channel protein, GIRK2. The functional consequences of this mutation are still in dispute. In this study we demonstrate the in vivo and in vitro rescue of weaver granule cells when NR1 NMDA subunits are eliminated in weaver NR1 double mutants. This rescue of weaver granule cells provides evidence that wvGIRK2 alone is not sufficient to cause granule cell death.

Disrupted retinal development in the embryonic belly spot and tail mutant mouse.

The Belly spot and tail (Bst) semidominant mutation, mapped to mouse Chromosome 16, leads to developmental defects of the eye, skeleton, and coat pigmentation. In the eye, the mutant phenotype is characterized by the presence of retinal colobomas, a paucity of retinal ganglion cells, and axon misrouting. The severity of defects in the Bst/+ retina is variable among individuals and is often asymmetric. In order to determine the role of the Bst locus during retinal morphogenesis, we searched for the earliest observable defects in the developing eye. We examined the retinas of Bst/+ and +/+ littermates from embryonic day 9.5 (E9.5) through E13.5 and measured retinal size, cell density, cell death, mitotic index, and cell birth index. We have found that development of the Bst/+ retina is notably dilatory by as early as E10.5. The affected retinas are smaller than their wildtype counterparts, and optic fissure fusion is delayed. In the mutant, there is a marked lag in the exit of retinal cells from the mitotic cycle, even though there are no observable differences in the rate of cellular proliferation or cell death between the two groups. We hypothesize that Bst regulates retinal cell differentiation and that variability of structural defects in the mutant, such as those affecting optic fissure fusion, is a reflection of the extent of developmental delay brought about by the Bst mutation.