Richard J. Mullen

Regional variation and absence of large neurons in the cerebellum of the staggerer mouse

Staggerer (sg) is a neurological mutant mouse in which the cerebellar granule cells degenerate after migration to the internal granule cell layer. In addition, the Purkinje cells are abnormal, being ectopic, smaller in size and without tertiary branchlet spines. In this study we report two new observations on this mutant: (1) the effects of the mutation show a regional variation in severity along the mediolateral axis. This variation is seen in the cross-sectional size of the tissue, the extent of cortical folding, as well as the density and cytological appearance of the medium-to-large cortical neurons (MLNs); (2) cell counts were done of 30-day-old mutants and of wild-type controls. The counts revealed that three-quarters of the MLNs of the cerebellar cortex are missing in staggerer. These findings cannot exclude the possibility that Golgi as well as Purkinje cells are absent since the two cannot be distinguished in staggerer. Depending on the number of Golgi cells present, between 60% and 90% of the wild-type number of Purkinje cells is missing in staggerer. The results are discussed in terms of their implications for other studies and for locating the site of staggerer gene action.

Staggerer chimeras: intrinsic nature of purkinje cell defects and implications for normal cerebellar development.

The site of gene action of the Staggerer mutation of mice was investigated with Staggerer in equilibrium or formed from wild-type chimeras. Homozygous Staggerer mice show severe locomotor difficulties due to cerebellar abnormalities which include degeneration of virtually all granule cells and cytological defects in Purkinje cells. Although the locomotor deficits of the mutant were not present in the chimeras, the presence of Staggerer cells affected cerebellar structure. The size and the extent of foliation of the chimeric cerebella were intermediate between wile-type and homozygous Staggerer. A normally proportioned granule cell layer was present. Using beta-glucuronidase as an independent determinant of a cells genotype, it was found that the genotypically Staggerer medium-to-large neurons expressed all of the light microscopic defects observable in these cells in the homozygous mutant. These defects include: (1) smaller size; (2) usually ectopic location; and (3) regional variation in the cytological appearance of the perikaryon. By contrast, all Purkinje cells which were genotypically wild-type appeared normal in size, in location and in their cytological appearance. Their density, however, was much reduced from wild-type. The effects of the Staggerer mutation on the granule, stellate and basket cells could not be directly assessed as the glucuronidase marker is not suitable for use with these cells. The Staggerer gene thus acts directly on Purkinje cells rather than via extracellular environmental changes. The findings are discussed in terms of their implications for normal cerebellar development.

Role of the pigment epithelium in inherited retinal degeneration analyzed with experimental mouse chimeras

Abstract In the study of inherited photoreceptor cell degenerations, a difficult problem is to determine the site of the primary genetic defect. It could be either intrinsic to the photoreceptor cells, the pigment epithelial cells or both, or extrinsic to these cells or to the eye as a whole. This problem was studied in mice with inherited retinal degeneration ( rd rd ) by analyzing the interaction of mutant and normal pigment epithelium with the underlying normal or degenerated retina in the eyes of experimental chimeric mice. Chimeras were produced by fusing eight-cell embryos of albino SJL ( rd rd ) mice with those of pigmented C57BL 10 (+/+) mice. The eyes of the resulting chimeric mice at 1·5–26·5 months of age had patches of normal retina interspersed with patches lacking photoreceptor cells. Pigment epithelial cells of both normal and mutant strains, identified by the presence or absence of melanosomes, were found overlying areas of both normal and degenerated retina. The same findings were obtained using two other strain combinations, C3H (pigmented, rd rd ) fused with BALB c (albino, +/+) and SJL (albino, rd rd ) fused with C57BL 10 × CBA F 1 hybrids (pigmented, +/rd). The synthesis of rod outer segment dises proceeded normally in photoreceptor cells underlying mutant pigment epithelial cells, as determined by autoradiographic analysis, and phagosomes were found in both mutant and normal pigment epithelial cells. The findings indicate that the pigment epithelial cell is not the primary target of the mutant rd gene in the mouse and localize the site of mutant gene action to the neural retina, presumably but not necessarily, to the photoreceptor cells.

Role of the staggerer gene in determining Purkinje cell number in the cerebellar cortex of mouse chimeras

Staggerer (sg) is a mutation in mice which causes severe cerebellar atrophy. Virtually all of the granule cells degenerate by the end of the first postnatal month. The Purkinje cells are also affected. They are reduced in number, ectopic in location, reduced in size and show a regional variation in their cytological appearance. The appearance of the cerebellar cortex of Staggerer in equilibrium wild-type chimeric mice has demonstrated that the reduced size, ectopia, and regional variation are always traits of cells whose nucleus is of sg/sg genotype and thus are due to factors intrinsic to the affected cell. This report is a quantitative analysis of the cerebellar cortex of two Staggerer chimeras. Sections were chosen from serially sectioned half cerebella. Using beta-glucuronidase activity as an independent marker of a cells genotype, cells were scored as belonging to 1 of 3 cell classes: wild-type Purkinje cell, wild-type Golgi cell, and Staggerer medium-to-large neuron. The results confirm that the total number of cells in the chimera (all 3 classes) is greater than that found in a homozygous Staggerer but less than that found in a wild-type. Further, the distribution of the two genotypes strongly suggests that the decreased number of cells in Staggerer and in the chimera is due to an intrinsic defect in the Staggerer cells and not to a general (extrinsic) compromising of cerebellar development. Possible implications for normal cerebellar development are discussed.

Neuronal influence on glial enzyme expression: Evidence from chimeric mouse cerebellum

Cerebella, variably deficient in Purkinje cells, were obtained from aggregation chimeras of either Lurcher or Purkinje cell degeneration mutants. These cerebella were used to analyze the expression of glycerol-3-phosphate dehydrogenase (GPDH) in Bergmann glia. Immunocytochemistry showed apparently normal GPDH expression only in Bergmann glia in the immediate vicinity of surviving Purkinje cells. The number of GPDH-positive Bergmann glia cells associated with isolated Purkinje cells was close to that expected, based on measurements in Golgi-stained, normal cerebella of the Bergmann glia cells domain. The results support the hypothesis that GPDH expression in Bergmann glia cells depends upon their sustained interaction with Purkinje cells, most likely involving direct cell-cell contact.

Nuclear morphology of ichthyosis mutant mice as a cell marker in chimeric brain

Abstract The nuclear morphology of certain neuronal populations from the mutant mouse, ichthyosis, is distinct from wild-type strains of mice. The granule cells of the cerebellum, cochlear nucleus, and olfactory bulb in ichthyosis mice have a much greater tendency for centralized clumping of nuclear heterochromatin. In the early postnatal nervous system many cells in migratory and germinal regions of the brain also express the ichthyosis phenotype. The retention of the ichthyosis phenotype in neurons of chimeric mice is documented. The prevalent expression of the ichthyosis phenotype in postnatal migratory and germinal regions of the brain would be particularly useful for studying cell interactions in the developing brain.

Annexin IV is a marker of roof and floor plate development in the murine CNS

Midline structures, such as the notochord and floor plate, are crucial to the developing central nervous system (CNS). Previously, we demonstrated that annexin IV is an excellent marker of midline structures. In the present study, we explore the possible role of annexin IV in development of the CNS midline. Using immunocytochemistry with an antibody to annexin IV, we have elucidated the temporal and spatial expression of this molecule. Annexin IV is present in the notochord at embryonic day (E) 8.5, prior to its expression in any structures within the neural tube. Subsequently, annexin IV is expressed by floor plate cells at E9.5. Annexin IV is also expressed in the roof plate, but not until E10.5. To determine if normal morphogenesis of these midline structures is essential for annexin IV expression, we analyzed two strains of mutant mice that have defective formation of either the floor or the roof plate. In Danforths short‐tail mice, the floor plate is absent from the caudal spinal cord, and annexin IV immunopositivity disappears at the level where the floor plate is missing. In curly tail mutant mice, there can be a failure of the neural tube to close, and in these regions there is no annexin IV expression in presumptive roof plate cells. Finally, annexin IV immunolabeling is present from the caudal spinal cord, through the brainstem up to the diencephalon and lamina terminalis. Thus, annexin IV is an excellent marker for differentiated midline cells, is temporally and spatially correlated with development of the floor and roof plates, and is expressed in a rostral‐caudal manner that supports the hypothesis that the floor plate extends the full length of the original neural tube.

Biochemical and genetic factors in the heat inactivation of murine beta-glucuronidase.

The enzymes coded for by two alleles at the glucuronidase structural locus (Gus) were compared in their response to pH, buffering anion, buffer molarity, ionic strength, and temperature. The heat-labile Gush gene product responded in a qualitatively similar but quantitatively reduced manner compared to the relatively heat-stable Gusb gene product. In all buffers tested, the enzyme was most heat stable at pH 5.0. Ranking of the various buffer anions tested, according to increasing heat stabilization, was water ≪ acetate ≤ phosphate < citrate. Varying the molarity of the buffers from 0.01 to 0.6 m at pH 5.0 revealed further differences among the buffers. Increasing ionic strength exerted a destabilizing force on the protein. The half-life of the enzyme decreased by as much as a hundredfold between 71 and 75 C. The Gush/Gush genotype also results in decreased activity levels in all tissues, reportedly because of decreased synthesis. The heat inactivation curves of Gusb/Gush heterozygotes were incompatible with any theoretical curve based on the assumption that the Gusb and Gush chromosomes in the heterozygote behave in a manner similar to that seen in the homozygotes.

Patterns of neuronal differentiation in neural tube mutant mice: curly tail and Pax3 splotch-delayed.

A battery of antibodies was used to assess development of the spinal cord and its neurons in mouse embryos with neural tube defects (NTDs). The two mutant strains examined, curly tail (ct) and splotch‐delayed (Pax3Sp‐d), develop an open neural tube for unrelated reasons, and thus provided for a complementary analysis. Five percent of embryos homozygous for the ct gene and 89% of embryos homozygous for the Pax3Sp‐d gene develop spina bifida in the lumbosacral region of the neuraxis. Expression of several neuronal antigens, including Islet‐1/2, polysialylated neural cell adhesion molecule (NCAM), neurofilaments, and a neuronal‐specific nuclear protein (NeuN) recognized by monoclonal antibody A60, were used as indicators of the level of differentiation of neuronal tissue. Immunohistochemical labeling suggests that early (embryonic days 12–15) neuronal differentiation in the dorsal and ventral region of the dysraphic neural tube occurs remarkably normally in both of the mutants. Similarly, labeling with antibodies to NCAM and neuroafilaments indicate that axonal development during early neurogenesis is unperturbed. Later stages of neuronal maturation, however, do not occur in the usual manner. Instead, the neuronal tissue begins a prodigious degeneration at embryonic day 17 (E17), so that by E18 only a rudimentary tissue remains. These results suggest that the aberrant morphology of the neural tube does not affect neuronal differentiation. However, the anomalous morphological and chemical environment may contribute to the neuronal degeneration observed at later stages.

Cell mixing in the spinal cords of mouse chimeras

With the aim of determining whether there is significant cell mixing during development of the spinal cord, experimental chimeric mice containing two genetically distinct cell populations were produced by aggregating BALB/c or BALB/c x LPT hybrid embryos with C3H/HeN embryos. The BALB/c and LPT hybrid spinal cord cells were distinguished histochemically from the C3H/HeN spinal cord cells by using beta-glucuronidase as an independent cell genotype marker. BALB/c and LPT hybrid cells have high levels of beta-glucuronidase activity, while the C3H/HeN cells have low levels. The spinal cords of the chimeras were dissected out regionally (i.e., cervical, thoracic, and lumbar areas) and were sectioned serially. Each region was then analyzed by scoring large- and medium-sized neuronal cell bodies (greater than or equal to 10 microns) whose genotypes were distinguished by their beta-glucuronidase levels. Observations of seven chimeric mice, with coat colors that varied from one extreme (5% albino) to the other (90% albino), suggest that each chimeric spinal cord is a relatively homogeneous population throughout its length. On average only 4 to 5 percentage point differences were observed when comparing left-right, cranial-caudal, and dorsal-ventral regions within a given chimera. The cell mixing, however, is not total, and regional variations were noted. Maximum left-right differences between different spinal cord levels never exceeded 18 percentage points, while in the entire cord the maximum left-right difference was 11 percentage points. When considering dorsal-ventral differences, 18 and 15 percentage points were observed within the spinal cord levels and the entire cord, respectively. However, when comparisons were made between smaller subregions (e.g., right-dorsal-cervical vs left-ventral-lumbar), larger differences of up to 30 percentage points were observed. In addition, the genotype proportions in the spinal cord were closely correlated with the visually estimated proportions of coat color genotypes.