| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online December 13, 2005 as doi:10.1096/fj.05-3785fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
* Laboratory of Neurophysiology,
IRIBHM, IBMM,
Laboratory of Movement Biomechanics Université Libre de Bruxelles, Brussels, Belgium; and
Laboratory of Electrophysiology, Université Mons-Hainaut, Mons, Belgium
1 Correspondence: Laboratory of Neurophysiology CP601, Université Libre de Bruxelles, route de Lennik 808, Brussels 1070, Belgium. E-mail: sschiffm{at}ulb.ac.be
SPECIFIC AIMS
Inactivation of gene coding for calretinin, a calcium binding protein acting as a fast calcium buffer, deeply altered the cerebellar physiology leading to motor discoordination. Although mechanistic hypotheses have been proposed to explain these alterations, the widespread expression of calretinin in the whole central nervous system and its expression in several neuronal types in the cerebellar circuitry preclude a precise mechanistic demonstration. The aim of this study was to determine to what extent the expression of calretinin in a specific neuronal cell type, the cerebellar granule cells, contributes to the phenotype detected in calretinin knockout (Cr–/–) mice. To achieve this objective, we constructed transgenic Cr–/– mice exhibiting a selective re-expression of Cr in granule cells (hereafter referred to as Cr-rescue) through the promoter function of the GABAA receptor 
6 subunit gene.
PRINCIPAL FINDINGS
1. Cr-rescue mouse lines generation and characterization of granule cells specific expression of calretinin
To generate Cr-rescue mice, we used as promoter a part of the GABAA receptor 6 subunit gene, which is capable of directing expression of a transgene in cerebellar granule cells. We constructed a calretinin transgene that consisted of the truncated GABAA-6 gene, an IRES, the calretinin encoding cDNA, and the SV40polyA (pm6-IRES-calretinin-polyA) (Fig. 1
A). Founders obtained were crossed with C57Bl6 Cr–/– mice and two times with either C57Bl6 Cr–/– mice or C57Bl6 WT mice to obtain F3 Cr–/– Tg+, F3 Cr–/– Tg–, and F3 Cr+/+ Tg– mice. Five different lines were analyzed using in situ hybridization, immunohistochemistry and immunoblot.
|
As described previously, in situ hybridization revealed a widespread brain expression of calretinin mRNA in WT mice whereas no expression was detected in Cr–/– mice (Fig. 1Ba, b
). In the brain of our 5 transgenic lines, a similar pattern of calretinin mRNA expression was observed (Fig. 1Bc, g
). In all 5 lines, this calretinin mRNA expression is restricted to the cerebellum, excepted in one lineage, where an additional slight expression in the pontine nuclei and occipital/retrosplenial cortex was found (Fig. 1Bc
). In these 5 lines, expression was confined to the granular layer (Fig. 1Bh, j
), confirming the calretinin mRNA re-expression in granule cells.
Calretinin immunoreactivity was distributed throughout WT brain, including cerebellar granule cells and parallel fibers in the molecular layer. As described, no labeling was observed in the brain of Cr–/– mice except a paradoxical calretinin immunoreactivity, which corresponds to calbindin-D28k, in Purkinje cells. In Cr-rescue mice, calretinin immunoreactivity was only detected in granule cells of the cerebellar cortex, including in the line showing a low level of calretinin mRNA expression in pontine nuclei. It is noteworthy that in all 5 lineages of Cr-rescue mice, the paradoxical calretinin-like immunoreactivity detected in Purkinje cells persists as in Cr–/– mice. In cerebellar cortex of WT mice, calretinin is not only expressed in granule cells but also in unipolar brush (UBCs) and Lugaro cells. To determine whether calretinin would have also been re-expressed in UBCs of our Cr-rescue mice, we performed a double immuno-detection of calretinin and glutamate receptor subunits GluR2/3, which are expressed in UBCs and Purkinje cells. UBCs were GluR2/3-positive and calretinin-negative in the different lineages of Cr-rescue mice. Calretinin is the only specific marker for the detection of Lugaro cell bodies in mouse cerebellum. However, we never observed calretinin immunoreactive cells presenting characteristics of Lugaro cells in the cerebellar cortex of Cr-rescue mice.
As described, immunoblot analysis using the anti-calretinin antibody revealed a band at 29 kDa corresponding to the calretinin molecular weight in cerebellum homogenates from WT mice whereas a band at 27 kDa that corresponds to calbindin was detected in Cr–/– mice. In the 5 lineages of Cr-rescue mice, we observed the 29 kDa band corresponding to calretinin and the 27 kDa band corresponding to calbindin. On the basis of these results, we decided to further study the 3 lines of Cr-rescue mice that expressed the higher level of calretinin in the cerebellum.
2.Restoration of a WT Purkinje cell firing and granule cell excitability
Since Cr–/– mice exhibited an alteration in Purkinje cell firing in vivo, the firing behavior of 552 identified Purkinje cells was analyzed in 10- to 12-month-old WT (n=78 cells in 4 animals), Cr–/– (n=150 cells in 9 animals) and Cr-rescue (n=324 cells in 15 animals) alert mice by single unit recording. As reported, the mean spontaneous simple spike firing rate was significantly enhanced in Cr–/– mice compared with WT mice (Fig. 2
A) whereas the complex spike firing rate was not statistically different (Fig. 2B
). In the 3 lines of Cr-rescue mice, the simple spike firing rate was lower than in Cr–/– mice and was restored to values not different from WT (Fig. 2A
). Durations of the complex spike and of the transient pause in simple spike firing after a complex spike were reduced in Cr–/– mice compared with the WT (Fig. 2C, D
). As for the simple spike firing rate, both durations were completely restored to WT values in the 3 different lines of Cr-rescue mice (Fig. 2C, D
).
|
A spontaneous high-frequency LFPO has been described in the cerebellum of Cr–/– mice and is absent in the cerebellum of WT mice. We looked for this oscillation in 6 Cr-rescue, 3 Cr–/– and 5 WT mice. All Cr–/– mice presented episodes of spontaneous LFPO (Fig. 2E
) with frequency of 162 ± 48 Hz and an oscillation index of 6.4 ± 4.7. Conversely, neither WT nor Cr-rescue mice presented episodes of LFPO (Fig. 2F
).
Intrinsic granule cell excitability was investigated on granule cells in brain slices using patch clamp, given that it was altered in Cr–/– mice. Fast regular repetitive spiking was obtained by injecting steps of depolarizing currents and action potentials frequency increased with the current intensity. The average frequency was measured and used to construct current-frequency plots, and the slope factor of these plots was used as a normalized measure of excitability. We observed that the mean slope factor for one line of rescued transgenic Cr–/– mice (4.8±0.8 Hz/pA, n=4) was equivalent to the one reported for WT mice (4.8±0.2 Hz/pA) and lower than the one recorded from Cr–/– mice (6.6±0.7 Hz/pA).
3. Cr-rescue mice exhibit a normal motor coordination
Previous work has revealed impaired motor coordination in Cr–/– mice mostly in aged mice. To test whether the rescue of calretinin expression in granule cells and the restoration of a normal Purkinje cell firing pattern result in a normal motor behavior, we examined motor coordination of Cr-rescue mice generated on a hybrid background by using the runway test. To discriminate Cr–/– from WT mice at the younger age of 3 months, we decreased the width of the runway (18 mm in the original test to 12 mm). In this condition, Cr-rescue and WT mice made fewer errors than Cr–/– mice (P <0.05), while no differences appeared between WT and Cr-rescue mice (P >0.05) (2 way repeated ANOVA).
CONCLUSIONS AND SIGNIFICANCE
In this report, we produced lines of transgenic mice rescuing specifically the expression of calretinin in granule cells of the cerebellum of Cr–/– mice. Using this model, we have shown that this selective rescue restored: 1) a WT Purkinje cells firing behavior leading to the absence of oscillation; 2) a normal neuronal excitability of granule cells in slices; and 3) a normal motor coordination compared with the alterations detected in Cr–/– mice.
The selective rescue of calretinin expression in granule cells of Cr–/– mice proves that its control of the Purkinje cells firing only occurs through a subtle regulation of the granule cell intrinsic excitability. Consequently, by showing that the level of calretinin expression specifically in granule cells regulates their excitability, the Purkinje cell firing and ultimately motor coordination, these results demonstrate that the fine tuning of granule cell excitability plays a crucial role in the control of information coding and storage in the cerebellum. Moreover, they show that by its fine control of calcium homeostasis, the calcium binding protein calretinin is an essential actor in granule cell to provide a correct computation in the cerebellar cortex.
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-3785fje;
This article has been cited by other articles:
|
|
 |
|
  L. Servais, R. Hourez, B. Bearzatto, D. Gall, S. N. Schiffmann, and G. Cheron Purkinje cell dysfunction and alteration of long-term synaptic plasticity in fetal alcohol syndrome PNAS, June 5, 2007; 104(23): 9858 - 9863. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
