| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Animal Model |
From the Basic Research Program,* the Laboratory Animal Sciences Program, Science Applications International Corporation-Frederick, Inc., National Cancer Institute at Frederick, Frederick, Maryland; the Laboratory of Immunobiology, Veterinary and Tumor Pathology Section, the Mouse Cancer Genetics Program, Neural Development Section, Computer and Statistical Services, Center for Cancer Research, National Cancer Institute at Frederick, Frederick, Maryland; the Clinical Center,¶ Clinical Electrophysiology and Inherited Heart Diseases Section,|| Cardiology Branch, National Heart, Lung, and Blood Institute, Epilepsy Research Section,** Division of Intramural Research, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland; and the Hamon Center for Therapeutic Oncology Research and the Department of Pathology,¶¶ University of Texas, Southwestern Medical Center, Dallas, Texas
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
2
-2 protein encoded by the gene is a regulatory subunit of voltage-dependent calcium channels and is expressed in brain, heart, and other tissues. Here we report that mice homozygous for targeted disruption of the Cacna2d2 gene exhibit growth retardation, reduced life span, ataxic gait with apoptosis of cerebellar granule cells followed by Purkinje cell depletion, enhanced susceptibility to seizures, and cardiac abnormalities. The Cacna2d2tm1NCIF null phenotype has much in common with that of Cacna1a mutants, such as cerebellar neuro-degeneration associated with ataxia, seizures, and premature death. A tendency to bradycardia and limited response of null mutants to isoflurane implicate 2-2 in sympathetic regulation of cardiac function. In summary, our findings provide genetic evidence that the 2-2 subunit serves in vivo as a component of P/Q-type calcium channels, is indispensable for the central nervous system function, and may be involved in hereditary cerebellar ataxias and epileptic disorders in humans.
1 protein that forms the channel pore and voltage-sensor. Core 1 subunits are used for dividing VDCCs into three major phylogenetic types, which are supported by pharmacological and functional classification: T-type (1G, 1H, and 1I); L-type (1C, 1D, 1F, and 1S); and the presynaptic channels-type (1A, 1B, and 1E).4
A variety of auxiliary regulatory subunits, including 2, ß, and 
, associate with the 1-subunits, forming different heteromeric functional VDCCs.5-7
The members of the 2 family include 2-1, which is ubiquitously expressed; 2-2, which is expressed in brain, heart, and other organs; and 2-3, whose transcripts are found mostly in the brain.8-10
2 subunits consist of two proteins that are encoded by a single gene, the product of which is post-translationally cleaved from a single polypeptide precursor.10,11
The discovery of worm and fly orthologs of 2-encoding genes12
demonstrates their conservation over billions of years of evolution, suggesting that the 2 protein is of fundamental importance in cellular function.
In humans, 2-2 is encoded by CACNA2D2 (GenBank Accession No. AF040709), which is contained, along with seven other genes, in a 120-kb region of chromosome 3p21.3 that is frequently deleted in lung, breast, and other cancers.12
Reduction or absence of CACNA2D2 expression in non-small cell lung cancer cell lines suggests that 2-2 itself may be a tumor suppressor protein. 2-2 is predicted to be a heavily glycosylated 175-kd protein whose single-pass transmembrane piece anchors the 2 protein to the membrane.13
Electrophysiological studies demonstrated that 2-2 might function as an auxiliary component of a variety of different VDCCs: L-type [Cav1.2 (1C)]; P/Q-type [Cav2.1 (1A)]; N-type [Cav2.2 (1B)]; R-type [Cav2.3 (1E)]; and T-type [Cav3.1 (1G) and Cav3.2 (1H)].8,9,14
Co-expression of 2-2 with each of these 1 subunits enhanced currents through the heteromeric channels. While these experiments characterized 2-2 as electrophysiologically versatile, its biological significance remained uncertain. Here we report that the Cacna2d2 null mice show ataxia, growth retardation, increased susceptibility to seizures, and a reduced life span. Phenotypic similarity with Cacna1a null mice and spontaneous Cacna1a mutants implicates 2-2 in regulation of P/Q-type currents as a component of Cav2.1 channels and suggests a possible involvement of the gene in human hereditary cerebellar ataxias and epilepsies. We also provide evidence that 2-2 expression in the central nervous system, dorsal sympathetic ganglia, and cardiac conductive tissue may be related to the observed disturbances in null mutants’ heart rate regulation.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
A replacement-type Cacna2d2-targeting vector was based on a mouse genomic clone isolated from a 129/SvJ lambda FIXII library (Stratagene, La Jolla, California). Cacna2d2 exons were identified by Southern hybridization, partial sequencing, and BLAST analysis. Mouse EST clone represented in GenBank by Accession No. AA000341 (Research Genetics, Huntsville, AL) was used as a probe. A 2629-bp fragment containing eight exonic fragments (positions 2517 to 3184 in the mouse Cacna2d2 mRNA AF247139) and a 6.3-kb fragment representing the 3' non-coding region of Cacna2d2 were integrated into the construct as 5' and 3' flanking sequences, respectively. The neo-gene with the phosphor-glycerol kinase 1 promoter and the bovine growth hormone polyadenylation sequence (pGKneobpA) was used as a positive selectable marker; the pGK-thymidine kinase cassette was used as a negative selectable marker.15
Electroporation and selection were performed using CJ7 ES cell line, as described elsewhere.16
DNAs derived from G418/FIAU-resistant ES clones were screened using a diagnostic EcoRI restriction enzyme sites and the 2.0-kb 3' probe external to the targeting sequence indicated in Figure 1A
. This probe detects a 
20-kb EcoRI fragment in wild-type DNA and an 8.3-kb EcoRI-fragment in a mutant allele. Two independent Cacna2d2-targeted cell lines were isolated and one of those, when injected into C57BL/6J blastocysts, transmitted the mutant allele to the offspring generating chimeras. Chimeras were then identified by the agouti coat color. Male chimeras were mated with C57BL/6J females and agouti-colored offspring were genotyped by tail clip. Chimeras that transmitted the Cacna2d2tm1NCIF allele through the germline were then mated to C57BL/6J females to establish lines.
|
RT-PCR on mouse RNA samples was performed using a SuperScript One-Step RT-PCR kit (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. Two primer sets were used for RT-PCR. One of them, 5'-GCATAACTATGATGTCACAC-3' and 5'-GGGGCTTGAAGATATAACCA-3', represented a Cacna2d2 cDNA fragment with positions 1742–2862. The other pair, 5'-TCTACTCTGTGTCCTCATTG-3' and 5'-TGAGTCTAGGGACTGTGGG-3', corresponded to positions 3113–4149 (GenBank Accession No. AF247139). Comparative PCR on human cardiovascular MTN cDNA panel (Clontech, Palo Alto, CA) was performed as suggested by the manufacturer with human primers 5'-GCATAACTATGACGTCACAC-3' and 5'-GGGGCTTGAAGACATAACCG-3' corresponding to the first set of mouse primers. Northern analysis and RT-PCR were performed on mouse brain mRNAs isolated with a FastTrack 2.0 kit (Invitrogen). A Cacna2d2 cDNA EST clone (GenBank Accession No. AA000341, positions 3237–5498 in AF042792) purchased from Genome Systems, Inc. (St. Louis, MO) was used as a 3' probe, while the RT-PCR product obtained with the first set of primers (see above) was used as a 5' probe. The same mouse EST clone was also used to generate 35S-labeled RNA probes for in situ hybridization on mouse tissues. These probes were synthesized using MAXIscript (Ambion Inc., Austin, TX) in vitro transcription kit and two primers. Primer 5'TAATACGACTCACTATAGGGAGAcctatgactatcaggcag-3' contained T7 RNA-polymerase promoter shown in uppercase and was used to produce sense probe. Primer 5'AATTAACCCTCACTAAAGGGAGAccacagtctgaggtatct-3' contained T3 RNA-polymerase promoter and was used to generate anti-sense probe. DNA template used in the RNA synthesis was eliminated with DNase I following the in vitro transcription reaction. For in situ hybridization, tissues of postnatal mice or embryos at stages 13.5 and 15 dpc were perfused and fixed as described.17 Briefly, the samples were hydrated, paraffin-embedded, and sectioned at 4 mm onto microscope slides. For pre-hybridization, the slides were deparaffinized in xylene, hydrated through a series of graded ethanol/DEPC-saline, put through a microwave RNA retrieval procedure, and the RNA was further unmasked by permeabilization with pronase. For hybridization, a riboprobe 750,000 cpm was added to the hybridization buffer,17 denatured at 95°C for 5 minutes, and incubated with the slides for 14 hours at 55°C. Slides were washed in the washing buffer, incubated with RNase A, dehydrated in graded ethanol rinses, and dried under vacuum. Dried slides were immersed into twofold-diluted K.5 nuclear emulsion (Polysciences, Warrington, PA), slowly dried at room temperature, and exposed for 14 days. The slides were then developed using D19 reagent (Eastman Kodak, Rochester, NY), and latent images fixed using Kodak Fixer. The slides were finally thoroughly rinsed, counterstained with hematoxylin (Richard-Allen, Kalamazoo, MI), dehydrated, and glass-covered using permanent mounting media. Visualization of signal was done on a Leitz Laborlux-S (Wetzlar, Germany) microscope.
Animal Testing
Animals used in these studies were maintained in facilities fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care and all procedures were performed under protocols approved by the NIH (Publication No. 86–23, 1985) and NCI Animal Care and Use Committees in strict compliance with the Guide for the Care and Use of Laboratory Animals of the National Research Council (National Academy Press, Washington, DC; http://www.nap.edu/readingroom/books/labrats).
Seizure Susceptibility Testing
Twelve- to 24-week-old mice were used. Pentylenetetrazol (PTZ) (Sigma-Aldrich, St. Louis, MO), in 0.9% NaCl solution, was administered by subcutaneous injection into a loose fold of skin on the right belly. Mice were observed for 30 minutes. Animals exhibiting a clonic spasm of at least 5 seconds duration were scored as positive for seizure occurrence. To determine the 50% convulsive dose (CD50) a range of PTZ doses were used spanning the CD50 value. Each mouse received only a single injection and the animal was immediately euthanized after exhibiting convulsions or at the end of the 30-minute observation period. In a separate series of experiments the behavioral seizure score was determined with increasing cumulative doses of PTZ. Mice (18 to 24 weeks old) were injected intraperitoneally with a 10 mg/kg dose of PTZ and observed for 30 minutes. Additional doses of 20, 40, and 60 mg/kg were administered at 30-minute intervals until tonic hind limb extension occurred, at which time the mice were immediately euthanized. PTZ typically induced a sequence of behaviors that were scored in each of the 30-minute intervals after dosing according to the following scale: 0, normal behavior; 1, behavioral arrest; 2, vocalization and limb twitches; 3, sustained fore limb clonus; 4, wild running and jumping; and 5, tonic hind limb extension.
Inverted Screen Test
A modification of the horizontal screen test was used as a measure of overall motor function.18 Mice were placed on a horizontally oriented grid (consisting of parallel 1.5-mm diameter rods situated 1 cm apart), and the grid was inverted. The time the animal remained on the underside of the grid was recorded. Mice that fell from the grid within <15 seconds were considered to have reduced strength or coordination.
Footprint Pattern Test
Mice with hind paws dipped in non-toxic ink were placed at one end of a dark tunnel measuring 10 cm x 10 cm x 50 cm, the bottom of which was lined with white paper. Mice walked down the tunnel, and their footprints were used to assess the ability to walk in a straight line, irregularities, and orientation of the hind paws.19
Echocardiography
Trans-thoracic echocardiography was performed using an Acuson Sequoia equipped with a 15-MHz linear transducer (15L8) in a phased-array configuration. Animals were placed on a heating pad (37°C) and imaged under isoflurane anesthesia in the left lateral decubitus position. Two-dimensional images were obtained from parasternal long- and short-axis views with optimized gain and depth settings. M-mode images were obtained from short-axis views at the papillary muscle level at a sweep speed of 100 mm/s. Two-dimensional and M-mode images were digitized and stored for off-line analysis. Cardiac measurements were performed off-line from the M-mode images using a leading edge-to-leading edge technique according to the American Society of Echocardiography guidelines.20 Measurements from five cardiac cycles were averaged to obtain the following parameters in diastole and systole: interventricular septum (IVSd, IVSs); posterior wall (PWd, PWs); and left ventricular internal diameter (LVIDd, LVIDs). The percentage of LV fractional shortening (LV% FS) was calculated as {(LVIDd – LVIDs/LVIDd} x 100. LV mass was calculated according to the uncorrected cube method using the equation LV mass (mg) = 1.055{(IVSd+LVIDd+PWd)3 – (LVIDd)3}, as previously reported.21
ECG Recordings
ECGs were recorded in conscious and sedated mice with different Cacna2d2 status using a recently developed noninvasive ECG recording device and software (Mouse Specifics, Boston, MA), as described previously.22 Conscious or isoflurane-sedated animals were positioned on the ECG platform with gel-coated electrodes embedded in its base. The amplified signals were digitized with 16-bit precision at a sample rate of 2000 samples/second. Only continuous recordings (>10 to 25 ECG complexes) generated from mice with three paws in contact with three electrodes were analyzed. The following parameters were measured: HR and RR interval, PR interval, QRS duration and maximum height, QT and corrected QT intervals.
Blood Pressure Measurements and Clinical Chemistry
Noninvasive blood pressure and serum electrolytes were measured on each of four null and five wild-type animals by Phenotyping Service of the Jackson Laboratory (West Sacramento, CA). Growth hormone, T3, T4, and IGF-1 serum levels in each of four null and four wild-type animals were assayed by Anilytics, Inc. (Gaithersburg, MD).
Pathology
Cacna2d2+/+, Cacna2d2+/–, and Cacna2d2–/– male and female mice from 22 days to 70 days of age were completely necropsied. All tissues were fixed in neutral buffered formalin, embedded in paraffin, sections cut at 4 to 6 µm, and stained with hematoxylin and eosin (H&E). Selected tissues were stained with luxol fast blue and Bodian stains. For immunohistochemistry of the brain, antibodies to glial fibrillary acidic protein (GFAP) (DAKO, Carpinteria, CA), Ki-67 (polyclonal, Novocastra, Newcastle on Tyne, UK), calbindin (D-28K, Sigma), and caspase-3 (Promega, Madison, WI) were used. The TUNEL method (ApopTag® In Situ Apoptosis Detection kit, Intergen, Purchase, NY) was used to identify apoptotic cells.
Statistics
Data in this study were evaluated using standard analysis of variance and Student’s t-tests, repeated measures analysis of variance, profile analysis, analysis of covariance, post-hoc tests, non-parametric methods, survival analysis, and graphical descriptive techniques. Data analyzed with parametric methods were routinely tested for conformance with homogeneity of variance and covariance requirements. Kaplan-Meier plots were constructed for survival, and differences in survival distributions were tested with the log rank test.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
We constructed a Cacna2d2-targeting vector, in which a 3445-bp XbaI/HindIII fragment (positions 6593434–6596878 on NT_039477) containing seven 3'-terminal exons of the gene was replaced by a neomycin-resistance gene (Figure 1A)
. This replacement eliminated the sequence encoding the entire subunit with a single transmembrane domain plus the 87 C-terminal amino acids of the 2 responsible for binding the subunit. Genotyping of 256 F2 offspring by Southern hybridization (Figure 1B)
revealed a Cacna2d2+/+: Cacna2d2+/–: Cacna2d2–/– distribution (64:130:62) that was very close to the expected Mendelian distribution. Northern blot analysis on brain mRNAs showed expression of the full-length transcript in heterozygotes but not null-mutant animals (Figure 1, C and D)
.
Growth Delay, Neurological Phenotype, and Premature Death in Null-Mutants
While Cacna2d2+/– mice appeared healthy, had normal life spans, and were not distinguishable from their Cacna2d2+/+ littermates in any assay, null-mutants could be reliably identified at approximately 4 weeks of age and older by dystonic episodes, ataxic waddling gait typical for cerebellar dysfunction, and reflexive clutching of hind limbs when lifted by their tails (Figure 2)
. Dystonic episodes usually started with characteristic extension of the hind limbs. During the next stage that lasted from a few minutes to more than one-hour, conscious animals were immobilized on their abdomen or side by whole-body extension. After recovery, the null-mutants walked in a poorly coordinated fashion, dragging their hind limbs and often listing to one side. They frequently showed myoclonic jerking of the limbs or whole body tremor, and tonic extension of the hind limbs. In four animals, generalized clonic seizures that lasted up to 30 minutes and affected the entire body were observed. The null-mutants developed growth delay starting at the age of 4 weeks (P < 0.0001 by analysis of covariance, profile analysis, and post-hoc t-tests). The mean body weights of Cacna2d2–/– mice were 1.5 to 2 times less than that of their wild-type age mates (Figure 3A)
. Cacna2d2–/– mice also had markedly reduced life spans (P < 0.0001 by the log rank test), with a median survival time of 150 days (Figure 3B)
. Among 50 Cacna2d2–/– mice studied, 23 (46%) died prematurely with no obvious cause of death. Most of these mice died at young ages, and no premature deaths were observed at ages older than 6 months (Figure 3B)
.
|
|
Motor impairment in null-mutants was evaluated with the inverted screen test, in which the animals were required to support their body weight by grasping a wire grid. None of 6 Cacna2d2–/– remained on the inverted screen for at least 15 seconds, whereas 3 of 6 of the Cacna2d2+/– and 6 of 7 of the Cacna2d2+/+ remained on the grid. The mean times on the grid were 5.5 ± 1.5 seconds, 56 ± 23 seconds, and 138 ± 83 seconds, respectively. Compared with wild-type control mice, Cacna2d2–/– mice had significant motor impairment (P = 0.006, Kolmogorov-Smirnov test). Heterozygotes were not significantly different (P = 0.713) from wild-type mice.
Cacna2d2 Expression in Mouse Tissues
In situ hybridization with mouse embryo and adult cerebellum sections showed that Cacna2d2 is expressed primarily in the central nervous system (Figure 4)
. In the postnatal cerebellum, a very high level of gene expression was detected in the Purkinje cell layer, while granular produced poor signals (Figure 4, A and B)
. Remarkably, Cacna2d2 expression was also detected outside the central nervous system, namely, in the dorsal sympathetic ganglia (Figure 4C)
. While no Cacna2d2 expression was detected in mouse embryonic heart, previous reports indicated that this gene is highly expressed in both mouse and human adult hearts.8-10
Indeed, using the RT-PCR assay we demonstrated that the CACNA2D2 mRNA is expressed mainly in four cardiac regions: auricles, aorta, interventricular septum, and atrioventricular node (Figure 5)
. These areas represent highly innervated parts of the heart’s conductive system. Limited levels of CACNA2D2 expression were also detected in the muscle tissue, such as apex and ventricles.
|
|
To explore the possibility of heart failure due to the 2-2 deficiency we evaluated null mutants’ heart function. First, histo-pathological study of the null mutants’ hearts revealed no signs of cardiomyopathy, apoptosis, or any other abnormality (data not shown). Second, heart function was assessed in vivo by surface electrocardiography (ECG) and trans-thoracic echocardiography. A trend to bradycardia was revealed in conscious null-mutants as compared to the Cacna2d2+/– and Cacna2d2+/+ control group (747 ± 53 versus 714 ± 73, control versus knockout, P = 0.07, Table 1
). No significant differences were observed in other major ECG parameters. Remarkably, isoflurane anesthesia produced different effects on heart rates in two cohorts. While in both cases isoflurane reduced heart rate, the null-mutants’ response was less profound than that of the control cohort (186 ± 37 versus 102 ± 44, control versus knockout, P = 0.01, Table 1
). Consistent with the previous histopathological evaluation, echocardiographic study showed no evidence of systolic abnormalities in the null mutants’ echocardiograms, and no significant differences in IVSd, IVSs, PWd, PWs, LVIDd, LVIDs, LV%FS, and LV mass were found between null mutants and wild-type animals (6 null-mutants and 6 wild-type animals; data not shown). No significant differences were detected in blood pressure and serum electrolyte concentrations between Cacna2d2–/– mice and a control group (potassium, sodium, magnesium, and calcium measurements on five wild-type and four homozygous mutant mice; data not shown).
|
As noted, Cacna2d2–/– mice exhibited sporadic behavioral seizures. To further pursue the possibility of an epileptic phenotype, seizure susceptibility was determined with the convulsant PTZ. For all genotypes, there was a dose-dependent increase in the fraction of animals exhibiting clonic seizures with PTZ doses in the range of 10 to 80 mg/kg (Figure 6A)
. However, the dose-response curve for Cacna2d2–/– mice was significantly shifted to the left compared with the curves for Cacna2d2+/– and Cacna2d2+/+ littermates, indicating that the null mutants have greater seizure susceptibility. In a separate series of experiments examining the behavioral seizure score with cumulative dosing of PTZ, all genotypes exhibited a dose-dependent increase in the seizure score with PTZ doses in the range 10 to 60 mg/kg, and all showed tonic hind limb extension at PTZ doses of 60 mg/kg or less. As shown in Figure 6B
, null mutant mice exhibited higher seizure scores than their littermates at 10 and 20 mg/kg cumulative doses of PTZ, indicating that they experience more severe seizures at these doses. Both studies indicate that the null mutants have enhanced seizure susceptibility.
|
At necropsy, the null mice were runty as compared to Cacna2d2+/+ or Cacna2d2+/– littermates, dehydrated, and had little fat. Because ataxia is a clinical manifestation of disturbance in coordinated motor activity, which may be caused by cerebellar impairment, we compared cerebellums isolated from Cacna2d2–/– and wild-type mice. Necropsy of mutant mice revealed no significant differences in the overall gross cerebellum anatomy or Purkinje cell layer architecture (Figure 7)
. Consistent lesions in the cerebellum, however, were detected in the granular layer. Apoptosis and thinning (loss of granule neurons) in the granule cell layers, especially in the vermis, was detected starting from day 22, becoming more profound later. Apoptosis in the granule cell layer was initially demonstrated with hematoxylin staining (Figure 8, A and B)
and then verified with the TUNEL method and caspase-3 antibody (Figure 8, C and D
, respectively). At P70 and later GFAP, immunoreactive glial cell processes became prominent in the granule cell layer (Figure 8, E and F)
. No apoptotic bodies were seen in Cacna2d2+/+ or Cacna2d2+/– mice (data not shown). While no Purkinje cells depletion was observed in Cacna2d2–/– mice at the age of about 1 month (Figure 7B)
, first signs of Purkinje cells depletion were noticed at P70 (Figure 8, A and B)
and became more prominent later, especially in aging mice (> 1-year-old, data not shown). Since mice became clinically ill when these lesions appeared and were sacrificed or died soon thereafter, we could not determine the natural course of the lesions except in one 10-week-old mouse that exhibited reactive gliosis of Bergman glia in H&E stained sections. Lesions in other tissues, including severe thymic atrophy and suppurative rhinitis, were sometimes found in null mice. Pathological evaluation of Cacna2d2–/– hearts revealed no obvious abnormalities.
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The results of our study unequivocally implicate 2-2 deficiency in cerebellar degeneration and ataxia, seizure susceptibility, and poor survival. Comparative analysis of this first 2-type subunit knockout model, the spontaneous ducky mutant23,24
that was recently shown to carry a Cacna2d2 gene rearrangement,25
and other previously characterized calcium neurological models facilitated identification of cells, currents, and mechanisms related to 2-2 involvement in cerebellar ataxia. Homozygous inactivation of the Cacna2d2 gene performed in this study produced a phenotype that had much in common with ducky and other epileptic models, such as tottering, rolling Nagoya, and Cacna1a Fcrtm1.26-29
The observed similarities include ataxic gait, seizures, signs of cerebellar degeneration, and premature death.
Despite the fact that Cacna2d2 normally shows broad expression in embryonic (Figure 4C)
and adult14
nervous tissue, neuronal degeneration observed in Cacna2d2 null-mutants was restricted to the cerebellum. Limited impact of gene inactivation on the brain and other parts of nervous system may be explained by functional redundancy and overlapping tissue distribution reported for 2 and 1 subunits.8-10,14
The pathology observed in the cerebellar area included thinning of the granular layer in young mice followed by loss of Purkinje cells at older ages. Even though Cacna2d2 expression was prominent in the Purkinje layer and barely detectable in granular cells (Figure 4, A and B
, and earlier observations 14
), the first sign of cerebellar degeneration in Cacna2d2 null mice was the depletion of the granule cell layer (P22) via caspase 3-dependent apoptosis (Figure 8)
. Only later was this process followed by gradual loss of Purkinje cells, with most of them gone by 52 weeks (data not shown). This observation is consistent with the view that granular cells’ survival during normal cerebellum development is largely dependent on forming proper connections with Purkinje cells30
and that perturbations in Purkinje cells during critical stages of cerebellar development often result in devastating effects on granule cells.31
Purkinje cell-dependent depletion of granular neurons was described also in another cerebellar neurodegenerative model, lurcher heterozygous mice.32
Remarkably, ducky mice showed a quite distinct pattern of cerebellar pathology that included reduction in the gross size and appearance of cerebellum, demyelination, and early abnormalities in Purkinje cells morphology.23-25,33
More severe cerebellar degeneration in ducky mutants may be associated with elimination of longer portions of the 2-2 protein (Table 2)
that may produce more acute effect on assembling, stability, and function of the entire multiprotein channel complex. Unfortunately, different genetic backgrounds of these two mutations (du on TKDU and du2J on C57BLKS/J) complicates direct comparison of the ducky phenotypes with Cacna2d2–/– (129/SvJ:C57BL/6J background). Nevertheless, the work on du and du2J 25,33
in conjunction with the present model provide strong evidence for a role of 2-2 in promoting cerebellar neurons survival and functional interaction between granular and Purkinje neurons.
|
Another important feature that relates Cacna2d2-deficient mice to Cacna1a mutants is seizure-like behavior.25-29
Spontaneous behavioral seizures in our model prompted us to use the chemoconvulsant PTZ to quantitatively assess seizure susceptibility/vulnerability. The null-mutants showed a significantly reduced PTZ threshold, demonstrating their enhanced seizure susceptibility. Given the electro-physiological flexibility of 2 regulatory subunits and insufficiency in their number as compared to VDCC core subunits (3 types versus 10), one may expect that each of three 2 subunits may be found in vivo in more than one calcium channel type. In addition to its association with P/Q-type (CaV2.1) channels as discussed above, it appears that 2-2 may be the regulatory subunit of T-type (CaV3.1) Ca2+ channels that underlie rhythmogenesis in some central nervous system neurons involved in epileptic activity.8,9,14
However, a recent study on mice with a targeted disruption of the CaV3.1 core subunit gene Cacna1g showed normal susceptibility to seizures.34
This observation does not directly support the association of 2-2 with T-type channels in vivo. Further characterization of cortical calcium channels and networks with 2-2 involvement is required to verify whether P/Q channels may be solely responsible for increased seizure susceptibility in our model.
Most channelopathies associated with seizures result from alterations in channel function that are expected to enhance neuronal excitability.35
An exception to this seems to be Ca2+ P/Q channelopathies where the epileptic phenotype occurs with reduced excitatory channel function. Different mechanisms were proposed to explain how a reduction in pre-synaptic P/Q currents might underlie spontaneous spike-and-wave discharges.36,37
Based on experiments on co-expression of 2-2 with a variety of VDCC core subunits and recordings made on 2-2-deficient Purkinje cells,7,8,14,24
the lack of the 2-2 auxiliary subunit in Cacna2d2–/– mice would be also expected to diminish the depolarizing current contributing to neurotransmitter release. While the net effect of this alteration is difficult to predict, the epileptic phenotype is consistent with other Ca2+ channelopathies where a reduced Ca2+ current caused by defects in Ca2+ channel proteins does exist.38
In this context, it is interesting to note that the antiepileptic drug gabapentin binds specifically and with high affinity to 2-1 and 2-2 proteins.39
While a role for this interaction in the anticonvulsant activity of gabapentin is yet to be definitively established,40
it is nevertheless intriguing that 2-2 does appear to regulate seizure susceptibility, at least in the mouse mutants, suggesting that the interaction could very well be of clinical relevance.
Overall, our data unambiguously show that Cacna2d2, which is widely expressed in CNS (cerebellum, cortex, thalamus, spinal cord, etc) and dorsal ganglia (Figure 4
and14
), is indispensable for central nervous system function and postnatal development. Based on the similarity between Cacna1a and Cacna2d2 null phenotypes and their co-expression in Purkinje cells, it may be assumed that products of these genes serve as components of the same calcium channel, ie, CaV2.1 (P/Q). Indeed, this conclusion is supported by the reduction in the Purkinje P-type currents reported in ducky homozygotes.25
Premature Death and Cardiac Function
Although the Cacna2d2–/– mice that died at early age were runty and not thriving, there were no apparent behavioral changes or lesions that would account for sudden death. Since cardiac arrest is one of the most plausible mechanisms of unexpected death at young ages and in epileptic patients,41-43
we assessed possible involvement of the Cacna2d2 product in cardiac function. Taking into consideration that 2-2 may serve as a regulatory subunit of multiple VDCCs,8-10
we first identified human heart areas and nervous tissue where 2-2 is expressed. Localization of Cacna2d2 transcripts in heart-innervating sympathetic ganglia (Figure 4A)
complemented reports on 2-2 expression in rat root ganglia44
and suggested that 2-2 may be involved in autonomic regulation of cardiac function. Indeed, examination of the baseline ECG from conscious Cacna2d2 null mice revealed a tendency to bradycardia (Table 1)
. Isoflurane anesthesia further unveiled abnormalities in null-mutants’ heart rate regulation, suggesting that 2-2 deficiency compromises response to the anesthetic. Volatile anesthesia produces its effect in the mammalian cerebral cortex by reduction in excitatory synaptic transmission, which, in turn, is caused by a decreased amount of transmitter glutamate in the synaptic cleft.45
The observed difference in heart rate modulation, therefore, may be connected with the inability of Cacna2d2 null-mutants to efficiently down-regulate synaptic calcium currents and the neurotransmitter release.
In summary, while our animal model provides, for the first time, evidence on 2-2 involvement in cardiac rhythm regulation, no obvious cardiac mechanism was detected to explain the propensity to sudden death in the null mice. Recently, using ribonuclease protection assay Chu and Best46
identified all three 2 subunits’ transcripts in rat heart atria. More studies into the structure and function of cardiac calcium channels are required to understand the exact roles of 2-1, 2-2, or 2-3 in heart and the extent of their functional redundancy. Creating double 2 mutants may help in answering this question and further assess the significance of 2 subunits for cardiac function.
Clinical Significance of the Cacna2d2tm1NCIF Model
In line with previously characterized Cacna2d2 ducky mutants, our model clearly implies a link between 2-2 and congenital neurological disorders. Mutations in the CACNA1A gene encoding the core 1A subunit of the P/Q type VDCC have been associated in humans with familial hemiplegic migraine, episodic ataxia type-2, autosomal dominant spinocerebellar ataxia (SCA6), and idiopathic generalized epilepsies.47-49
It is of utmost importance, therefore, to see if mutations in the human CACNA2D2 gene from the 3p21.3 chromosomal region can produce effects similar to those observed in our animal model. The Cacna2d2tm1NCIF model may also serve as a valuable tool for in-depth study of calcium-dependent neuronal death and development of therapies for P-type-channels-associated congenital disorders.
Note added in proof
While this manuscript was in review, a novel spontaneous Cacna2d2 mutant, Cacna2d2entla, was characterized and crossed into a C57BL/6J background.50
This mutation produced an in-frame full-length 2-2 protein with a 39 amino acids duplication in exon 3. The subunit was capable of incorporation into cellular membrane but failed to undergo proper processing to covalently bind the subunit. Cacna2d2entla homozygotes showed a neurological phenotype and cerebellar morphology similar to our model.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part with federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. NO1-CO-12400.
Accepted for publication May 17, 2004.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2 subunit. J Neurosci 1999, 19:684-691(2) auxiliary subunit gene (CACNA2D2). J Biol Chem 2000, 275:12237-122422 subunit subtypes. J Membr Biol 2001, 184:35-43[Medline]
2 and are encoded by the same gene. J Biol Chem 1990, 265:14738-14741 2 subunit in current stimulation and subunit interaction. Neuron 1996, 16:431-440[Medline]
2-2 subunit. Eur J Neurosci 2000, 12:1217-1226[Medline]
2 -2 protein with abnormal function. J Biol Chem 2002, 277:7684-7693(1G) T-type Ca(2+) channels. Neuron 2001, 31:35-45[Medline]
(2) subunits-structure and gabapentin binding. Mol Pharmacol 2001, 59:1243-12482-subunits from rat atria and the differential regulation of their expression by IGF-1. J Mol Cell Cardiol 2003, 35:207-215[Medline]
1A-voltage-dependent calcium channel. Nat Genet 1997, 15:62-69[Medline]
This article has been cited by other articles:
 |
|
 |
|
  D. Angeloni Molecular analysis of deletions in human chromosome 3p21 and the role of resident cancer genes in disease Brief Funct Genomic Proteomic, May 24, 2007; (2007) elm007v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-H. Tang, J.-W. Lee, M. G. Galvez, L. Robillard, S. E. Mole, and H. A. Chapman Murine cathepsin f deficiency causes neuronal lipofuscinosis and late-onset neurological disease. Mol. Cell. Biol., March 1, 2006; 26(6): 2309 - 2316. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||
