Articles in PresS. Am J Physiol Cell Physiol (November 23, 2005). doi:10.1152/ajpcell.00247.2005
Dominant-Negative Effects of Human P/Q-type Ca2+ Channel
Mutations Associated with Episodic Ataxia Type 2
Chung-Jiuan Jeng1, Yu-Ting Chen2, Yi-Wen Chen2, Chih-Yung Tang2*
School of Medicine, Fu Jen Catholic University, Hsin-Chuang, Taipei County,
Taiwan1; Department of Physiology, College of Medicine, National Taiwan University,
Taipei, Taiwan2
*To whom correspondence should be addressed:
Dr. Chih-Yung Tang, Department of Physiology, College of Medicine, National
Taiwan University, Taipei, Taiwan. Phone number: 886-2-23562215. Fax:
886-2-23964350. E-mail: [email protected]
Running head: Dominant-negative effects of missense and nonsense EA2 mutations
1
Copyright © 2005 by the American Physiological Society.
ABSTRACT
Episodic ataxia type 2 (EA2) is an inherited autosomal dominant disorder related
to cerebellar dysfunction and is associated with mutations in the pore-forming α1A
subunits of human P/Q-type calcium channels (CaV2.1 channels). The majority of
EA2 mutations result in significant loss-of-function phenotypes. Whether EA2
mutants may display dominant-negative effects in human, however, remains
controversial. To address this issue, five EA2 mutants in the long isoform of human
α1A subunits were expressed in Xenopus oocytes to explore their potential
dominant-negative effects. Upon coexpressing the cRNAs of α1A WT with each α1A
mutant in molar ratios ranging from 1:1 to 1:10, the amplitude of barium currents
through wild-type CaV2.1 channels decreased significantly as the relative molar ratio
of α1A mutants increased, suggesting the presence of an α1A mutant-specific
suppression effect. When we coexpressed α1A WT with proteins not known to interact
with CaV2.1 channels, no significant suppression effects were observed. Furthermore,
increasing the amount of auxiliary subunits resulted in partial reversal of the
suppression effects in nonsense, but not missense EA2 mutants. On the other hand,
when we repeated the same coinjection experiments of α1A WT and mutant by using a
splice variant of α1A subunit that contained a considerably shorter carboxyl-terminus
(the short isoform), no significant dominant-negative effects were noted until we
2
enhanced the relative molar ratio to 1:10. Taken together, these results indicate that
for human wild-type CaV2.1 channels comprising the long α1A subunit isoform, both
missense and nonsense EA2 mutants indeed display prominent dominant-negative
effects.
KEYWORDS: Channelopathy, Voltage clamp, Xenopus oocytes, Cerebellum, Splice
variants
3
INTRODUCTION
Voltage-gated P/Q-type Ca2+ channels (CaV2.1 channels) comprise a principal
pore-forming α1A subunit, in combination with auxiliary α2δ, β, and perhaps γ
subunits (5). CaV2.1 channels play an important role in neurotransmitter release,
especially at cerebellar Purkinje cells and at somatic motor nerve terminals of
mammals (47, 51, 52, 60). CaV2.1 channels also participate in determining membrane
excitability of the dendrosomatic region of postsynaptic neurons (49, 53). The
expression of α1A subunits is distributed widely throughout the brain with especially
prominent density in the cerebellum (28, 57), highlighting the importance of CaV2.1
channels in the neurophysiological function of the cerebellum. For example, CaV2.1
channels have been shown to play a pivotal role in mediating glutamate release from
the parallel fiber nerve terminals of granule cells and in generating complex Ca2+
spikes in the dendritic trees of Purkinje cells (30, 34, 41, 46), both of which serve as
indispensable elements of the induction of long-term depression (16, 20), a process
that is believed to underlie the synaptic mechanism of cerebellar motor learning (19,
39).
Episodic ataxia type 2 (EA2) is a rare autosomal dominant neurological disorder
with distinct signs of cerebellar dysfunctions, including interictal nystagmus and
episodes of ataxia lasting hours (21). Molecular genetic analyses have linked EA2 to
4
mutations in a gene in chromosome 19p, CACNA1A, which encodes human α1A
subunits of CaV2.1 channels (36). To date, more than twenty mutations in human α1A
subunits have been identified from familial or sporadic EA2 patients, the majority of
which are nonsense (truncation) mutations and the rest missense mutations and
aberrant splicing (21, 24). So far, only a few missense EA2 mutants have been shown
to express functional human CaV2.1 channels. All of these missense mutations
displayed significantly smaller ionic currents compared with human wild-type CaV2.1
channels in the heterologous expression system (22, 45, 56), indicating that virtually
all EA2 mutants result in loss-of-function phenotypes. Thus, haploinsufficiency has
long been suggested to be one of the major molecular mechanisms underlying EA2
(14, 36, 56).
Whether dominant-negative effects also underlie the molecular mechanism of
EA2 mutations, however, remains controversial. One EA2 nonsense mutant (R1824x),
when introduced into the rabbit α1A subunit for heterologous expression, displayed
pronounced dominant-negative effects (25). When introduced into the human α1A
subunit, the same EA2 nonsense mutants, however, was shown not to possess
significant dominant-negative effects (18). Similarly, one EA2 nonsense mutant that
predicts a truncation mutation with a premature stop at the S1 segment of domain III
exhibited prominent dominant-negative effects when introduced into the rat α1A
5
subunit (37), whereas another EA2 nonsense mutant that also predicts a truncation
mutation at the same transmembrane segment failed to demonstrate any
dominant-negative effects when introduced into the human α1A subunit (56). These
discrepancies raise the possibility that dominant-negative effects may involve species
differences and question whether dominant-negative effects really occur in human
EA2
patients.
Furthermore,
whether
missense
EA2
mutants
possess
dominant-negative effects on their wild-type counterpart has never been determined.
The purpose of this study is to test the hypothesis that EA2 mutants may exert
dominant-negative effects on human CaV2.1 channels. We systemically assessed the
functional interaction between five EA2 mutants and wild-type human α1A subunit.
Our results strongly suggest that, for both missense and nonsense mutations, EA2
phenotypes may occur as a result of dominant-negative effects. The potency of the
EA2 suppression effect, however, varies significantly between two splice variants of
human α1A subunits.
6
MATERIALS AND METHODS
Molecular biology
Full-length cDNAs for all constructs of the long isoform of human α1A subunits
(GenBank accession number AF004884), as well as those for α2δ and β4 subunits,
were kindly prepared and generously provided by Dr. Joanna Jen (Dept. of Neurology,
UCLA, USA)(22). In human cerebellum, approximately two-thirds of α1A splice
variants is the long isoform (44). As indicated in the Results section, in one line of
experiment we also employed the short isoform of human α1A subunit (equivalent to
GenBank accession number AF004883, but with one splice deletion of the tripeptide
VEA)(56), which was kindly provided by Dr. Jörg Striessnig (Department of
Pharmacology and Toxicology, Institute of Pharmacy, University of Innsbruck,
Austria). To avoid confusion, we adopt a recently published nomenclature system for
the location of different mutations in α1A subunits (21). To increase the level of
protein expression in Xenopus oocytes, cDNAs for α1A (the long form) and α2δ
subunits were subcloned, using BamHI and XbaI sites, into the pGEMHE vector
(kindly provided by Dr. Emily Liman, Univ. of Southern California, USA) (29).
For in vitro transcription, cDNA was linearized with XbaI (for α1A and α2δ
subunits) or with HindIII (for β4 subunits). Capped cRNAs were transcribed in vitro
from the linearized cDNA template with the mMessage mMachine T7 kit (Ambion,
7
Austin, TX, USA). Concentration of cRNA was determined by gel electrophoresis and
verified with spectrophotometry (GeneQuant Pro RNA/DNA Calculator, Amersham
Biosciences, Piscataway, NJ, USA).
cRNA injection in Xenopus oocytes
Adult female Xenopus laevis (African Xenopus Facility, Knysna, South Africa)
were anesthetized by immersion in Tricaine (1.5 g/liter). Ovarian follicles were
removed from Xenopus frogs, cut into small pieces, and incubated in ND96 solution
[(in mM) 96 NaCl, 2 KCl, 1.8 MgCl2, 1.8 CaCl2, and 5 HEPES, pH 7.2]. To remove
the follicular membrane, Xenopus oocytes were incubated in Ca2+-free ND96
containing collagenase (2 mg/ml) on an orbital shaker (~200 rpm) for about 60-90
min at room temperature. After several washes with collagenase-free, Ca2+-free ND96,
Xenopus oocytes were transferred to ND96. Stage V-VI Xenopus oocytes were then
selected for cRNA injection.
For all cRNA injection paradigms, the total volume of injection was always 41.4
nl per oocyte. To examine the phenotype of wild-type or mutant human Cav2.1
channels, Xenopus oocytes were injected with a 1:1:1 molar ratio combination of α1A,
α2δ, and β4 cRNAs, up to a maximal cRNA amount of 81 ng per oocyte. For
coexpression experiments, it is imperative to find a submaximal cRNA concentration
for wild-type Cav2.1 channels that will allow us to add an extra amount of cRNA for
8
mutant α1A subunits. Therefore, we empirically set up a fixed concentration of cRNA
for injection, known as the standard cRNA cocktail, in which α1A, α2δ, and β4 cRNAs
were mixed in a molar ratio of 1:2:2, and the final injection amount for α1A, α2δ, and
β4 cRNAs was about 3.8ng, 3.8ng, and 1.9 ng per Xenopus oocyte, respectively.
Where necessary, a 6-fold dilution of the standard cRNA cocktail, which is called the
low cRNA cocktail, was instead used for Xenopus oocyte injection. Injected Xenopus
oocytes were stored at 16°C in ND96 and functionally assayed 2-5 days after cRNA
injection.
Electrophysiology and data analysis
For functional studies, Xenopus oocytes were transferred to a recording bath
containing Cl--free Ba2+ solution of the following composition (in mM): 40 Ba(OH)2,
50 NaOH, 2 CsOH, 5 HEPES (pH 7.4 with methanesulfonic acid). To minimize the
contribution of endogenous Ca2+-activated Cl- currents, added to the bath solution was
niflumic acid (0.4 mM), which has been shown to display potent blocking effects on
endogenous Ca2+-activated Cl- currents in Xenopus oocytes (58). The removal of
contaminating Cl- currents by niflumic acid was verified by the suppression of slow
inward tail currents during repolarization after a depolarizing test pulse (6). The bath
volume was about ~200 μl. An agarose bridge was used to connect the bath solution
with a ground chamber (containing 3M KCl) into which two ground electrodes were
9
inserted. Borosilicate electrodes (0.1 –1 MΩ) used in voltage recording and current
injection were filled with 3M KCl. Barium currents through Cav2.1 channels were
acquired, using the conventional two-electrode voltage-clamp technique with an
OC-725C oocyte clamp (Warner, Hamden, CT, USA). Data were filtered at 1 kHz
(OC-725C oocyte clamp) and digitized at 100 μs per point using the Digidata 1332A /
pCLAMP 8.0 data acquisition system (Axon Instruments, Foster City, CA, USA). The
holding potential was set at -90 mV, and 70-ms test pulses were typically applied
from -80 to +70 mV, in 10 mV increments. Passive membrane properties were
compensated using the –P/4 leak subtraction method provided by the pCLAMP 8.0
software. All recordings were performed at room temperature (20-22 °C).
Data analyses were performed via built-in analytical functions of the pCLAMP
8.0 software. Peak barium current amplitudes measured at each test voltage were
plotted against voltage (current-voltage curve, I-V curve). Apparent reversal
potentials (Vrev) were then determined by extrapolating the I-V curve to the voltage
axis (typically about +65 mV), and macroscopic channel conductance (G) at each test
potential was calculated by the equation: G = I / (V - Vrev), where I is the peak current
amplitude and V the test potential. The voltage dependence of activation was
determined by fitting conductance-voltage (G-V) curves with a simple Boltzmann
equation: G/Gmax = 1 / {1 + exp[(V0.5 - V) / k ]}, where Gmax is the maximum
10
conductance, V0.5 the half-maximal voltage for activation, and k the slope factor of the
G-V curve.
All values were presented as mean ± SEM. The significance of the difference
between two means was tested using the Student’s t test, whereas means from
multiple groups were compared using the one-way ANOVA analysis. All statistical
analyses were performed with the Origin 7.0 software (Microcal Software,
Northampton, MA, USA).
11
RESULTS
Functional expression of wild-type and mutant α1A subunits in Xenopus oocytes
We focus on five EA2-related mutant α1A subunits (α1A mutants) that have been
previously reported. Three of them involve nonsense mutations that introduce a
premature stop codon at the site of mutation (R1281x, R1549x, R1669x) (23, 24, 62)
while the other two are missense mutations (F1406C, E1761K)(8, 22). As shown in
Figure 1A, all the five mutations take place in the domain III to IV region of the α1A
subunit. The functional properties of some of these mutants have been studied before,
using cell line expression systems such as human embryonic kidney (HEK) cells.
However, an EA2 mutant that is nonfunctional when expressed in a HEK-derived cell
line was found to exhibit detectable barium currents upon being expressed in Xenopus
oocytes (56). We therefore re-examine the functional phenotype of the five
aforementioned mutants, using the Xenopus oocyte expression system. Figure 1B
illustrates the typical barium currents recorded from Xenopus oocytes injected with
either wild-type α1A subunit (α1A WT) or missense α1A mutants. Auxiliary α2δ and β4
subunits were coinjected with individual α1A subunits into Xenopus oocytes in the
molar ratio of 1:1:1 (α1A:α2δ:β4). The maximal inward barium currents through
human CaV2.1 channels were usually recorded at the test potentials of +10 to +20 mV.
Consistent with a previous functional analysis using COS7 cells as the expression
12
system (22), the missense mutation F1406C produced dramatically reduced currents
in Xenopus oocytes. The other missense mutation, E1761K, however, exhibited no
significant barium currents in Xenopus oocytes. To the best of our knowledge, this is
the first functional study to demonstrate that the mutation E1761K renders human
CaV2.1 channels nonfunctional. Figure 1C shows the scatter plot of the amplitudes of
barium currents at +20 mV recorded from oocytes expressing wild-type or mutant
CaV2.1 channels. No detectable currents in Xenopus oocytes were seen in the three
nonsense mutations (Fig. 1C), which is the typical phenotype to be expected of
truncation mutations. Thus, our functional expression studies show that all five α1A
mutants under investigation result in a significant loss of channel function in Xenopus
oocytes.
Since EA2 is an autosomal dominant disease, most EA2 patients are expected to
carry heterozygous CACNA1A genotype: one copy each of the wild-type and mutant
allele. As far as the molecular mechanism of EA2 is concerned, a loss-of-function
phenotype of the mutant CACNA1A gene will certainly support haploinsufficiency as
a reasonable mechanism causing the neurological disorder. It is likely, however, that
α1A mutants may negatively influence the proper function of α1A WT by, for example,
disrupting channel assembly via competing for the binding of auxiliary subunits,
which are known to up-regulate the functional expression of voltage-gated Ca2+
13
channels in the heterologous expression system (2, 50). We therefore wanted to
explore the possibility that dominant-negative effects may serve as a major molecular
mechanism underlying EA2.
Coexpression of equal-molar ratios of α1A WT and α1A mutants in Xenopus oocytes
By definition, pure haploinsufficiency effects will be expected of nonfunctional
α1A mutants that display neither functional nor structural interactions with α1A WT.
Accordingly, upon coexpressing α1A WT and α1A mutant in a 1:1 molar ratio in
Xenopus oocytes, a significant reduction of the Ca2+ channel function (such as the
amplitude of barium currents) relative to that of expressing α1A WT alone, will
support the idea that the α1A mutant possesses dominant-negative effects. On the other
hand, if the current amplitude of α1A WT turns out to be not significantly affected by
the presence of the α1A mutant, it will be counted as evidence against
dominant-negative effects. We thus set forth to perform experiments on the
coexpression of α1A WT and α1A mutant.
We first empirically established a standard concentration of cRNA for Xenopus
oocyte injection, which we called the standard cRNA cocktail (see Method section), to
ensure that the amplitude of barium currents through human wild-type CaV2.1
channels is large enough to conduct detailed analyses. In the standard cRNA cocktail,
a fixed concentration of cRNAs for α1A WT, α2δ, and β4 subunits were mixed in a
14
molar ratio of 1:2:2, prior to injection into oocytes. In other words, the standard cRNA
cocktail can be regarded as a control for haploinsufficiency effects since it is
equivalent to coexpressing α1A WT with a hypothetical nonfunctional α1A mutant that
does not exert dominant-negative effects. To evaluate the dominant-negative effects of
the five EA2 mutants, Xenopus oocytes were coinjected with α1A WT and α1A mutant
cRNAs mixed in the molar ratio of 1:1; i.e., one copy of the cRNA for an α1A mutant
was added to the standard cRNA cocktail. As stated above, for all cRNA injection
paradigms, the total volume of injection per oocyte was identical (41.4 nl).
As shown in the representative data in Figure 2A, the amplitude of barium
currents recorded from Xenopus oocytes coexpressing α1A WT with any of the five
EA2 mutants was significantly smaller than that for oocytes expressing α1A WT alone.
Since it is not uncommon to observe a wide range of variation in the amplitude of
barium currents through wild-type CaV2.1 channels among different oocytes (see Fig.
1C), we adopted a normalization procedure to perform an objective comparison of the
expression levels of CaV2.1 channels. With the same batch of oocytes on a given day
of experiment, the average value of the peak barium current amplitudes at +20 mV
was calculated from ten or more oocytes expressing wild-type CaV2.1 channels. The
average value was then used to normalize the peak +20-mV current amplitudes
measured from Xenopus oocytes subjected to different coinjection paradigms.
15
Normalized data from different batches of Xenopus oocytes on different dates were
later pooled together for comprehensive analyses. In addition, to avoid biased data
due to unhealthy Xenopus oocytes, we only analyzed data collected from experiments
in which there was no apparent difference in the viability of Xenopus oocytes between
different injection paradigms. As shown in Figure 2B, in the presence of nonsense and
missense EA2 mutants, the current amplitude of wild-type CaV2.1 channels decreased
by about 60-80%, suggesting that the functional expression of wild-type CaV2.1
channels is significantly lower in the presence of α1A mutants.
Since the gating of CaV2.1 channels is determined by the membrane potential
experienced by the voltage sensors located in α1A subunits, it is possible that the
foregoing reduction in barium currents may arise from the alteration of the
voltage-dependence of activation of α1A WT by the EA2 mutants. To examine this
possibility, we compared the steady-state activation property (conductance-voltage
curve, G-V curve) of CaV2.1 channels. As shown in Figure 3 and Table 1, no
significant shift in the G-V curves was noticed in the absence or presence of α1A
mutants, consistent with the idea that the EA2 mutants failed to affect the voltage
sensitivity of wild-type CaV2.1 channels. Likewise, other biophysical properties of
human CaV2.1 channels, such as activation and inactivation kinetics, were not
significantly affected by the presence of the EA2 mutants (data not shown).
16
Therefore, coexpression of α1A WT with any of the five EA2 mutants under
investigation results in a significant suppression of the current amplitudes of
wild-type Cav2.1 channels that cannot be explained by a haploinsufficiency effect,
which is consistent with the idea that the EA2 mutants may exert dominant-negative
effects.
Increasing the relative molar ratios of α1A mutants reveals significant
dominant-negative effects of EA2 mutations
The decreased amplitude of Cav2.1 currents in the presence of α1A mutants
presumably reflects an α1A mutant-specific suppression of the functional expression
of wild-type Cav2.1 channels. If there is indeed an α1A mutant-specific suppression
effect, an increase in the relative molar ratio α1A mutant should enhance this
suppression effect. We therefore address this issue by testing the effect of
coexpressing α1A WT and α1A mutants in the molar ratios of 1:1, 1:3, 1:5, and 1:10,
by increasing the amount of α1A mutant cRNA added to the standard cRNA cocktail.
For all cRNA injection paradigms, the total volume of injection per oocyte was
identical (41.4 nl).
Figure 4A illustrates the representative barium currents recorded from Xenopus
oocytes coinjected with different molar ratios of cRNAs for wild-type and R1281x
α1A subunits, clearly demonstrating that R1281x mutant decreased the expression of
17
wild-type Cav2.1 channels in a dose-dependent manner. Compared to the mean
current amplitude at +20 mV for the expression of α1A WT alone, coexpression with
the α1A mutant in the molar ratio of 1:1, 1:3, 1:5, and 1:10 resulted in the normalized
amplitude of about 36%, 21%, 16%, and 14%, respectively (Fig. 4B), indicating that
the amplitude of barium currents decreased as the relative molar ratio of R1281x
increased. The presence of an increased amount of R1281x did not result in any
significant shift in the current-voltage relationship (I-V curves) of barium currents
(Fig. 4C), suggesting that the foregoing reduction of barium current is not due to a
change in the voltage-dependence of wild-type CaV2.1 channels. In addition, there
was no apparent difference in the viability of oocytes between different coinjection
paradigms (data not shown), suggesting that the observed decline in the CaV2.1
channel expression did not arise from a deterioration of the viability of Xenopus
oocytes as a result of an increased amount of injected cRNA.
Similar suppression effects on α1A WT were also observed in the other nonsense
as well as missense mutants (Fig. 5). For instance, with F1406C mutant, the
normalized mean current amplitude at +20 mV decreased from about 38% at 1:1
coexpression ratio to only 3% at 1:10. Likewise, with R1669x mutant, the mean
current amplitude at 1:1 and 1:10 molar ratio was about 22% and 4%, respectively.
18
Increased molar coexpression of the EA2 mutants did not significantly shift the
position of the I-V curves of CaV2.1 channels along the voltage axis (data not shown).
An alternative explanation for the suppression effects of the EA2 mutants
discussed above is the exhaustion of the cellular machinery for protein synthesis as a
result of excessive amount of cRNA injected into Xenopus oocytes. To address this
issue, we repeated the same experiments by coexpressing wild-type CaV2.1 channels
with two different types of proteins: densin-180 and rat olfactory channels (rolf).
Densin-180 is a 180-kDa transmembrane protein that is tightly associated with the
postsynaptic density in neurons and is postulated to function as a synaptic adhesion
molecule (1, 55). As depicted in Figure 6A, the average size of barium currents
through wild-type CaV2.1 channels did not decrease significantly in the presence of an
excessive amount of densin-180. Since all previously reported heterologous
expressions of densin-180 involve mammalian expression systems only, one may
argue that the lack of suppression effect reported here could be interpreted as a failure
of the functional expression of densin-180 in Xenopus oocytes. In light of this
possibility, we then performed the coexpression experiment using rolf, which is a
cyclic nucleotide-gated ion channel that can be functionally expressed in oocytes (48).
As shown in Figure 6B, the mean current amplitude of wild-type CaV2.1 channels did
not decrease significantly in the presence of an excessive amount of rolf either. Taken
19
together, these results are consistent with the idea that coexpressions of α1A WT and
α1A mutants fail to deplete the cellular machinery for protein synthesis in Xenopus
oocytes.
Another possibility to explain the suppression effects of the EA2 mutants is that
there exist in Xenopus oocytes an α1A subunit-specific suppression effect on either the
cellular machinery for protein synthesis or post-translational processing of CaV2.1
channels per se; i.e., both α1A WT and α1A mutant may confer this suppression effect.
To test this hypothesis, we should have perhaps evaluated the effect of adding up to
10-fold extra amount of α1A WT cRNA for oocyte injection. However, since the
amount of α1A WT cRNA is rather high in the standard cRNA cocktail, the viability of
Xenopus oocytes significantly deteriorated by merely doubling the amount of α1A WT
cRNA (equivalent to 1:1 coinjection ratio), presumably a consequence of
Ca2+-mediated cytotoxic effects as a result of an over expression of wild-type CaV2.1
channels in Xenopus oocytes. We therefore performed this experiment using the low
cRNA cocktail for Xenopus oocyte injection, in which the amount of cRNA for α1A,
α2δ, and β4 subunits was only one sixth of that for the standard cRNA cocktail. As
shown in Figure 7A, at low cRNA concentration, increasing the amount of α1A WT
cRNA injected in Xenopus oocytes did not lead to the suppression of Cav2.1 currents.
Instead, the expression of Cav2.1 channels became increasingly higher. For example,
20
when we enhanced the amount of cRNA for α1A WT by 6-fold (coinjection ratio of
1:5), which is equivalent to the amount of α1A WT in the standard cRNA cocktail, the
mean amplitude of barium currents at +20 mV increased by about 83%. When we
enhanced the amount of α1A WT cRNA further by 11-fold (coinjection ratio of 1:10),
the expression of Cav2.1 channels did not increase accordingly but was still 57%
higher than the control condition. This apparent reversal of the trend of increased
expression of Cav2.1 channels at 1:10 coinjection ratio, which is equivalent to
doubling the amount of α1A WT cRNA in the standard cRNA cocktail, was concurrent
with presence of the deterioration of the viability of Xenopus oocytes as a result the
presumable Ca2+-mediated cytotoxic effects mentioned above, and thus should not be
attributed to a suppression of the expression of Cav2.1 channels. In contrast, when we
coexpressed wild-type and mutant α1A subunits in 1:1 molar ratio using the low cRNA
cocktail, the suppression effect of EA2 mutations was still significant, ranging from
50-88% (Fig. 7B). In fact, except for E1761K, the suppression effects of EA2
mutations using the low cRNA cocktail were actually more prominent than those
using the standard cRNA cocktail (compare Fig. 2B & 7B). Taken together, our results
indicate that the decreased expression of human wild-type CaV2.1 channels in the
presence of the EA2 mutants is indeed conferred by the specific dominant-negative
effects of the α1A mutants.
21
Enhancing the amount of auxiliary subunits partially reverses dominant-negative
effects of nonsense EA2 mutants
It is known that one of the major roles of Ca2+ channel auxiliary subunits is to
facilitate membrane insertion of α1A subunits, thereby ensuring proper functional
expression of Cav2.1 channels (2, 50). Consequently, EA2 mutants may suppress the
functional expression of their wild-type counterpart by competing for the availability
of auxiliary α2δ and β4 subunits. To test this hypothesis, we ask if the suppression
effect bestowed by the five EA2 mutants can be reversed by the presence of an
excessive amount of auxiliary α2δ and β4 subunits. As depicted in Figure 8A, with the
three nonsense mutants, up to 4-fold increase of the auxiliary subunits resulted in a
small (about 20%) reversal of the dominant-negative effects. This partial reversal of
suppression effects is not likely to be due to a direct up-regulation of the expression of
α1A WT by the auxiliary α2δ and β4 subunits, since the expression level of wild-type
Cav2.1 channels did not significantly alter with up to 8-fold increase of auxiliary
subunits (data not shown). On the contrary, in the presence of additional auxiliary
subunits, the two missense mutants showed no significant change in the suppression
effects (Fig. 8B). Thus, our results demonstrate that competition for the availability of
auxiliary α2δ and β4 subunits can partially account for the dominant-negative effects
of nonsense, but not missense mutants.
22
The dominant-negative effect of R1281x is dramatically reduced for the short
isoform of human α1A subunits
R1281x mutant predicts a truncation mutation with a premature stop at the S1
segment of domain III (see Fig. 1A). Our finding that R1281x displays potent
dominant-negative effects (Fig. 4) is in direct contrast to a previous report showing
that R1281x cRNA coinjection in Xenopus oocytes failed to exhibit significant
suppression effect on the current amplitude of human wild-type CaV2.1 channels (56).
One difference in the methodology between the two studies involves different splice
isoforms of human α1A subunits. The α1A subunit we have used in the
above-mentioned experiments is the so-called long isoform of CACNA1A (GenBank
accession number AF004884) that is about 234-amino acids longer than the short
isoform used by Wappl et al. (GenBank accession number AF004883); the major
difference between the two structural variants is that the long isoform contains a long
carboxyl terminus with 11 consecutive glutamine residues (15). Incidentally, another
EA2 nonsense mutant R1824x, which predicts a premature stop at the domain IV S6
segment resulting in complete loss of the carboxyl terminus, displayed pronounced
dominant-negative effects when it was introduced into the rabbit α1A subunit for
heterologous expression (25), but was shown not to possess significant
23
dominant-negative effects upon being introduced into the short isoform of human α1A
subunit (18). Therefore, these discrepancies raise the possibility that splice variants of
CACNA1A may display differential interactions between α1A WT and EA2 mutants.
To address this possibility, we repeated the coexpression experiment for α1A WT
and R1281x by using the short isoform of α1A subunit for both the WT and mutant.
Upon coexpressing α1A WT and R1281x cRNAs in 1:1 molar ratio in Xenopus
oocytes, there was a small (about 24%) and statistically insignificant (p>0.05)
reduction in the mean amplitude of barium currents relative to that of expressing α1A
WT alone (Fig. 9). Similarly, no significant suppression effects were observed for
coinjection ratios of 1: 3 and 1:5 either. A prominent dominant-negative effect,
however, was discerned when we increased the coinjection ratio further to 1:10,
whereby the mean current amplitude at +20 mV decreased by about 67%, which was
quantitatively similar to the extent of current reduction conferred by equal-molar
coexpression of α1A WT and R1281x long-isoforms (see Fig. 4). These data indicate
that the dominant-negative effect of R1281x mutation is dramatically reduced for the
short isoform of human α1A subunits, consistent with the idea that the two splice
variants of CACNA1A display differential interactions between α1A WT and EA2
mutants.
24
DISCUSSION
Both missense and nonsense EA2 mutants exhibit dominant-negative effects on
human wild-type CaV2.1 channels comprising the long α1A subunit isoform
EA2 is an autosomal dominant neurological disorder associated with mutations
in the pore-forming α1A subunits of human CaV2.1 channels. Since the majority of
EA2 mutations result in a loss-of-function phenotype, both haploinsufficiency and
dominant-negative effects may contribute to the mechanism leading to the cerebellar
dysfunctions as seen in EA2 patients. In theory, both haploinsufficiency and
dominant-negative effects may result in the functional dominance of a
loss-of-function mutation. Haploinsufficiency-conferred dominance is usually not a
direct result of the reduced gene expression level per se, rather it typically arises from
an ineffective feedback system for the regulatory mechanism to compensate the
reduced expression, as exemplified by many human diseases caused by null mutations
of transcription factors (42, 54). On the other hand, dominant-negative effects involve
the interference of the function of a normal protein by the mutant one. Interestingly,
different mutations of the same gene may lead to either haploinsufficiency or
dominant-negative effects (31, 43).
To the best of our knowledge, we have presented the first evidence in human
CaV2.1 channels that EA2 mutants, three nonsense mutations (R1281x, R1549x,
25
R1669x) as well as two missense mutations (F1406C, E1761K), display prominent
dominant-negative effects. Our studies suggest that the intense suppression effects
conferred by both the missense and nonsense EA2 mutants are well beyond the
expectation of simple haploinsufficiency effects. Previously, two other EA2 nonsense
mutants, when introduced into rabbit or rat α1A subunits, have also been shown to
exhibit significant dominant-negative effects (25, 37). The inheritance modes of these
seven EA2 mutations include both de novo and familial (autosomal dominant)
mutations, suggesting that the dominant-negative effect may be a general feature of
EA2 mutations. We thus propose that dominant-negative effects may serve as the
major molecular mechanism underlying EA2 phenotypes.
Our proposal is consistent with the findings from previous studies involving
genetic elimination of the α1A gene in mice. Homozygous null (α1A -/-) mice
developed progressive neurological deficits such as ataxia and dystonia, and died
within four weeks after birth (13, 26). In contrast, heterozygous α1A +/- mice were
phenotypically normal (13, 26), suggesting that the decrease of α1A WT conferred by
the single-alleled knock-out fails to produce detectable neurological dysfunction in
mice. Our coexpression results clearly indicate that the reduction of the amount of α1A
WT bestowed by the dominant-negative effects of EA2 mutants will be significantly
more than that of the heterozygous α1A +/- condition, but also significantly less than
26
that of the homozygous α1A -/- condition. This may explain why EA2 patients do not
display normal motor function like the α1A +/- mice, and why neither do they develop
the severe and lethal neurological deficits found in α1A -/- mice.
Interestingly, the present study demonstrates that the dominant-negative effect of
EA2 mutants is notably more potent in human CaV2.1 channels comprising the long
α1A subunit isoform. The long and short splice variants are generated by the
differential usage of the splice acceptor at the boundary of intron 46 and exon 47 of
CACNA1A (44, 64); the short α1A subunit isoform contains a stop codon immediately
after exon 46, whereas the long isoform contains a longer carboxyl tail encoded by
exon 47. In human cerebellum, approximately two-thirds of the transcripts of
CACNA1A is the long isoform (44). One potential functional distinction between these
two isoforms has been implicated in a recent report showing that alternative splicing
at the carboxyl terminus confers different modes of Ca2+/Calmodulin-dependent
facilitation of CaV2.1 channels (7). Our demonstration that the two splice variants of
CACNA1A display differential interactions between α1A WT and EA2 mutants further
highlights the physiological significance of splice-related functional diversity. We
have shown here that for the short splice variant, the suppression effect of R1281x
was not significant until the relative molar ratio was enhanced to 1:10. Based on the
assumption that the genes for α1A WT and the EA2 mutant share comparable
27
transcription efficiency in vivo, an important implication of our data is that the
dominant-negative effect of EA2 mutants is physiologically negligible for the short
α1A subunit isoform. Since the major structural distinction between the two α1A
subunit isoforms lies in the length of the carboxyl terminus, identification of the
molecular mechanism underlying this differential EA2 suppression effect now
emerges as a critical challenge for the future.
That the suppression effect of EA2 mutants varies considerably between the two
splice variants may constitute a plausible rationale for reconciling our findings with
those from the two previous reports (25, 56) showing that EA2 mutants fail to exhibit
detectable dominant-negative effects on human wild-type CaV2.1 channels comprising
the short α1A subunit isoform (see the last section of Results). Our conclusion that the
short α1A subunit isoform of R1281x exhibits considerable suppression effect when
coinjected in 1:10 molar ratio, however, contrasts with the observation by Wappl et al.
that a 20-fold excess of R1281x coinjection in Xenopus oocytes failed to exhibit any
detectable effect (56). One difference in the methodology between the two studies lies
in the auxiliary β subunit subtype: Wappl et al. used the β1 subunit, whereas we chose
the β4 subunit. A recent report demonstrated that a rat EA2-like α1A truncation
mutation that also predicts a premature stop at the S1 segment of domain III,
displayed marked dominant-negative effects on rat wild-type CaV2.1 channels
28
comprising β4 subunits (37). Similarly, a rabbit 95-kD two-domain form of α1A
subunit also showed significant dominant–negative effects on rabbit wild-type CaV2.1
channels comprising β1 subunits (3). Thus, whether the difference in β subunit
subtype could explain the discrepancy between the two studies remains to be clarified.
Differences among heterologous expression systems used by various labs may
also contribute to the foregoing conflicting results. One important issue that needs to
be addressed in the future, therefore, is whether the dominant-negative effects we
observed in oocytes also exist in human beings. The dominant-negative effect of the
rat EA2-like α1A truncation mutant was present in both oocyte and COS-7 expression
systems (37), and the suppression effect of the rabbit two-domain isoform was
demonstrated in a mammalian cell line (3). Hence, our results here are unlikely to be
an oocyte-specific phenomenon.
Potential molecular mechanisms underlying the dominant-negative effects of
EA2 mutants
What is the mechanism underlying the dominant-negative effects of the five EA2
mutants? At least two scenarios may account for a reduction of the macroscopic
current amplitudes of Ca2+ channels: 1) altered biophysical properties and 2) defective
biosynthesis processes. We have shown that the reduction of the current amplitude of
wild-type CaV2.1 currents in the presence of EA2 mutants is not due to an alteration
29
of the voltage-dependent gating property of the channel (Fig. 3). Previous single
channel analyses revealed that the gating kinetics of an EA2 missense mutant was
significantly different from that for wild-type (56); it is still unknown, however,
whether this EA2 mutant possesses any dominant-negative effect. A non-functional
two-domain isoform of rabbit α1B (N-type) subunit has previously been shown to
display dominant-negative effect on rabbit wild-type CaV2.2 channels; when these
channels were functionally examined at single channel level, the biophysical
properties of wild-type CaV2.2 channels were found to remain virtually the same in
the absence or presence of the truncated construct (38). Therefore, it is still an open
question whether the single channel properties, such as gating kinetics and single
channel conductance, of human wild-type CaV2.1 channels may be modified in the
presence of EA2 mutants.
In addition to an alteration of biophysical property, the mechanism of
dominant-negative effect may also be explained by a decreased number of normal
CaV2.1 channels in the plasma membrane, i.e. a faulty biosynthesis of wild-type
CaV2.1 channels in the presence of EA2 mutants. The biosynthesis process of ion
channels can be divided into two general steps (9): biogenesis (such as protein folding
and coassembly of multiple subunits) and membrane trafficking (including clustering
and localization).
30
It is still not clear whether the truncated α1A subunits encoded by the nonsense
mutations R1281x, R1549x, and R1669x can be properly inserted into the plasma
membrane. Depending on the location of premature termination codons within an
mRNA, mRNAs derived from nonsense mutations may trigger nonsense-mediated
mRNA decay, thereby preventing the production of the encoded truncated proteins
(33). Consequently, some nonsense EA2 mutants may not be expressed at significant
levels due to this post-transcriptional mRNA quality control mechanism. Alternatively,
granted that normal transcription and translation processes do take place, a misfolded
protein like truncated channels may fail to pass protein quality control mechanisms
taking place in the endoplasmic reticulum (ER), resulting in defective membrane
trafficking
and
increased
protein
degradation
(10,
32).
Similar
to
the
dominant-negative effects of mutations in cardiac HERG K+ channels (12, 27, 63),
through yet-to-be identified mechanisms, truncated EA2 mutants may cause α1A WT
to misfold, thereby preventing proper trafficking and increasing protein degradation
of the wild-type CaV2.1 channel. Furthermore, an accumulation of misfolded protein
in ER may trigger an ER-mediated translation inhibition mechanism known as the
unfolded protein response (17). A recent study on the mechanism of the
dominant-negative effect of a rat EA2-like α1A nonsense mutant suggests that the
truncated construct may interact with the cognate full-length α1A WT to activate an
31
unfolded protein response, leading to the suppression of the translation of α1A WT
(37). More experiments will be needed to determine if the human EA2 truncation
mutants we studied here can also activate a similar unfolded protein response.
The auxiliary α2δ and β4 subunits play an essential role in facilitating the
membrane targeting of α1A subunits (2, 50). Accordingly, EA2 mutants may suppress
the functional expression of Cav2.1 channels by competing for the availability of
auxiliary subunits. Increasing the amount of auxiliary α2δ and β4 subunits by 4-fold
resulted in a small reduction of the dominant-negative effects of the three truncation
mutants, whilst the same treatment did not significantly affect the suppression effects
of the two missense mutants (Fig. 8), suggesting that competition for the availability
of auxiliary α2δ and β4 subunits may contribute to the dominant-negative effects of
nonsense EA2 mutants. Further studies will be required to determine whether
competition for other chaperon proteins may also be involved in the suppression
effect of EA2 mutants.
Our observation that competition for auxiliary subunits can partially account for
the dominant-negative effects of nonsense, but not missense mutants implies that the
dominant-negative effects conferred by missense and nonsense mutants may not
necessarily share the same molecular mechanism. In fact, the missense mutants
F1406C and E1761K may not necessarily encode misfolded proteins. The
32
loss-of-function phenotype of F1406C may reflect a significant alteration of the
biophysical property of the channel conferred by the mutation located at the outer
pore region (the S5-S6 linker) of domain III (see Fig. 1A). Similarly, with E1761K,
the mutation site corresponds to one of the four conserved glutamate residues within
the pore that dictate the divalent-cation selectivity of the channel (11, 61). In rabbit
cardiac L-type Ca2+ channels, individual replacements of each of the four conserved
glutamate residues with lysine resulted in functional channels that were significantly
more permeable to monovalent cations (61). Although there is no evidence yet, we
speculate that both F1406C and E1761K mutants are likely to form properly folded
α1A subunits that upon coassembly with the auxiliary subunits, are targeted for the
plasma membrane. Whether (or how) the translation inhibition mechanism is
applicable to the dominant-negative effects of missense mutants requires further
investigation. In addition, it will be of great interest to determine if other identified
EA2 missense mutants also possess dominant-negative effects.
For missense (and perhaps nonsense) EA2 mutants that retain proper membrane
trafficking properties, their loss-of-function phenotypes imply that they will behave
like virtually silent channels in the plasma membrane. The dominant-negative effects
of these EA2 mutants can then be explained by a competition between wild-type and
silent CaV2.1 channels for a limited number of “channel slots” in the membrane. In
33
other words, in the presence of EA2 mutants, the whole-cell Ca2+ current amplitude
decreases significantly because considerably less functional CaV2.1 channels gain the
access to a common membrane trafficking pathway which does not distinguish
between wild-type and functionally silent channels. Such a slot theory has recently
been put forward by Cao et al. (4), who elegantly demonstrated that human wild-type
and mutant CaV2.1 channels could compete for specific channel type-preferring slots
located at presynaptic terminals, and that the mutant channels’ impairment in
contributing to neurotransmission matched precisely with their deficiency in
supporting whole-cell Ca2+ current density. Notably, one of the mutants tested by Cao
et al. involves a quadruple-mutation (E4A) of the four glutamate residues at the
selectivity filter, similar to the missense mutant E1761K employed in the present
study.
One remarkable implication of the slot theory is that the dominant-negative
effects of EA2 mutants may also affect the clustering and subcellular localization of
wild-type CaV2.1 channels. In human cerebellum, where the long α1A subunit splice
variant is the dominant isoform (44), quantitative and qualitative alterations in the
expression pattern of CaV2.1 channels will lead to dire consequences such as
erroneous wiring of synaptic inputs from parallel fibers and climbing fibers during
synaptogenesis (35, 40), and aberrant burst firing patterns of Purkinje cells (46, 59).
34
Since Purkinje cells are the only output neurons of the cerebellar cortex, the foregoing
neuropathological scenarios may in turn contribute to the development of EA2-related
ataxic symptoms.
35
ACKNOWLEDGEMENTS
We are indebted to Drs. Joanna C. Jen and Jijun Wan for providing WT and mutant
cDNA clones, for subcloning the constructs into pGEMHE vectors, and for valuable
discussions throughout the whole process. We thank Drs. Jörg Striessnig and Martina
Sinnegger-Brauns for preparing and providing the short isoform of α1A cDNAs. We
also thank Dr. Ting-Chi Tang for critically reviewing the manuscript.
GRANTS
This work was supported by research grants from National Science Council, Taiwan,
to C.J.J. (NSC92-2320-B-002-085) and C.Y.T. (NSC91-2320-B-002-117).
36
REFERENCE
1. Apperson ML, Moon IS, and Kennedy MB. Characterization of densin-180, a
new brain-specific synaptic protein of the O-sialoglycoprotein family. J Neurosci 16:
6839-6852, 1996.
2. Arikkath J and Campbell KP. Auxiliary subunits: essential components of the
voltage-gated calcium channel complex. Curr Opin Neurobiol 13: 298-307, 2003.
3. Arikkath J, Felix R, Ahern C, Chen CC, Mori Y, Song I, Shin HS, Coronado
R, and Campbell KP. Molecular characterization of a two-domain form of the
neuronal voltage-gated P/Q-type calcium channel alpha(1)2.1 subunit. FEBS Lett 532:
300-308, 2002.
4. Cao YQ, Piedras-Renteria ES, Smith GB, Chen G, Harata NC, and Tsien
RW. Presynaptic Ca2+ channels compete for channel type-preferring slots in altered
neurotransmission arising from Ca2+ channelopathy. Neuron 43: 387-400, 2004.
5. Catterall WA. Structure and regulation of voltage-gated Ca2+ channels. Annu
Rev Cell Dev Biol 16: 521-555, 2000.
6. Charnet P, Bourinet E, Dubel SJ, Snutch TP, and Nargeot J. Calcium
currents recorded from a neuronal alpha 1C L-type calcium channel in Xenopus
oocytes. FEBS Lett 344: 87-90, 1994.
7. Chaudhuri D, Chang SY, DeMaria CD, Alvania RS, Soong TW, and Yue DT.
Alternative splicing as a molecular switch for Ca2+/calmodulin-dependent facilitation
of P/Q-type Ca2+ channels. J Neurosci 24: 6334-6342, 2004.
8. Denier C, Ducros A, Durr A, Eymard B, Chassande B, and
Tournier-Lasserve E. Missense CACNA1A mutation causing episodic ataxia type 2.
Arch Neurol 58: 292-295, 2001.
9. Deutsch C. Potassium channel ontogeny. Annu Rev Physiol 64: 19-46, 2002.
10. Ellgaard L, Molinari M, and Helenius A. Setting the standards: quality control
in the secretory pathway. Science 286: 1882-1888, 1999.
11. Ellinor PT, Yang J, Sather WA, Zhang JF, and Tsien RW. Ca2+ channel
selectivity at a single locus for high-affinity Ca2+ interactions. Neuron 15: 1121-1132,
1995.
12. Ficker E, Dennis AT, Obejero-Paz CA, Castaldo P, Taglialatela M, and
Brown AM. Retention in the endoplasmic reticulum as a mechanism of
dominant-negative current suppression in human long QT syndrome. J Mol Cell
Cardiol 32: 2327-2337, 2000.
13. Fletcher CF, Tottene A, Lennon VA, Wilson SM, Dubel SJ, Paylor R,
Hosford DA, Tessarollo L, McEnery MW, Pietrobon D, Copeland NG, and
Jenkins NA. Dystonia and cerebellar atrophy in Cacna1a null mice lacking P/Q
37
calcium channel activity. Faseb J 15: 1288-1290, 2001.
14. Guida S, Trettel F, Pagnutti S, Mantuano E, Tottene A, Veneziano L, Fellin T,
Spadaro M, Stauderman K, Williams M, Volsen S, Ophoff R, Frants R, Jodice C,
Frontali M, and Pietrobon D. Complete loss of P/Q calcium channel activity caused
by a CACNA1A missense mutation carried by patients with episodic ataxia type 2.
Am J Hum Genet 68: 759-764, 2001.
15. Hans M, Urrutia A, Deal C, Brust PF, Stauderman K, Ellis SB, Harpold
MM, Johnson EC, and Williams ME. Structural elements in domain IV that
influence biophysical and pharmacological properties of human alpha1A-containing
high-voltage-activated calcium channels. Biophys J 76: 1384-1400, 1999.
16. Hansel C, Linden DJ, and D'Angelo E. Beyond parallel fiber LTD: the
diversity of synaptic and non-synaptic plasticity in the cerebellum. Nat Neurosci 4:
467-475, 2001.
17. Harding HP, Calfon M, Urano F, Novoa I, and Ron D. Transcriptional and
translational control in the Mammalian unfolded protein response. Annu Rev Cell Dev
Biol 18: 575-599, 2002.
18. Imbrici P, Jaffe SL, Eunson LH, Davies NP, Herd C, Robertson R,
Kullmann DM, and Hanna MG. Dysfunction of the brain calcium channel CaV2.1
in absence epilepsy and episodic ataxia. Brain 127: 2682-2692, 2004.
19. Ito M. Cerebellar long-term depression: characterization, signal transduction,
and functional roles. Physiol Rev 81: 1143-1195, 2001.
20. Ito M. The molecular organization of cerebellar long-term depression. Nat Rev
Neurosci 3: 896-902, 2002.
21. Jen J, Kim GW, and Baloh RW. Clinical spectrum of episodic ataxia type 2.
Neurology 62: 17-22, 2004.
22. Jen J, Wan J, Graves M, Yu H, Mock AF, Coulin CJ, Kim G, Yue Q,
Papazian DM, and Baloh RW. Loss-of-function EA2 mutations are associated with
impaired neuromuscular transmission. Neurology 57: 1843-1848, 2001.
23. Jen J, Yue Q, Nelson SF, Yu H, Litt M, Nutt J, and Baloh RW. A novel
nonsense mutation in CACNA1A causes episodic ataxia and hemiplegia. Neurology
53: 34-37, 1999.
24. Jen JC and Baloh RW. Genetics of episodic ataxia. Adv Neurol 89: 459-461,
2002.
25. Jouvenceau A, Eunson LH, Spauschus A, Ramesh V, Zuberi SM, Kullmann
DM, and Hanna MG. Human epilepsy associated with dysfunction of the brain
P/Q-type calcium channel. Lancet 358: 801-807, 2001.
26. Jun K, Piedras-Renteria ES, Smith SM, Wheeler DB, Lee SB, Lee TG, Chin
H, Adams ME, Scheller RH, Tsien RW, and Shin HS. Ablation of P/Q-type Ca(2+)
38
channel currents, altered synaptic transmission, and progressive ataxia in mice lacking
the alpha(1A)-subunit. Proc Natl Acad Sci U S A 96: 15245-15250, 1999.
27. Kagan A, Yu Z, Fishman GI, and McDonald TV. The dominant negative
LQT2 mutation A561V reduces wild-type HERG expression. J Biol Chem 275:
11241-11248, 2000.
28. Kulik A, Nakadate K, Hagiwara A, Fukazawa Y, Lujan R, Saito H, Suzuki N,
Futatsugi A, Mikoshiba K, Frotscher M, and Shigemoto R. Immunocytochemical
localization of the alpha 1A subunit of the P/Q-type calcium channel in the rat
cerebellum. Eur J Neurosci 19: 2169-2178, 2004.
29. Liman ER, Tytgat J, and Hess P. Subunit stoichiometry of a mammalian K+
channel determined by construction of multimeric cDNAs. Neuron 9: 861-871, 1992.
30. Llinas R, Sugimori M, Lin JW, and Cherksey B. Blocking and isolation of a
calcium channel from neurons in mammals and cephalopods utilizing a toxin fraction
(FTX) from funnel-web spider poison. Proc Natl Acad Sci U S A 86: 1689-1693,
1989.
31. Lux A, Gallione CJ, and Marchuk DA. Expression analysis of endoglin
missense and truncation mutations: insights into protein structure and disease
mechanisms. Hum Mol Genet 9: 745-755, 2000.
32. Ma D and Jan LY. ER transport signals and trafficking of potassium channels
and receptors. Curr Opin Neurobiol 12: 287-292, 2002.
33. Maquat LE. Nonsense-mediated mRNA decay: splicing, translation and mRNP
dynamics. Nat Rev Mol Cell Biol 5: 89-99, 2004.
34. Mintz IM, Sabatini BL, and Regehr WG. Calcium control of transmitter
release at a cerebellar synapse. Neuron 15: 675-688, 1995.
35. Miyazaki T, Hashimoto K, Shin HS, Kano M, and Watanabe M. P/Q-type
Ca2+ channel alpha1A regulates synaptic competition on developing cerebellar
Purkinje cells. J Neurosci 24: 1734-1743, 2004.
36. Ophoff RA, Terwindt GM, Vergouwe MN, van Eijk R, Oefner PJ, Hoffman
SM, Lamerdin JE, Mohrenweiser HW, Bulman DE, Ferrari M, Haan J,
Lindhout D, van Ommen GJ, Hofker MH, Ferrari MD, and Frants RR. Familial
hemiplegic migraine and episodic ataxia type-2 are caused by mutations in the Ca2+
channel gene CACNL1A4. Cell 87: 543-552, 1996.
37. Page KM, Heblich F, Davies A, Butcher AJ, Leroy J, Bertaso F, Pratt WS,
and Dolphin AC. Dominant-negative calcium channel suppression by truncated
constructs involves a kinase implicated in the unfolded protein response. J Neurosci
24: 5400-5409, 2004.
38. Raghib A, Bertaso F, Davies A, Page KM, Meir A, Bogdanov Y, and Dolphin
AC. Dominant-negative synthesis suppression of voltage-gated calcium channel
39
Cav2.2 induced by truncated constructs. J Neurosci 21: 8495-8504, 2001.
39. Raymond JL and Lisberger SG. Hypotheses about the neural trigger for
plasticity in the circuit for the vestibulo-ocular reflex. Prog Brain Res 124: 235-246,
2000.
40. Rhyu IJ, Abbott LC, Walker DB, and Sotelo C. An ultrastructural study of
granule cell/Purkinje cell synapses in tottering (tg/tg), leaner (tg(la)/tg(la)) and
compound heterozygous tottering/leaner (tg/tg(la)) mice. Neuroscience 90: 717-728,
1999.
41. Sabatini BL and Regehr WG. Control of neurotransmitter release by
presynaptic waveform at the granule cell to Purkinje cell synapse. J Neurosci 17:
3425-3435, 1997.
42. Seidman JG and Seidman C. Transcription factor haploinsufficiency: when
half a loaf is not enough. J Clin Invest 109: 451-455, 2002.
43. Shovlin CL, Hughes JM, Scott J, Seidman CE, and Seidman JG.
Characterization of endoglin and identification of novel mutations in hereditary
hemorrhagic telangiectasia. Am J Hum Genet 61: 68-79, 1997.
44. Soong TW, DeMaria CD, Alvania RS, Zweifel LS, Liang MC, Mittman S,
Agnew WS, and Yue DT. Systematic identification of splice variants in human
P/Q-type channel alpha1(2.1) subunits: implications for current density and
Ca2+-dependent inactivation. J Neurosci 22: 10142-10152, 2002.
45. Spacey SD, Hildebrand ME, Materek LA, Bird TD, and Snutch TP.
Functional implications of a novel EA2 mutation in the P/Q-type calcium channel.
Ann Neurol 56: 213-220, 2004.
46. Swensen AM and Bean BP. Ionic mechanisms of burst firing in dissociated
Purkinje neurons. J Neurosci 23: 9650-9663, 2003.
47. Takahashi T and Momiyama A. Different types of calcium channels mediate
central synaptic transmission. Nature 366: 156-158, 1993.
48. Tang CY and Papazian DM. Transfer of voltage independence from a rat
olfactory channel to the Drosophila ether-a-go-go K+ channel. J Gen Physiol 109:
301-311, 1997.
49. Tank DW, Sugimori M, Connor JA, and Llinas RR. Spatially resolved
calcium dynamics of mammalian Purkinje cells in cerebellar slice. Science 242:
773-777, 1988.
50. Trimmer JS. Regulation of ion channel expression by cytoplasmic subunits.
Curr Opin Neurobiol 8: 370-374, 1998.
51. Uchitel OD, Protti DA, Sanchez V, Cherksey BD, Sugimori M, and Llinas R.
P-type voltage-dependent calcium channel mediates presynaptic calcium influx and
transmitter release in mammalian synapses. Proc Natl Acad Sci U S A 89: 3330-3333,
40
1992.
52. Urbano FJ, Piedras-Renteria ES, Jun K, Shin HS, Uchitel OD, and Tsien
RW. Altered properties of quantal neurotransmitter release at endplates of mice
lacking P/Q-type Ca2+ channels. Proc Natl Acad Sci U S A 100: 3491-3496, 2003.
53. Usowicz MM, Sugimori M, Cherksey B, and Llinas R. P-type calcium
channels in the somata and dendrites of adult cerebellar Purkinje cells. Neuron 9:
1185-1199, 1992.
54. Veitia RA. Exploring the etiology of haploinsufficiency. Bioessays 24: 175-184,
2002.
55. Walikonis RS, Oguni A, Khorosheva EM, Jeng CJ, Asuncion FJ, and
Kennedy MB. Densin-180 forms a ternary complex with the (alpha)-subunit of
Ca2+/calmodulin-dependent protein kinase II and (alpha)-actinin. J Neurosci 21:
423-433, 2001.
56. Wappl E, Koschak A, Poteser M, Sinnegger MJ, Walter D, Eberhart A,
Groschner K, Glossmann H, Kraus RL, Grabner M, and Striessnig J. Functional
consequences of P/Q-type Ca2+ channel Cav2.1 missense mutations associated with
episodic ataxia type 2 and progressive ataxia. J Biol Chem 277: 6960-6966, 2002.
57. Westenbroek RE, Sakurai T, Elliott EM, Hell JW, Starr TV, Snutch TP, and
Catterall WA. Immunochemical identification and subcellular distribution of the
alpha 1A subunits of brain calcium channels. J Neurosci 15: 6403-6418, 1995.
58. White MM and Aylwin M. Niflumic and flufenamic acids are potent reversible
blockers of Ca2(+)-activated Cl- channels in Xenopus oocytes. Mol Pharmacol 37:
720-724, 1990.
59. Womack MD and Khodakhah K. Dendritic control of spontaneous bursting in
cerebellar Purkinje cells. J Neurosci 24: 3511-3521, 2004.
60. Wu LG, Westenbroek RE, Borst JG, Catterall WA, and Sakmann B. Calcium
channel types with distinct presynaptic localization couple differentially to transmitter
release in single calyx-type synapses. J Neurosci 19: 726-736, 1999.
61. Yang J, Ellinor PT, Sather WA, Zhang JF, and Tsien RW. Molecular
determinants of Ca2+ selectivity and ion permeation in L-type Ca2+ channels. Nature
366: 158-161, 1993.
62. Yue Q, Jen JC, Thwe MM, Nelson SF, and Baloh RW. De novo mutation in
CACNA1A caused acetazolamide-responsive episodic ataxia. Am J Med Genet 77:
298-301, 1998.
63. Zhou Z, Gong Q, Epstein ML, and January CT. HERG channel dysfunction
in human long QT syndrome. Intracellular transport and functional defects. J Biol
Chem 273: 21061-21066, 1998.
64. Zhuchenko O, Bailey J, Bonnen P, Ashizawa T, Stockton DW, Amos C,
41
Dobyns WB, Subramony SH, Zoghbi HY, and Lee CC. Autosomal dominant
cerebellar ataxia (SCA6) associated with small polyglutamine expansions in the alpha
1A-voltage-dependent calcium channel. Nat Genet 15: 62-69, 1997.
42
Table 1. The steady-state voltage-dependence property of wild-type CaV2.1
channels in the absence or presence of equal-molar coexpressions of α1A mutants.
The conductance-voltage (G-V) curve of each oocyte expressing CaV2.1 currents was
fitted with a simple Boltzmann equation: 1 / {1 + exp[(V0.5 - V) / k ]}, where V0.5 is
the half-maximal voltage for activation, and k the slope factor of the G-V curve. The
number of oocytes analyzed for each coexpression paradigm is shown in the
parentheses.
V0.5
k
(28)
3.5 ± 1.9
4.5 ± 0.3
WT+R1281x (17)
5.8 ± 1.7
4.6 ± 0.5
WT+F1406C (12)
4.1 ± 1.1
4.1 ± 0.4
WT+R1549x (15)
3.9 ± 1.6
4.9 ± 0.7
WT+R1669x (13)
2.8 ± 0.9
5.1 ± 0.3
WT+E1761K (16)
3.9 ± 1.3
4.7 ± 0.9
WT
43
FIGURE LEGENDS
Figure 1. Expression of wild-type and mutant human CaV2.1 channels in
Xenopus oocytes.
(A) The schematic location of the five EA2 mutations under investigation within the
α1A subunit. Single-lettered notation is used here to denote the amino acid at the
indicated position. A premature stop codon is represented by the letter X. (B)
Representative barium currents recorded from Xenopus oocytes expressing wild-type
(WT), F1406C, or E1761K CaV2.1 channels. The holding potential was –90 mV. The
pulse protocol comprised 70-ms depolarizing test pulses ranging from –60 to +20 mV,
with 10-mV increments. (C) The distribution of the peak current amplitudes at +20
mV measured from oocytes injected with water (H2O-injected) as well as those
expressing various constructs of CaV2.1 channels. The numbers in the parentheses
refer for to the number of oocytes recorded for each construct. The mean peak current
amplitude at +20 mV for WT, R1281x, F1406C, R1549x, R1669x, E1761K, and
H2O-injected oocytes was (mean±SEM, in μA) 1.04±0.60, 0.04±0.03, 0.12±0.09,
0.02±0.02, 0.04±0.02, 0.03±0.03, and 0.03±0.03, respectively.
44
Figure 2. Equal-molar coexpression with α1A mutants significantly decreased the
current amplitudes of wild-type CaV2.1 channels in Xenopus oocytes.
(A) Representative barium currents recorded from Xenopus oocytes subjected to
different cRNA coinjection paradigms. Wild-type (WT) CaV2.1 channels correspond
to oocytes injected with the standard cRNA cocktail, in which α1A WT, α2δ, and β4
cRNAs were mixed in a molar ratio of 1:2:2 (see text and the method section for more
details). For coexpression experiments, one copy of the cRNA for an α1A mutant was
added to the standard cRNA cocktail for oocyte injections, resulting in 1:1 molar ratio
of α1A WT and α1A mutant. The holding potential was –90 mV. The pulse protocol
comprised 70-ms depolarizing test pulses ranging from –60 to +50 mV, with 10-mV
increments. (B) Normalized mean barium current amplitudes of wild-type CaV2.1
channels in the absence or presence of α1A mutants. The peak +20-mV current
amplitude measured from each Xenopus oocyte coexpressing α1A WT and α1A mutant
was normalized by the corresponding mean +20-mV barium current amplitude of
wild-type CaV2.1 channels measured from the same batch of oocytes on the same day
of experiment; i.e., the mean +20-mV barium current amplitude of wild-type CaV2.1
channels on any given day of experiment was always normalized as unity. Normalized
data from different batches of Xenopus oocytes or different days of experiments were
later pooled together for comprehensive analyses. In the presence of nonsense or
45
missense EA2 mutants, the current amplitude of wild-type CaV2.1 channels became
significantly smaller (*, t test: p < 0.05). The number of oocytes measured for each
coexpression paradigm is shown in the parentheses.
Figure 3. The voltage activation curves of wild-type CaV2.1 channels were not
significantly affected by the presence of α1A mutants.
Macroscopic channel conductance (G) at each test potential (V) was calculated by the
equation: G = I / (V - Vrev), where I is the peak barium current amplitudes measured at
each test voltage, and Vrev the apparent reversal potentials determined by
extrapolating the I-V curve to the voltage axis. For each oocyte, the channel
conductance at each test potential was normalized by the corresponding peak
conductance.
Figure 4. The effects of enhanced coexpression of R1281x mutant on the current
amplitudes of wild-type CaV2.1 channels in Xenopus oocytes.
(A) Representative barium currents recorded from Xenopus oocytes coexpressing
different molar ratios of R1281x mutant. To achieve the indicated molar ratios of α1A
WT and α1A mutant, appropriate amount of the cRNA for R1281x was added to the
standard cRNA cocktail for oocyte injections. The holding potential was –90 mV. The
46
pulse protocol comprised 70-ms depolarizing test pulses ranging from –60 to +50 mV,
with 10-mV increments. (B) Normalized mean barium current amplitudes of
wild-type CaV2.1 channels in the presence of different molar ratios of R1281x. Data
were normalized by the mean +20-mV barium current amplitude of wild-type CaV2.1
channels in the absence of R1281x (1:0), using the same procedure as described in
Figure 2B. The mean current amplitudes of coexpressing α1A WT and R1281x in the
molar ratios 1:1, 1:3, 1:5, and 1:10 were significantly different (One-way ANOVA: p
< 0.05). Furthermore, when individually compared with the coexpression ratio 1:1,
the mean current amplitude for the molar ratio 1:3, 1:5, or 1:10 was significantly
smaller (*, t test: p < 0.05), indicating that the amplitude of barium currents decreased
as the relative molar ratio of R1281x increased. The number of oocytes measured for
each coexpression paradigm is shown in the parentheses. (C) Normalized I-V curves
of wild-type CaV2.1 channels in the presence of different molar ratios of mutation
R1281x. The peak barium current amplitude measured at each test voltage was
normalized by the mean current amplitude at +20 mV of wild-type CaV2.1 channels
(WT). Increased molar coexpression of R1281x failed to significantly shift the
position of the I-V curves of CaV2.1 channels along the voltage axis.
Figure 5. The effects of enhanced coexpression of missense and nonsense EA2
47
mutants on the current amplitudes of wild-type CaV2.1 channels in Xenopus
oocytes.
Normalized mean barium current amplitudes at +20 mV of wild-type CaV2.1 channels
in the presence of different molar ratios of mutants F1406C, R1549x, R1669x, and
E1761K. The mean current amplitudes of coexpressing α1A WT and α1A mutants in
the molar ratios 1:1, 1:3, 1:5, and 1:10 were significantly different (One-way ANOVA:
p < 0.05). For all four EA2 mutants, the mean current amplitude for the coexpression
ratio 1:3, 1:5, or 1:10 was significantly smaller than that for the molar ratio 1:1 (*, t
test: p < 0.05), demonstrating that the amplitude of barium currents decreased as the
relative molar ratio of EA2 mutants increased. The number of oocytes measured for
each coexpression paradigm is shown in the parentheses.
Figure 6. Densin-180 and rolf failed to display dominant-negative effects on
wild-type CaV2.1 channels in Xenopus oocytes.
Normalized mean barium current amplitudes at +20 mV of CaV2.1 channels in the
presence of different molar ratios of densin-180 (A) and rat olfactory channel (rolf)
(B). The mean amplitudes of barium currents through CaV2.1 channels in the absence
(1:0) or presence (1:1, 1:3, 1:5, 1:10) of densin-180 / rolf were not significantly
different (One-way ANOVA: p > 0.05). The number of oocytes measured for each
48
coexpression paradigm is shown in the parentheses.
Figure 7. Increased amount of α1A WT cRNA injection did not reduce the mean
current amplitudes of Cav2.1 channels.
(A) Normalized mean barium current amplitudes recorded from Xenopus oocytes
expressing different concentrations of α1A WT cRNA. Here, the low cRNA cocktail,
in which the amount of cRNA for α1A, α2δ, and β4 subunits was only one sixth of that
for the standard cRNA cocktail, was used as the control group (1:0) for oocyte
injections. Appropriate amount of additional α1A WT cRNA was then added to the
low cRNA cocktail to mimic the miscellaneous coexpression paradigms of α1A WT
and α1A mutants described in Figures 4 and 5. The mean amplitudes of barium
currents recorded from Xenopus oocytes expressing different concentrations of α1A
WT cRNA (1:0, 1:1, 1:3, 1:5, and 1:10) were significantly different (One-way
ANOVA: p < 0.05). In particular, enhancing the amount of α1A WT cRNA by 6 fold
(1:5) or 11 fold (1:10) resulted in mean CaV2.1 current amplitudes that were
significantly larger (*, t test: p < 0.05) than that in the control group (1:0). Data were
normalized by the mean +20-mV barium current amplitude of wild-type CaV2.1
channels (1:0), using the same procedure as described in Figure 2B. The number of
oocytes measured for each coexpression paradigm is shown in the parenthesis. (B)
49
Equal-molar coexpressions of EA2 mutants with wild-type CaV2.1 channels using the
low cRNA cocktail. The mean barium current amplitudes in the presence of EA2
mutants were significantly smaller (*, t test: p < 0.05) than that of expressing
wild-type CaV2.1 channels alone. The number of oocytes measured for each
coexpression paradigm is shown in the parentheses.
Figure 8. Effects of enhancing the molar ratio of auxiliary subunits on the
dominant-negative effects of EA2 mutants
(A) Increasing the amount of cRNAs for auxiliary subunits partially reversed the
suppression effects of nonsense EA2 mutants. Data were normalized by the mean
+20-mV barium current amplitude of wild-type CaV2.1 channels using the standard
cRNA cocktail (1:0:2:2), in which cRNAs for α1A WT, α2δ, and β4 subunits were
mixed in a molar ratio of 1:2:2, prior to injection into oocytes. The mean current
amplitudes of equal-molar coexpression of nonsense EA2 mutants using the standard
(1:1:2:2), double (1:1:4:4), and quadruple (1:1:8:8) auxiliary concentrations were
significantly different (One-way ANOVA: p < 0.05). Asterisks denote a significant
increase (t test: p < 0.05) in the current amplitude relative to that for the standard
auxiliary concentration (1:1:2:2).
(B) Increasing the amount of cRNAs for
auxiliary subunits did not affect the suppression effects of missense EA2 mutants. The
50
mean current amplitudes for the standard (1:1:2:2), double (1:1:4:4), and quadruple
(1:1:8:8) auxiliary concentrations were not significantly different (One-way ANOVA:
p > 0.05). The number of oocytes measured for each coexpression paradigm is shown
in the parentheses.
Figure 9. The dominant-negative effect of R1281x is considerably diminished for
the short α1A subunit isoform.
The short isoform of α1A subunits were employed for both α1A WT and R1281x.
Mean barium current amplitudes of wild-type CaV2.1 channels in the presence of
different molar ratios of R1281x were normalized by that for wild-type CaV2.1
channels in the absence of the mutant (1:0), using the same procedure as described in
Figure 2B. The mean current amplitudes of coexpressing α1A WT and R1281x in the
molar ratios 1:0, 1:1, 1:3, and 1:5 were not significantly different (One-way ANOVA:
p > 0.05). In contrast, the mean current amplitude of coexpressing α1A WT and
R1281x in the molar ratio of 1:10 was significantly smaller (*, t test: p < 0.05) than
that
for
expressing
α1A
WT
alone
(1:0),
indicating
that
a
significant
dominant-negative effect is only present if the relative molar ratio of R1281 cRNA
was notably high. As a negative control, also shown was the mean current amplitude
for the coexpression of α1A WT and rolf in the ratio of 1:10. The number of oocytes
51
measured for each coexpression paradigm is shown in the parentheses.
52
Fig. 1
Jeng et al.
53
Fig. 2
Jeng et al.
54
Fig. 3
Jeng et al.
55
Fig. 4
Jeng et al.
56
Fig. 5
Jeng et al.
57
Fig. 6
Jeng et al.
58
Fig. 7
Jeng et al.
59
Fig. 8
Jeng et al.
60
Fig. 9
Jeng et al.
61