Functional impact of alternative splicing of human T-type Cav3.3 calcium channels.

Low-voltage-activated T-type (Cav3) Ca2+ channels produce low-threshold spikes that trigger burst firing in many neurons. The CACNA1I gene encodes the Cav3.3 isoform, which activates and inactivates much more slowly than the other Cav3 channels. These distinctive kinetic features, along with its brain-region-specific expression, suggest that Cav3.3 channels endow neurons with the ability to generate long-lasting bursts of firing. The human CACNA1I gene contains two regions of alternative splicing: variable inclusion of exon 9 and an alternative acceptor site within exon 33, which leads to deletion of 13 amino acids (Delta33). The goal of this study is to determine the functional consequences of these variations in the full-length channel. The cDNA encoding these regions were cloned using RT-PCR from human brain, and currents were recorded by whole cell patch clamp. Introduction of the Delta33 deletion slowed the rate of channel opening. Addition of exon 9 had little effect on kinetics, whereas its addition to Delta33 channels unexpectedly slowed both activation and inactivation kinetics. Modeling of neuronal firing showed that exon 9 or Delta33 alone reduced burst firing, whereas the combination enhanced firing. The major conclusions of this study are that the intracellular regions after repeats I and IV play a role in channel gating, that their effects are interdependent, suggesting a direct interaction, and that splice variation of Cav3.3 channels provides a mechanism for fine-tuning the latency and duration of low-threshold spikes.

Cloning and expression of the human T-type channel Ca(v)3.3: insights into prepulse facilitation.

The full-length human Ca(v)3.3 (alpha(1I)) T-type channel was cloned, and found to be longer than previously reported. Comparison of the cDNA sequence to the human genomic sequence indicates the presence of an additional 4-kb exon that adds 214 amino acids to the carboxyl terminus and encodes the 3' untranslated region. The electrophysiological properties of the full-length channel were studied after transient transfection into 293 human embryonic kidney cells using 5 mM Ca(2+) as charge carrier. From a holding potential of -100 mV, step depolarizations elicited inward currents with an apparent threshold of -70 mV, a peak of -30 mV, and reversed at +40 mV. The kinetics of channel activation, inactivation, deactivation, and recovery from inactivation were very similar to those reported previously for rat Ca(v)3.3. Similar voltage-dependent gating and kinetics were found for truncated versions of human Ca(v)3.3, which lack either 118 or 288 of the 490 amino acids that compose the carboxyl terminus. A major difference between these constructs was that the full-length isoform generated twofold more current. These results suggest that sequences in the distal portion of Ca(v)3.3 play a role in channel expression. Studies on the voltage-dependence of activation revealed that a fraction of channels did not gate as low voltage-activated channels, requiring stronger depolarizations to open. A strong depolarizing prepulse (+100 mV, 200 ms) increased the fraction of channels that gated at low voltages. In contrast, human Ca(v)3.3 isoforms with shorter carboxyl termini were less affected by a prepulse. Therefore, Ca(v)3.3 is similar to high voltage-activated Ca(2+) channels in that depolarizing prepulses can regulate their activity, and their carboxy termini play a role in modulating channel activity.