Mid-channel proteolysis of the L-type voltage gated calcium channel and the potential role of amyloid-beta precursor protein

L-type voltage-gated calcium channels are involved in many important physiological processes, including muscle contraction, hormone secretion and neuronal gene expression. These channels are regulated by many different mechanisms to tightly control calcium influx. Our lab has uncovered a new form of L-type channel regulation that involves the proteolysis of the channel in the main body of the alpha1 subunit in response to increased intracellular calcium, channel activity and age. I investigated the immediate and long-term functional impact of mid-channel proteolysis on the CaV1.2 channel. Mid-channel proteolysis causes an acute change in gating and a decrease in channel activity over a longer time scale. Fragment channels result from proteolysis, and these fragments associate on the plasma membrane to form functional channels. These L-type fragment channels exhibit different biophysical properties than full-length CaV1.2. While fragment channels must combine so that all four domains are present to be functional, non-complimentary pairs containing more than four domains still produce discernible current. L-type fragment channels co-immunoprecipitate with the full-length CaV1.2, indicating that fragments bind to either the alpha1 subunit or the channel complex. Some of these fragments cause a shift in inactivation and in the I--V curve of the channel, and one fragment comprising Domain IV and the C-terminus (fragment C2) inhibits full-length CaV1.2 in a dominant negative manner. These results demonstrate the functional effects of mid-channel proteolysis.;L-type mid-channel proteolysis increases with animal age. Therefore, to identify the protease responsible for mid-channel proteolysis, I turned to proteases involved in aging diseases. Amyloid-beta precursor protein (APP), a protein implicated in Alzheimer's disease (AD), modulates L-type channels and is itself extensively proteolyzed. One of those proteases is presenilin, the catalytic component to gamma-secretase. I found that APP dramatically reduced human CaV1.2 current in Xenopus oocytes. The current-voltage relationship and inactivation profiles of Ca V1.2 in the presence of APP mirrored those of the fragment channels. Moreover, a gamma-secretase inhibitor, DAPT, completely reversed this effect. When an AD APP mutant was co-expressed with CaV1.2, currents were further diminished. Astonishingly, an APP mutant that protects against AD had the opposite effect, allowing larger CaV1.2 currents than wild-type APP.;Western blots stained with an antibody against CaV1.2 revealed a ∼100 kD band when APP was coexpressed with the channel, which was absent in oocytes solely expressing CaV1.2. DAPT application reversed this effect, indicating the band was a product of presenilin proteolysis. A putative cut site was found on the alpha1 subunit that would produce a band similar in size to the one observed in Western blots. When this site was mutated, the ∼100 kD band no longer appeared when Ca V1.2 was coexpressed with APP. Unfortunately, the CaV1.2 II--III loop antibody was later found to cross-react with APP. Therefore, additional experiments are necessary to determine whether the ∼100 kD band is CaV1.2 Interestingly, APP induced mid-channel proteolysis was detected in primary neurons using imaging techniques. While the mechanism for APP-induced inhibition of the channel is still unresolved, my data clearly shows this effect is mediated by presenilin. Whether or not presenilin is responsible for cutting CaV1.2 remains to be resolved.