Junctional Cleft [Ca]i Measurements using Novel Cleft-Targeted Ca Sensors

Cardiac SR Ca release is controlled locally by [Ca]i in the junctional cleft. Despite its importance for excitation-contraction coupling (ECC), there is no method for direct cleft [Ca]i measurements. We constructed novel cleft-targeted [Ca]i sensors by attaching FKBP12.6 to the genetically-encoded Ca sensor GCaMP2. FKBPs12.6 binds with high affinity and specificity to ryanodine receptors (RyRs) and targets the sensor to the junctions. We also constructed a FKBP-GCaMP2 variant with low Ca affinity (Kd≈10 μM; FKBP-GCaMP-Low). Targeted or untargeted GCaMP2 sensors were expressed in rat cardiac myocytes by adenoviral infection. Both FKBP-GCaMP2s express in a striated pattern, spaced ∼2 μm apart (at T-tubules), while the un-tagged GCaMP2s are more uniformly distributed. Ca transients recorded with FKBP-GCaMP-Low are ∼2-fold larger and have a much faster upstroke (time-to-peak=46±5 vs. 90±7 ms) and decay (decay time=254±19 vs. 421±42 ms at 0.5 Hz) compared to those reported by un-targeted GCaMP-Low. Thus, we can detect a [Ca]Cleft transient with much larger amplitude and faster kinetics than global [Ca]Bulk transient during ECC. During diastole SR Ca leak (or influx) may raise local [Ca]Cleft above [Ca]Bulk (distant from Ca sources). We used high-affinity FKBP-GCaMP and GCaMP to compare diastolic [Ca]Cleft to [Ca]Bulk. Blockade of SR Ca leak and sarcolemmal Ca influx (with 1 mM tetracaine & 0Na/0Ca solution) should equalize [Ca]Cleft to [Ca]Bulk. Thus, a larger fluorescence decline (ΔF/F0) in [Ca]Cleft vs. [Ca]Bulk upon block provides a measure of this diastolic [Ca]i gradient. Indeed, ΔF/F0 was significantly larger for FKBP-GCaMP2 vs. GCaMP2 (0.55±0.10 vs. 0.13±0.02), indicating that [Ca]Cleft > [Ca]Bulk during diastole. In conclusion, we developed sensors for measuring [Ca]Cleft vs. [Ca]Bulk and provided measures of a standing [Ca]i gradient during diastole and dynamic differences during ECC.