Differential control of calcium homeostasis and vascular reactivity by Ca2+/calmodulin-dependent kinase II

Abstract

The multifunctional Ca(2+)/calmodulin-dependent kinase II (CaMKII) is activated by vasoconstrictors in vascular smooth muscle cells (VSMC), but its impact on vasoconstriction remains unknown. We hypothesized that CaMKII inhibition in VSMC decreases vasoconstriction. Using novel transgenic mice that express the inhibitor peptide CaMKIIN in smooth muscle (TG SM-CaMKIIN), we investigated the effect of CaMKII inhibition on L-type Ca(2+) channel current (ICa), cytoplasmic and sarcoplasmic reticulum Ca(2+), and vasoconstriction in mesenteric arteries. In mesenteric VSMC, CaMKII inhibition significantly reduced action potential duration and the residual ICa 50 ms after peak amplitude, indicative of loss of L-type Ca(2+) channel-dependent ICa facilitation. Treatment with angiotensin II or phenylephrine increased the intracellular Ca(2+) concentration in wild-type but not TG SM-CaMKIIN VSMC. The difference in intracellular Ca(2+) concentration was abolished by pretreatment with nifedipine, an L-type Ca(2+) channel antagonist. In TG SM-CaMKIIN VSMC, the total sarcoplasmic reticulum Ca(2+) content was reduced as a result of diminished sarcoplasmic reticulum Ca(2+) ATPase activity via impaired derepression of the sarcoplasmic reticulum Ca(2+) ATPase inhibitor phospholamban. Despite the differences in intracellular Ca(2+) concentration, CaMKII inhibition did not alter myogenic tone or vasoconstriction of mesenteric arteries in response to KCl, angiotensin II, and phenylephrine. However, it increased myosin light chain kinase activity. These data suggest that CaMKII activity maintains intracellular calcium homeostasis but is not required for vasoconstriction of mesenteric arteries.

Keywords: Ca2+/calmodulin-dependent protein kinase type 2; L-type Ca2+ channel; calcium signaling; myometrial contraction.

Figure 1. CaMKII inhibition reduces LTCC activity in VSMC

(A) Representative examples of stimulated action potentials in WT and TG SM-CaMKIIN VSMC. (B) Action potential duration (APD) measured at 90% reduction in current from the upstroke (n=5 each). (C, D) Current-density/voltage relationships for whole-cell I_Ca recorded in VSMC from WT (n=7), TG SM-CaMKIIN (n=9) in 10 mM BaCl₂. Average data for peak I_Ca density in response to a voltage command pulse to 0 mV. (E, F) Residual I_Ca 50 ms after peak amplitude (R50) (n=10 for each genotype). (G, H) I_Ca at baseline (black) and after 5 repetitive depolarizations (red). Repetitive depolarizations were induced by depolarizing to +10 mV at −70 holding potential with 0.5 Hz (n=6 in WT, n=7 in TG SM-CaMKIIN). (I) Summary data for relative peak I_Ca in response to a train of 10 voltage command pulses. (J) Relative peak current I_Ca histogram of the 5^th command pulse normalized to I of the 1^st Ca command pulse. (K) Representative phosphorylated (p-Cavβ3) and total β3 subunit (Cavβ3) and total Cav1.2α1c immunoblots from mesenteric artery lysates. Lysates from 5-7 mice were pooled per lane. (L) Summary data for Cavβ3 phosphorylation, normalized to total Cavβ2 and for Cav1.2α1c normalized to β-actin (n=3) * p<0.05 compared to WT. WT (black lines or bars), TG SM-CaMKIIN (grey).

Figure 2. CaMKII inhibition reduces agonist-induced increase in [Ca²⁺]_i

(A) [Ca²⁺]_i in mesenteric VSMC after 100 nM Ang-II (arrow) was measured by fluorescence emission using Fura-2 at 340 or 380 nm excitation. Each trace represents the average of transients from 10 to 15 cells from 4–5 different preparations. (B) [Ca²⁺]_i plotted as % increase over baseline (n=10-15). Nif, nifedipine. (C) [Ca²⁺]_i after 1μM Bay K8644 (n=10). (D) [Ca²⁺]_i after 10 μM phenylephrine (PE). (n=10). * p<0.05 compared to WT, # p<0.05 to Ca²⁺ treatment. WT (black lines or bars), TG SM-CaMKIIN (grey).

Figure 3. CaMKII inhibition reduces SR Ca²⁺ content, phospholamban phosphorylation, and SERCA2 activity

(A) SR Ca²⁺ content after addition of 10 mM caffeine and 10 μM thapsigargin (arrow) by Fura-2 fluorescent imaging. Each trace represents the average of transients in 10 to 15 cells from 4-5 different preparations. (B) Fold change in SR Ca²⁺ content (n=10). (C) Representative phospho-phospholamban Thr17 (p-PLB) and total phospholamban (PLB) immunoblots in mesenteric artery lysates. In all immunoblots, lysates from 5-7 mice were pooled per lane. (D) Quantitation of (C) (n=5). (E) SERCA2 activity assays, assessed by inorganic phosphate release (Pi), in aortic lysates (n=3 independent experiments with 3 samples per experiment). (F) qrtPCR for SERCA2 mRNA. (G) Representative SERCA2 immunoblots in mesenteric artery lysates. (H) Quantitation of (G) (n=5). In all immunoblots, lysates from 5-7 mice were pooled per lane. * p<0.05 compared to WT. WT (black lines or bars), TG SM-CaMKIIN (grey).

Figure 4. CaMKII inhibition does not decrease agonist-stimulated vasoconstriction or myogenic tone

(A-F) Vasoconstriction of second-order mesenteric arteries to (A, B) KCl, (C) Ang-II, (D) phenylephrine (PE), (E) 0.1 μM Bay K8644, and (F) 10 mM caffeine. (n=5-15 mice per group) (G) Vasoconstriction to PE or (H) Ang-II after pretreatment with 0.1 μM nifedipine. Data are expressed as the difference compared to baseline. (n=6 mice per group). (I) Myogenic tone (%) in denuded mesenteric vessels (n=7). * p<0.05 compared to WT. WT (black lines or bars), TG SM-CaMKIIN (grey).

Figure 5. CaMKII inhibition increases MLCK but not ROCK activity

(A) ROCK activity assay by ELISA for MYPT Thr696 phosphorylation. The ROCK-dependent phosphorylation was inhibited with Y-27632 (10 μM). (B) Vasoconstriction to Ang-II with ROCK inhibitor H-1152 (100 nM) (n=4-6). (C) Representative immunoblots for p-MLCK, MLCK and β-actin (n=5). (D) Densitometry of immunoblots in C. (E) MLCK kinase assay. (F) Representative immunoblots for p-MLC20 and MLC20 (n=6). Ponceau Red-stained gel to demonstrate equal protein loading. (G) Densitometry of immunoblots in F. * p<0.05 compared to WT. WT (black lines or bars), TG SM-CaMKIIN (grey).