deltaPKC inhibition or varepsilonPKC activation repairs endothelial vascular dysfunction by regulating eNOS post-translational modification

Abstract

The balance between endothelial nitric oxide synthase (eNOS)-derived nitric oxide (NO) and reactive oxygen species (ROS) production determines endothelial-mediated vascular homeostasis. Activation of protein kinase C (PKC) has been linked to imbalance of the eNOS/ROS system, which leads to endothelial dysfunction. We previously found that selective inhibition of delta PKC (deltaPKC) or selective activation of epsilon PKC (varepsilonPKC) reduces oxidative damage in the heart following myocardial infarction. In this study we determined the effect of these PKC isozymes in the survival of coronary endothelial cells (CVEC). We demonstrate here that serum deprivation of CVEC increased eNOS-mediated ROS levels, activated caspase-3, reduced Akt phosphorylation and cell number. Treatment with either the deltaPKC inhibitor, deltaV1-1, or the varepsilonPKC activator, psivarepsilonRACK, inhibited these effects, restoring cell survival through inhibition of eNOS activity. The decrease in eNOS activity coincided with specific de-phosphorylation of eNOS at Ser1179, and eNOS phosphorylation at Thr497 and Ser116. Furthermore, deltaV1-1 or psivarepsilonRACK induced physical association of eNOS with caveolin-1, an additional marker of eNOS inhibition, and restored Akt activation by inhibiting its nitration. Together our data demonstrate that (1) in endothelial dysfunction, ROS and reactive nitrogen species (RNS) formation result from uncontrolled eNOS activity mediated by activation of deltaPKC or inhibition of varepsilonPKC; (2) inhibition of deltaPKC or activation of varepsilonPKC corrects the perturbed phosphorylation state of eNOS, thus increasing cell survival. Since endothelial health ensures better tissue perfusion and oxygenation, treatment with a deltaPKC inhibitor and/or an varepsilonPKC activator in diseases of endothelial dysfunction should be considered.

Figure 1. δV1-1 or ψεRACK decrease ROS levels in CVEC

ROS production in CVEC was measured by the DHE/HPLC assay. CVECs were treated with 1 µM δV1-1 or ψεRACK, or 10% BCS, in presence or absence of 200 µM L-NAME in 0.1% BCS, for 15 min (A) or 6 h (B). Results are expressed as peak area/mg protein ± S.D. *p<0.05 and **p<0.01 vs. 0.1% BCS.

Figure 2. Serum deprivation promotes endothelial cell dysfunction

Akt phosphorylation in serum-deprived CVEC exposed to 0.1% BCS for 15 min (A) or 6 h (B). Total Akt is used for normalization. The graphs represent quantification of gels. Akt activity is expressed as percent ± SD. ***p<0.001 vs. 0.1% BCS. C. Cleaved caspase-3 in serum-deprived CVEC exposed to 0.1% BCS for 6 h. Actin is used for normalization. Total caspase-3 is shown as control of loading. The graph represents quantification of gels. Caspase-3 activity is expressed as percent vs. 0.1% BCS. ***p<0.001 vs. 0.1% BCS. The gels shown in this and following figures are representative of three with similar results.

Figure 3. Effect of serum on δPKC and εPKC activation

Expression of δPKC (A) and εPKC (B) in CVEC exposed to 0.1% or 10% BCS. Subcellular distribution of δPKC (C) and εPKC (D) in CVEC exposed to 0.1% or 10% BCS for 15 min. The graphs represent quantification of percent enzyme translocation, calculated as (particulate fraction / particulate fraction + soluble fraction) * 100 ± S.D. **p<0.001 or ***p<0.001 vs. 10% BCS. Translocation of δPKC (E) and εPKC (F) in CVEC exposed to δV1-1 or ψεRACK (0.1-10 µM, 15 min). *p<0.05, **p<0.01 and ***p<0.001 vs. 0.1% BCS. In all blots, total actin is shown as loading control.

Figure 3. Effect of serum on δPKC and εPKC activation

Expression of δPKC (A) and εPKC (B) in CVEC exposed to 0.1% or 10% BCS. Subcellular distribution of δPKC (C) and εPKC (D) in CVEC exposed to 0.1% or 10% BCS for 15 min. The graphs represent quantification of percent enzyme translocation, calculated as (particulate fraction / particulate fraction + soluble fraction) * 100 ± S.D. **p<0.001 or ***p<0.001 vs. 10% BCS. Translocation of δPKC (E) and εPKC (F) in CVEC exposed to δV1-1 or ψεRACK (0.1-10 µM, 15 min). *p<0.05, **p<0.01 and ***p<0.001 vs. 0.1% BCS. In all blots, total actin is shown as loading control.

Figure 4. δV1-1 or ψεRACK promote Akt activation and cleaved caspase-3 inhibition

A. CVEC were serum-deprived for 24 h and then exposed to 10 % BCS or δV1-1 or ψεRACK or TAT in 0.1% BCS for 24 h. Data are reported as number of cells ± S.D. *p<0.05 and ***p<0.001 vs. 0.1% BCS. B. Akt activity in CVEC exposed to 1mM dV1-1 or ψεRACK in 0.1% BCS for 15 min. The graph represents quantification of gels (Arbitrary Densitometric Units). ***p<0.001 vs. 0.1% BCS. C. Cleaved caspase-3 in CVEC exposed to 1µM δV1-1 or ψεRACK in 0.1% BCS for 6 h. The graph represents quantification of gels (percent ± S.D. **p<0.01 and ***p<0.001 vs. 0.1% BCS). D. Nitrated-Akt in CVEC treated with 1µM dV1-1 or ψεRACK in 0.1% BCS for 15 min. The graph represents quantification of gels (Arbitrary Densitometric Units). ***p<0.001 vs. 0.1% BCS. Nitrated (E) and phosphorylated (F) Akt were measured after treatment of SIN-1 (200 µM, 15 min) in the presence or absence of both peptides. Results are reported as A.D.U. (Arbitrary Densiometric Unit). *p<0.05 and **##p<0.01 vs. δV1-1; ^§§p<0.01 vs. ψεRACK.

Figure 4. δV1-1 or ψεRACK promote Akt activation and cleaved caspase-3 inhibition

A. CVEC were serum-deprived for 24 h and then exposed to 10 % BCS or δV1-1 or ψεRACK or TAT in 0.1% BCS for 24 h. Data are reported as number of cells ± S.D. *p<0.05 and ***p<0.001 vs. 0.1% BCS. B. Akt activity in CVEC exposed to 1mM dV1-1 or ψεRACK in 0.1% BCS for 15 min. The graph represents quantification of gels (Arbitrary Densitometric Units). ***p<0.001 vs. 0.1% BCS. C. Cleaved caspase-3 in CVEC exposed to 1µM δV1-1 or ψεRACK in 0.1% BCS for 6 h. The graph represents quantification of gels (percent ± S.D. **p<0.01 and ***p<0.001 vs. 0.1% BCS). D. Nitrated-Akt in CVEC treated with 1µM dV1-1 or ψεRACK in 0.1% BCS for 15 min. The graph represents quantification of gels (Arbitrary Densitometric Units). ***p<0.001 vs. 0.1% BCS. Nitrated (E) and phosphorylated (F) Akt were measured after treatment of SIN-1 (200 µM, 15 min) in the presence or absence of both peptides. Results are reported as A.D.U. (Arbitrary Densiometric Unit). *p<0.05 and **##p<0.01 vs. δV1-1; ^§§p<0.01 vs. ψεRACK.

Figure 5. δV1-1 and ψεRACK modulate eNOS phosphorylation and activity

A. Phosphorylation of eNOS on Ser1179, Thr497, Ser116 or Ser635 in CVEC exposed to 10% BCS, or 1 µM δV1-1 or ψεRACK in 0.1% BCS for 15 min. Total eNOS is used for normalization. The graph represents quantification of gels (Arbitrary densitometric units). ***p<0.001 vs. 0.1% BCS. B. cGMP production in CVEC treated with 10% BCS, or 1 µM δ V1-1 or ψεRACK in 0.1% BCS for 15 min. Data are expressed as (pg/ml)/mg protein. **p<0.01 vs. 0.1% BCS. or (C) eNOS phosphorylation (see above for phosphorylation sites) in CVEC exposed to 10% BCS, or 1 µM δV1-1 or ψεRACK in 0.1% BCS for 6h. Total eNOS is used for normalization. The graph represents quantification of gels (Arbitrary densitometric units). **p<0.01 and ***p<0.001 vs. 0.1% BCS. D. cGMP production in CVEC treated with 10% BCS, or 1 µM δ V1-1 or ψεRACK in 0.1% BCS for 6 h. Data are expressed as (pg/ml)/mg protein. ^#p<0.05 and ^##p<0.01 vs. 0.1% BCS.

Figure 6. eNOS inhibition controls Akt activity

CVEC were silenced for bovine eNOS and transfected with the phospho-mimetic or the dephospho-mimetic form of human eNOS in Thr495 (A) or Ser114 (B) (T495D and T495A or S114D and S114A, respectively). To verify cell transfection, western blot for eNOS Thr495 (A, upper panels) or Ser114 phosphorylation (B, upper panels) was carried out and normalized for total eNOS In these cells cGMP levels were: (pg/mg protein): 3.5 ± 0.4 for S114D and 3.3 ± 0.2 for T495D, p<0.01 vs. Vector 7.2 ± 0.3; 7.5 ± 0.2 for S114A and 6.9 ± 0.4 for T495A) of and ROS/RNS levels were (peak area/mg protein): 96 ± 4 for S114D and 98 ± 2 for T495D, p<0.01 vs. Vector 135 ± 3; 129 ± 4 for S114A and 132 ± 4 for T495A). The graphs represent quantification of phospho-Akt/total Akt gels (Arbitrary Densitometric Units); ***p<0.001 vs. Vector and ^##p<0.01 vs. T495D or S114D respectively. C. Akt activity in CVEC exposed to L-NAME (15 min, 200 µM). The graph represents quantification of gels (Arbitrary Densitometric Units).**p<0.01 vs. 0.1% BCS D Cleaved caspase-3 in CVEC exposed to L-NAME (6 h, 200 µM) in 0.1% BCS. The graph represents quantification of gels (percent ± S.D.). ***p<0.001 vs. 0.1% BCS. E Nitrated-Akt in CVEC treated with L-NAME (15 min, 200 µM) in 0.1% BCS. The graph represents quantification of gels (Arbitrary Densitometric Units). ***p<0.001 vs. 0.1% BCS. In the presence of L-NAME the number of cells was: 153 ± 6, 200 µM L-NAME, vs. 0.1% BCS 105 ± 4; p<0.01.

Figure 6. eNOS inhibition controls Akt activity

CVEC were silenced for bovine eNOS and transfected with the phospho-mimetic or the dephospho-mimetic form of human eNOS in Thr495 (A) or Ser114 (B) (T495D and T495A or S114D and S114A, respectively). To verify cell transfection, western blot for eNOS Thr495 (A, upper panels) or Ser114 phosphorylation (B, upper panels) was carried out and normalized for total eNOS In these cells cGMP levels were: (pg/mg protein): 3.5 ± 0.4 for S114D and 3.3 ± 0.2 for T495D, p<0.01 vs. Vector 7.2 ± 0.3; 7.5 ± 0.2 for S114A and 6.9 ± 0.4 for T495A) of and ROS/RNS levels were (peak area/mg protein): 96 ± 4 for S114D and 98 ± 2 for T495D, p<0.01 vs. Vector 135 ± 3; 129 ± 4 for S114A and 132 ± 4 for T495A). The graphs represent quantification of phospho-Akt/total Akt gels (Arbitrary Densitometric Units); ***p<0.001 vs. Vector and ^##p<0.01 vs. T495D or S114D respectively. C. Akt activity in CVEC exposed to L-NAME (15 min, 200 µM). The graph represents quantification of gels (Arbitrary Densitometric Units).**p<0.01 vs. 0.1% BCS D Cleaved caspase-3 in CVEC exposed to L-NAME (6 h, 200 µM) in 0.1% BCS. The graph represents quantification of gels (percent ± S.D.). ***p<0.001 vs. 0.1% BCS. E Nitrated-Akt in CVEC treated with L-NAME (15 min, 200 µM) in 0.1% BCS. The graph represents quantification of gels (Arbitrary Densitometric Units). ***p<0.001 vs. 0.1% BCS. In the presence of L-NAME the number of cells was: 153 ± 6, 200 µM L-NAME, vs. 0.1% BCS 105 ± 4; p<0.01.

Figure 7. δV1-1 and ψεRACK modulate eNOS protein-protein interaction

CVEC were treated with ψεRACK (15 min, 1µM), then immunoprecipitated for anti-Hsp90 (A) or Cav-1 (B) and immunoblotted for eNOS or εPKC. CVEC were treated with δV1-1 (15 min, 1 µM), then immunoprecipitated with anti-Cav-1 (C) or Hsp90 (D) and immunoblotted for δPKC or eNOS. Cells were treated with δV1-1 or ψεRACK (15 min, 1µM) then immunoprecipitated for anti-eNOS and immunoblotted for εPKC (E) or δPKC (F). Also in panel F are levels of δPKC and eNOS in the respective initial lysate.

Figure 8. Schematic model of relationship between eNOS and PKC isoforms in low serum and after treatment with δV1-1 or ψεRACK

A. In nutrient deprivation conditions δPKC is activated and associated with Cav-1, while inactive εPKC is localized in the cytoplasm and associated with eNOS and Hsp90. In this condition, eNOS is activated, high levels of ROS/RNS are produced, and Akt is nitrated. B. Treatment with δV1-1 or with ψεRACK shifted εPKC to the cytoplasm and εPKC translocates to the membrane, respectively. Active ePKC associates with eNOS and Cav-1, and promotes inhibition of eNOS by increasing eNOS phosphorylation at Thr497 and Ser116 sites. Inhibition of eNOS results in decreased ROS and RNS production and increased phosphorylation of Akt.