The involvement of protein kinase C-ε in isoflurane induced preconditioning of human embryonic stem cell - derived Nkx2.5⁺ cardiac progenitor cells

The methods for hESCs culture, the generation of hESCs-derived CPCs and characterization of differentiated CPCs were the same as those of previous studies [5, 10, 11]. This study did not include personal information because human embryonic stem cells were obtained from the National Stem Cell Bank. Institutional review board of Institute of Reproductive Medicine and Population approval was obtained (number: IRMP-IRB2009112) and individual patient consent was waived as there were no specific human participants.

In brief, H1 hESC (WA01) was cultured on a feeder cell layer of mouse embryonic fibroblasts inactivated with mitomycin C (Invitrogen, Carlsbad, CA, USA). Colonies of hESCs were passaged every week by a mechanical micro-dissection method. After 60 % of the plates were covered with undifferentiated cell colonies for over 7 days, the cells were made into small clumps with incubation in the collagenase solution. They were then moved to suspension culture in a 24-well non-treated dish with a differentiation medium and were soaked for 4 days until the cells became embryoid bodies (EBs). A total of 20 ng/ml of bone morphogenetic protein 4 (R&D Systems, Minneapolis, MN, USA) was mixed with the differentiation medium for the suspension culture. The EBs were moved into Matrigel-coated 35-mm plates (BD Biosciences, San Jose, CA, USA) and expanded for 2–3 weeks. Immunostaining with antibodies directed against human Nkx2.5 (mouse monoclonal, R&D Systems) was done to detect differentiated CPCs. 4, 6-diamidino-2-phenylindole, DAPI (Vector Laboratories, Burlingame, CA, USA) was used to stain all nuclei. DAPI-positive cells were counted as the total number of cells and the number of Nkx2.5-positive cells was expressed as a percentage of the total number of cells. The images from 10 random microscopic fields of each plate at × 200 magnification were obtained.

A cell survival study was performed using a previously reported protocol [5, 12]. The differences were the drugs for the experiments. The protocols for the experiments are illustrated in Fig. 1. In brief, differentiated CPC clusters were separated with 0.25 % of trypsin-ethylenediaminetetraacetic acid solution for 5 min and were washed in the differentiation medium. Several drops of suspension including CPCs were placed in a chamber on the stage of an inverted microscope (ECLIPSE TS100; Nikon, Tokyo, Japan) until cells were settled. We stained those with 0.5 ml of 0.4 % trypan blue solution (Sigma-Aldrich, St. Louis, MO, USA) for 2 min followed by washout with glucose-free Tyrode solution (132 mM NaCl, 10 mM, HEPES, 5 mM KCl, 1 mM CaCl₂ and 1.2 mM MgCl₂ adjusted to pH 7.4). Round cells that excluded trypan blue were considered living. We checked the location of living cells by a chamber bottom with a labelled grid. Approximately one hundred living CPCs were counted for 10 min in each experiment. For example, N=7 refers to the counting the numbers of about 700 living CPCs. After the cell count, the plate was washed with glucose-free Tyrode for 30 min, then the CPCs were exposed to 200 mM H₂O₂ and 100 mM FeSO₄ · 7H₂O (Sigma-Aldrich) for 20 min to simulate the oxidative stress. The CPCs were washed out with glucose-free Tyrode solution for 20 min. Then the remaining living CPCs were counted after staining with trypan blue. For APC, the cells were exposed to isoflurane dissolved Tyrode solution for 20 min before exposure to the oxidative stress. We changed all materials with a potential to contact CPCs such as plastic tubing, plastic syringes, cover glasses and slide glasses to new ones in every experiment to prevent contamination between treatment groups. Moreover we randomized the order of experiments daily and the researchers who counted the cell deaths with a microscope were blinded regarding drug treatment.

Isoflurane was prepared by the same method as described in the previous study [5]. Briefly, it was dissolved in glucose-free Tyrode solution by sonication and delivered to CPCs by airtight syringes. We exposed CPCs to 0.25 mM, 0.5 mM and 1.0 mM isoflurane. To confirm the concentrations of isoflurane which varied ± 10 % of the reported value, analyses were performed by gas chromatography (Shimadzu, Kyoto, Japan).

A PKC-ε inhibitor, cell permeable PKC-ε V1-2 peptide (εV1-2) and a selective PKC-ε activator, pseudo-receptors for activated C-kinase epsilon (εψRACK) were kindly provided gratis by the laboratory of Daria Mochly-Rosen (Department of Chemical and Systems Biology, Stanford University, CA). We dissolved an isoform-nonspecific PKC inhibitor, chelerythrine (1 μM; Sigma-Aldrich), and εV1-2 (1 μM) in the superfusing solution during oxidative stress to examine the effects of the PKC and PKC-ε on cell survival [13]. An isoform-nonspecific PKC activator, 4β-phorbol 12-myristate 13-acetate (PMA, 10 nM) and a selective PKC-ε activator, pseudo-receptors for activated C-kinase epsilon (εψRACK, 2 μM) were mixed with the superfusion solution for oxidative stress to illuminate the effects of the PKC and their isoform on cell survival. PMA was dissolved in dimethyl sulfoxide and εψRACK was dissolved in deionized distilled water. All stock solutions were diluted to the required concentrations in the superfusing buffer immediately before usage.

Confocal microscopic examinations were done with five plates to identify CPCs derived from hESCs with early cardiac marker (Nkx2.5). The proportion of the cells stained with Nkx2.5 was 95 ± 3 % of the total cell number (Fig. 2).

Glucose-free Tyrode solution did not have an effect on the death rate of CPCs in the time-control group (9.6 ± 5.4 %, n = 7). Oxidative stress increased the CPCs death rate to 31.4 ± 10.2 % (n = 11). For the evaluation of the preconditioning effect of isoflurane, three concentrations were used. Preconditioning with 0.25 mM of isoflurane did not lower the death rate of CPCs (36.7 ± 18.0 %, n = 10). However, 0.5 mM and 1.0 mM of isoflurane decreased CPCs death rate to 12.7 ± 9.3 % (n = 7) and 12.0 ± 7.7 % (n = 7) respectively without a significant difference between the two concentrations (Fig. 3).

PMA and chelerythrine themselves had no effect on the death rate of CPCs in the time-control groups (11.7 ± 5.9 %, n = 7 and, 8.9 ± 3.8 %, n = 7). PMA had a protective effect on CPCs when they were under oxidative stress: the death rates of CPCs were 16.0 ± 3.2 % (n = 8). When CPCs were treated with PMA and 0.5 mM of isoflurane, PMA did not reduce or potentiate the protective effect of isoflurane (12.7 ± 7.0 %, n = 7) (Fig. 4a). Chelerythrine did not show additional effects on the death rate of CPCs under oxidative stress (24.1 ± 6.1 %, n = 8). However, it abolished the preconditioning effects of isoflurane on CPCs under oxidative stress (27.6 ± 13.5 %, n = 13) (Fig. 4b).

There were no significant effects on the death rate of CPCs by εψRACK and εV1-2 themselves in the time-control group (7.8 ± 2.4 %, n = 8, and 12.8 ± 5.7 %, n = 8). εψRACK had a protective effect on CPCs when they were under oxidative stress: the death rates of CPCs were 10.6 ± 3.6 % (n = 8). When εψRACK was added after isoflurane wash out, it did not reduce or potentiate the protective effect of 0.5 mM of isoflurane (7.8 ± 3.7 %, n = 7) (Fig. 5a). εV1-2 did not show additional effects on the death rate of CPCs when they were under oxidative stress (25.2 ± 8.0 %, n = 8). However, it abolished the preconditioning effect of isoflurane on CPCs under oxidative stress (25.9 ± 8.7 %, n = 8) (Fig. 5b).

Our data showed the possibilities of using the activator of PKC or PKC-ε instead of preconditioning by artificially inducing ischemia or by exposure to anesthetics for the prevention of ischemic injury, making the procedure more convenient by simple injection of drugs. However, the study hinges on the specificity of the PKC inhibitors and activators, and is largely pharmacologic in nature. The lack of control peptides is a limitation. The safety of the drugs should be elucidated further before clinical application. For example, PMA, a non-specific PKC activator is known to be a carcinogen [27]. In addition, nonspecific activators of PKC should be used with caution, although our data showed protective effects, because some of isoforms are known to have the opposite effects [18]. Accordingly PKC-ε activator like εψRACK might be a promising choice in this aspect, however further evaluation concerning efficacy and toxicity in human beings is also required. Another limitation was whether the study with hESCs could reflect the situation of the real human hearts in vivo as this study was exclusively performed in tissue culture experiments, which is required for future studies in the clinical setting.