Optical-electronic hybrid Fourier convolutional neural network based on super-pixel complex-valued modulation.

An optical-electronic hybrid convolutional neural network (CNN) system is proposed and investigated for its parallel processing capability and system design robustness. It is regarded as a practical way to implement real-time optical computing. In this paper, we propose a complex-valued modulation method based on an amplitude-only liquid-crystal-on-silicon spatial light modulator and a fixed four-level diffractive optical element. A comparison of computational results of convolutions between different modulation methods in the Fourier plane shows the feasibility of the proposed complex-valued modulation method. A hybrid CNN model with one convolutional layer of multiple channels is proposed and trained electrically for different classification tasks. Our simulation results show that this model has a classification accuracy of 97.55% for MNIST, 88.81% for Fashion MNIST, and 56.16% for Cifar10, which outperforms models using only amplitude or phase modulation and is comparable to the ideal complex-valued modulation method.

Inactivation of Klf5 by zinc finger nuclease downregulates expression of pluripotent genes and attenuates colony formation in embryonic stem cells.

Recent studies suggest that Klf5 is required to maintain embryonic stem (ES) cells in an undifferentiated state. However, whether Klf5 can be inactivated by novel fusion technology of zinc finger nucleases (ZFN) has never before been examined. Therefore, we used ZFN technology to target the Klf5 gene in mouse ES cells, and examined the effects of the Klf5 gene on the expression of pluripotency-related genes, Oct3/4, Nanog, and Sox2 and on the self-renewal of ES cells. In Klf5-ZFN-transfected cells, expression of the Klf5 mRNA was downregulated by ~80% compared to the control. Furthermore, expression of the Oct3/4 and Nanog mRNAs was significantly decreased in the Klf5-ZFN-targeted cells. RT-PCR analysis, however, showed no significant change in the level of Sox2 mRNA, but a decreased trend was evident in the Klf5-ZFN-targeted cells. Moreover, we observed the spontaneous differentiation of Klf5-ZFN-transfected cells and quantitative analysis revealed a significant decrease in colony formation in Klf5-ZFN-transfected cells. In conclusion, our data suggest that ZFN methodology is an effective approach to target the Klf5 gene and that Klf5 plays an important role in the maintenance of ES cell self-renewal.

Induced pluripotent stem (iPS) cells repair and regenerate infarcted myocardium.

Cardiac myocyte differentiation reported thus far is from iPS cells generated from mouse and human fibroblasts. However, there is no article on the generation of iPS cells from cardiac ventricular specific cell types such as H9c2 cells. Therefore, whether transduced H9c2 cells, originally isolated from embryonic cardiac ventricular tissue, will be able to generate iPS cells and have the potential to repair and regenerate infarcted myocardium remains completely elusive. We transduced H9c2 cells with four stemness factors, Oct3/4, Sox2, Klf4, and c-Myc, and successfully reprogrammed them into iPS cells. These iPS cells were able to differentiate into beating cardiac myocytes and positively stained for cardiac specific sarcomeric α-actin and myosin heavy chain proteins. Following transplantation in the infarcted myocardium, there were newly differentiated cardiac myocytes and formation of gap junction proteins at 2 weeks post-myocardial infarction (MI), suggesting newly formed cardiac myocytes were integrated into the native myocardium. Furthermore, transplanted iPS cells significantly (p < 0.05) inhibited apoptosis and fibrosis and improved cardiac function compared with MI and MI+H9c2 cell groups. Moreover, our iPS cell derived cardiac myocyte differentiation in vitro and in vivo was comparable to embryonic stem cells in the present study. In conclusion we report for the first time that we have H9c2 cell-derived iPS cells which contain the potential to differentiate into cardiac myocytes in the cell culture system and repair and regenerate infarcted myocardium with improved cardiac function in vivo.

Inhibition of plasminogen activator inhibitor-1 expression in vascular smooth muscle cells by protoporphyrins through a heme oxygenase-independent mechanism.

Heme oxygenase-1 (HO-1), a rate-limiting enzyme in heme catabolism, has been shown to play a regulatory role in the expression of plasminogen activator inhibitor-1 (PAI-1), a risk factor for vascular disease. Accordingly, we examined the effect of protoporphyrins, both HO inhibitors and activators, on PAI-1 expression in human vascular smooth muscle cells (VSMCs). Tin-protoporphyrin (SnPP) markedly inhibited the transforming growth factor beta1 (TGFbeta1)-induced expression of PAI-1 protein. Protoporphyrins, whether they are inhibitors or activators of HO, produced a similar inhibitory effect. However, SnPP had no effect on the level of PAI-1 mRNA transcripts. Knockdown of human HO-1 with a specific siRNA did not reduce the PAI-1 protein level in TGFbeta1-treated cells. In addition, the proteasome inhibitor lactacystin reversed the inhibitory effect of SnPP on PAI-1 protein expression. Both cobalt-protoporphyrin (CoPP) and CoCl2 markedly induced HO-1 expression. However, CoPP did not affect PAI-1 gene expression, whereas CoCl2 upregulated PAI-1 mRNA in a dose-dependent manner. Our results demonstrate that protoporphyrins can block the TGFbeta1-mediated induction of PAI-1 protein in VSMCs and that this inhibitory effect is independent of HO activity.

Oxidative stress enhances phosphorylation of p53 in neonatal rat cardiomyocytes.

p53 is an important regulator of cell growth and apoptosis and its activity is regulated by phosphorylation. Accordingly, in neonatal rat cardiomyocytes we examined the involvement of p53 in H(2)O(2)-induced apoptosis. Treatment with 50-100 microM H(2)O(2) markedly induced apoptosis in cardiomyocytes, as assessed by gel electrophoresis of genomic DNA. To examine whether H(2)O(2) increases p53 phosphorylation in cardiomyocytes, we utilized an antibody that specifically recognizes phosphorylated p53 at serine-15. The level of phosphorylated p53 was markedly increased by 100 microM H(2)O(2) at 30 and 60 min. Using specific protein kinase inhibitors we examined the involvement of protein kinases in p53 phosphorylation in response to H(2)O(2) treatment. However, staurosporine, a broad spectrum inhibitor of protein kinases, SB202190, a specific p38 kinase inhibitor, PD98059, a MAP kinase inhibitor, wortmannin, an inhibitor of DNA-PK and PI3 kinase, SP600125, a JNK inhibitor and caffeine,an inhibitor of ATM and ATR, failed to prevent the H(2)O(2)-induced phosphorylation of p53. cDNA microarray revealed that H(2)O(2) markedly increased expression of several p53 upstream modifiers such as the p300 coactivator protein and several downstream effectors such as gadd45, but decreased the expression of MDM2, a negative regulator of p53. Our results suggest that phosphorylation of p53 at serine-15 may be an important signaling event in the H(2)O(2)-mediated apoptotic process.

Phosphorylation and hypoxia-induced heme oxygenase-1 gene expression in cardiomyocytes.

BACKGROUND: Heme oxygenase-1 (HO-1) is a stress protein and the rate-limiting enzyme in heme degradation. We sought to examine the notion that protein kinases and phosphatases through phosphorylation and dephosphorylation modulate the HO-1 expression in cardiomyocytes under hypoxic conditions. METHODS AND RESULTS: Exposure of neonatal rat cardiomyocytes to hypoxia markedly induced the HO-1 expression, as assessed by Northern blot, Western blot, and transfection assay. The hypoxia-induced HO-1 expression was blocked by the kinase inhibitors staurosporine and SB202190 in a dose-dependent manner. Hypoxia decreased the activity of phosphatase-1 (PP-1). To examine the effect of PP-1 inhibition on HO-1 expression we used the phosphatase inhibitor okadaic acid (OA) and an antisense vector. OA treatment or overexpression of the antisense PP-1 transcript markedly induced HO-1 expression. Furthermore, transfection assay using HO-1 promoter constructs revealed the involvement of the nuclear factor kB (NF-kB) and Activator protein-1 (AP-1) in the hypoxia-induced activation of the HO-1 gene. The HO-1 promoter activity was modulated by OA under normoxic conditions or staurosporine under hypoxia. CONCLUSIONS: Our results suggest that activation of protein kinases and downregulation of PP-1 activity contribute to the hypoxia-induced HO-1 gene expression and that the proximal HO-1 promoter region containing NF-kB and AP-1 binding sites is likely to play a role in the transcriptional activation of the HO-1 gene in cardiomyocytes in response to hypoxic stress.

Mitochondrial Ca2+ flux and respiratory enzyme activity decline are early events in cardiomyocyte response to H2O2.

Oxidative stress is involved in mitochondrial apoptosis, and plays a critical role in ischemic heart disease and cardiac failure. Exposure of cardiomyocytes to H(2)O(2) leads to oxidative stress and mitochondrial dysfunction. In this study, we investigated the temporal order of mitochondrial-related events in the neonatal rat cardiomyocyte response to H(2)O(2) treatment. At times ranging from 10 to 90 min after H(2)O(2) treatment, levels were determined for respiratory complexes I, II, IV and V, and citrate synthase activities, mitochondrial Ca(2+) flux, intracellular oxidation, mitochondrial membrane potential and apoptotic progression. Complexes II and IV activity levels were significantly reduced within 20 min of H(2)O(2) exposure while complexes I and V, and citrate synthase were unaffected. Mitochondrial membrane potential declined after 20 and 60 min of H(2)O(2) exposure while intracellular oxidation, declining complex I activity and apoptotic progression were detectable only after 60 min. Measurement of mitochondrial Ca(2+) ([Ca(2+)](m)) using rhodamine 2 detected an early accumulation of [Ca(2+)](m) occurring between 5 and 10 min. Pretreatment of cardiomyocytes with either ruthenium red or cyclosporin A abrogated the H(2)O(2)-induced decline in complexes II and IV activities, indicating that [Ca(2+)](m) flux and onset of mitochondrial permeability transition pore opening likely precede the observed early enzymatic decline. Our findings suggest that [Ca(2+)](m) flux represents an early pivotal event in H(2)O(2)-induced cardiomyocyte damage, preceding and presumably leading to reduced mitochondrial respiratory activity levels followed by accumulation of intracellular oxidation, mitochondrial membrane depolarization and apoptotic progression concomitant with declining complex I activity.

Adenoviral SERCA1 overexpression triggers an apoptotic response in cultured neonatal but not in adult rat cardiomyocytes.

Although adenoviral SERCA gene transfer improves cardiac function in the failing heart, there is controversy regarding the cytotoxic effect of such overexpression. We sought to examine the effect of SERCA1 overexpression on neonatal (NRCMs) and adult rat cardiomyocytes (ARCMs). Cultured NRCMs and ARCMs were infected with adenoviral vector expressing EGFP (Ad.EGFP) or SERCA1 (Ad.SERCA1). Gel electrophoresis and microarray analysis were performed to examine DNA fragmentation and apoptosis-related genes, respectively. Northern and western blot were used to examine the expression level of SERCA. The optimal viral titer for Ad.EGFP was found to be 2-5 pfu/cell in NRCMs and 100 pfu/cell in ARCMs. Infection of NRCMs with 4 pfu/cell of Ad.SERCA1 resulted in loss of cells and DNA fragmentation, while no apoptosis was detected in Ad.EGFP-infected cells. Gene array analysis also revealed that Ad.SERCA1 induced expression of apoptosis-related genes in NRCMs. However, neither loss of cell viability nor DNA fragmentation and induction of apoptosis-related genes was observed in ARCMs infected with 100 pfu/cell of Ad.SERCA1. Following the optimal infection, the SERCA1 expression level in NRCMs was 4-fold higher than that in ARCMs. Interestingly, the endogenous SERCA2 protein in NRCMs was completely replaced by exogenous SERCA1 protein. Furthermore, the activity of Ca2+ ATPase was increased by 4-fold in NRCMs but only by 1.5-fold in ARCMs. Our results indicate that adenoviral SERCA1 gene transfer has appreciably different effects on NRCMs and ARCMs. SERCA1 overexpression triggers an apoptotic response in NRCMs but not in ARCMs.

Effects of PP1/PP2A inhibitor calyculin A on the E-C coupling cascade in murine ventricular myocytes.

Calyculin A was used to examine the importance of phosphatases in the modulation of cardiac contractile magnitude in the absence of any neural or humoral stimulation. Protein phosphatase (PP)1 and PP2A activity, twitch contractions, intracellular Ca(2+) concentration ([Ca(2+)](i)) transients, action potentials, membrane currents, and myofilament Ca(2+) sensitivity were measured in isolated mouse ventricular myocytes. Calyculin A (125 nM) inhibited PP1 and PP2A by 50% and 85%, respectively, whereas it doubled the twitch magnitude and increased twitch duration by 50% in field-stimulated cells. Calyculin A-evoked increases in L-type Ca(2+) current (70%) and the resulting [Ca(2+)](i) transient (83%) explain the positive inotropic response. However, increases in twitch and action potential durations did not result from increased myofilament Ca(2+) sensitivity or K(+) current inhibition, respectively. Comparison of the effects of calyculin A and isoproterenol on [Ca(2+)](i) transients and twitch contractions revealed that calyculin A had a much smaller lusitropic effect than the beta-agonist, indicating that calyculin A did not significantly increase sarcoplasmic reticulum Ca(2+) reuptake. Thus while cardiac contractile magnitude is controlled by a steady-state kinase/phosphatase balance, this regulation is not equally operative at all of the steps in the excitation-contraction coupling pathway and may in fact be most important to the regulation of the L-type Ca(2+) channel.