Adenovirus-delivered short hairpin RNA targeting PKCα improves contractile function in reconstituted heart tissue

Journal of Molecular and Cellular Cardiology 43 (2007) 371 – 376 www.elsevier.com/locate/yjmcc

Brief communication

Adenovirus-delivered short hairpin RNA targeting PKCα improves contractile function in reconstituted heart tissue Ali El-Armouche a,⁎,1 , Jasmin Singh a,1 , Hiroshi Naito a,1 , Katrin Wittköpper a , Michael Didié a , Alexander Laatsch b , Wolfram-Hubertus Zimmermann a , Thomas Eschenhagen a,⁎ a

b

Institute of Experimental and Clinical Pharmacology and Toxicology, University Medical Center Hamburg-Eppendorf, Hamburg, Martinistraße 52, D-20246, Germany Institute of Biochemistry and Molecular Biology II, University Medical Center Hamburg-Eppendorf, Hamburg, D-20246, Germany Received 19 March 2007; received in revised form 23 April 2007; accepted 29 May 2007 Available online 7 June 2007

Abstract PKCα has been shown to be a negative regulator of contractility and PKCα gene deletion in mice protected against heart failure. Small interfering (si)RNAs mediate gene silencing by RNA interference (RNAi) and may be used to knockdown PKCα in cardiomyocytes. However, transfection efficiencies of (si)RNAs by lipofection tend to be low in primary cells. To address this limitation, we developed an adenoviral vector (AV) driving short hairpin (sh)RNAs against PKCα (Ad-shPKCα) and evaluated its potential to silence PKCα in neonatal rat cardiac myocytes and in engineered heart tissues (EHTs), which resemble functional myocardium in vitro. A nonsense encoding AV (Ad-shNS) served as control. Quantitative PCR and Western blotting showed 90% lower PKCα-mRNA and 50% lower PKCα protein in Ad-shPKCα-infected cells. EHTs were infected with Ad-shPKCα on day 11 and subjected to isometric force measurements in organ baths 4 days later. Mean twitch tension was N 50% higher in Ad-shPKCα compared to Ad-shNS-infected EHTs, under basal and Ca2+- or isoprenaline-stimulated conditions. Twitch tension negatively correlated with PKCα mRNA levels. In summary, AV-delivered shRNA mediated highly efficient PKCα knockdown in cardiac myocytes and improved contractility in EHTs. The data support a role of PKCα as a negative regulator of myocardial contractility and demonstrate that EHTs in conjunction with AV-delivered shRNA are a useful model for target validation. © 2007 Elsevier Inc. All rights reserved. Keywords: RNA interference; Adenovirus; PKCα; Contractility; Engineered heart tissue

1. Introduction The PKC family of Ca2+ and/or lipid-activated serine– threonine kinases functions downstream of many membraneassociated signal transduction pathways [1]. In the human heart, PKCα was shown to be the dominantly expressed conventional PKC isoform [2]. PKCα has been recently identified as an important regulator of cardiac contractility and Ca2+ handling [3]. PKCα gene-deleted mice were shown to be hypercontractile, whereas transgenic mice overexpressing PKCα were hypocon-

⁎ Corresponding authors. Tel.: +49 40 42803 2180; fax: +49 40 42803 5925. E-mail addresses: [email protected] (A. El-Armouche), [email protected] (T. Eschenhagen). 1 Authors contributed equally. 0022-2828/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.yjmcc.2007.05.021

tractile. Furthermore, enhancement in cardiac contractility associated with PKCα knockout or with pharmacological PKCα inhibition protected against heart failure in several experimental models [2,3]. These data suggest that inhibition of PKCα may serve as a novel therapeutic strategy for enhancing cardiac contractility in heart failure. Given the limited specificity of current small molecule inhibitors of PKC isoforms, RNA interference (RNAi)mediated gene silencing [4] could serve as an alternative approach. RNAi utilizes sequence-specific double-stranded small interfering RNA (siRNA) to silence gene expression in mammalian cells [5]. To overcome notoriously low transfection efficiencies of siRNAs in primary cardiac cells and even more in intact heart tissues, we developed a novel adenoviral vector (AV) driving transcription of short hairpin RNA (shRNA). Its efficacy was tested both in neonatal rat cardiac myocytes and three-

372

A. El-Armouche et al. / Journal of Molecular and Cellular Cardiology 43 (2007) 371–376

dimensional engineered heart tissues (EHT) to monitor contracile consequences of the shRNA-mediated gene manipulation. 2. Methods 2.1. Cell culture and transfection Neonatal rat cardiac myocytes (NRCM) were isolated from 1–3 day old neonates as described previously [6]. NRCM were infected with Ad-EGFP 48 h before transfection as described earlier [6]. NIH/3T3 cells (mouse fibroblasts) were grown in DMEM supplemented with 10% FCS and transfected at 50–80% confluency. Transfection of NIH/3T3 and NRCM was carried out using Polyfect (Qiagen) with 3′-rhodamine-labeled control siRNA (UUCUCCGAACGUGUCACGUdTdT, Qiagen) or 3′rhodamine-labeled EGFP-22 siRNA (GCAAGCUGACCCUGAAGUUCAU, Qiagen) according to the manufacturer's protocol. Transfection efficiencies were determined after 24 h by fluorescence microscopy using a Zeiss LSM 510. 2.2. Generation of PKCα RNAi expression plasmid and adenovirus To obtain a vector-based suppression of PKCα, a selfannealing oligonucleotide targeting both rat and murine PKCα (target sequence: AACTCACAGACTTCAACTTCCTCATG) and a scrambled siRNA (GCTCGCGTATGGCGGATAGTACTAG) were inserted into the recently described short hairpin RNA expression vector pALsh [7]. The sh-sequence against PKCa was inserted downstream of an H1 promotor as described [7]. A scrambled sequence was used accordingly to generate a control vector. Recombinant replication-deficient AV was generated utilizing a bicistronic pAdEasy-1 system expressing GFP under a separate CMV promoter as previously described [8]. AV was propagated in HEK 293 cells and purified by cesium chloride density gradient centrifugation. The titer of biologically active virus was determined by infecting a known number of NRCM with serial dilutions of the GFP-containing viruses. A multiplicity of infection (MOI) of 1 was defined as the amount of AV necessary to positively transfect 100% of cells after 48 h as determined by GFP fluorescence. 2.3. Engineered heart tissue and isometric force measurement Circular engineered heart tissue (EHT) was prepared as described previously [9]. EHTs represent three-dimensional heart tissue-like structures that exhibit spontaneous, regular and synchronous beating and allow measurement of force of contraction under isometric conditions. After 11 days in culture, EHTs were infected with AdshPKCα/GFP or AdshNS/GFP control (MOI 50) by adding the respective amount of virus stock to the culture medium. Isometric force of contraction was measured 4 days later in organ baths containing gassed Tyrode's solution. Electrically stimulated EHTs were stretched to Lmax and inotropic and lusitropic responses to cumulative concentrations of calcium (0.2–2.8 mM) and isoprenaline (0.1–1000 nM) were recorded.

2.4. Quantitative real-time RT-PCR Total RNA (2 μg) isolated from NIH/3T3, NRCM and EHTs was reverse transcribed using Moloney murine leukemia virus reverse transcriptase. cDNA was amplified with the TaqMan system (ABIPrism 7900, Applied Biosystems) as described previously [10]. Primers and probes for rat PKCα (for: CCTTTCCTTCGGCGTCTCA, rev: TGTAGTATTCACCCTCCTCTTGGTT, probe: AAGATGCCAGCCAGTGGATGGTACAAGT) and rat GAPDH (for: AACTCCCTCAAGATTGTCAGCAA, rev: CAGTCTTCTGAGTGGCAGTGATG, probe: ATGGACTGTGGTCATGAGCCCTTCCA) were designed to cross an intron/exon boundary to eliminate genomic DNA amplification. 2.5. Western blot analysis Western blotting was performed as described previously [6] with monoclonal primary antibodies against PKCα (1:200, Santa Cruz) and GAPDH (1:2000, Hytest). 2.6. Statistical analysis Unpaired Student's t tests (two-sided) were used to compare means between 2 independent groups/samples. For correlation analysis the Spearman–Rank test was used. Concentration– response curves were analyzed by repeated-measures ANOVA. Average data are presented as mean ± SEM. Differences were considered significant when p b 0.05. 3. Results 3.1. Feasibility of RNAi in cardiac myocytes using siRNA transfection NRCM were infected with a GFP-only AV (MOI 3, 24 h) followed by Polyfect-mediated transfection of rhodaminelabeled siRNA targeting GFP. Several conditions including different siRNA (0.6–15 μg), Polyfect amounts (10–20 μl) and incubation times (24–48 h) were tested. However, neither condition allowed GFP silencing in NRCM (Fig. 1A), likely because the siRNA was not taken up in NRCM (intracellular rhodamine-signal after 8 h; Fig. 1B, top). The general feasibility of the siRNA approach was confirmed in NIH-3T3 cells. Here, transfection efficiency reached N 40% (intracellular rhodaminesignal after 8 h; Fig. 1B, bottom). Consequently, the plasmids pALsh [7] encoding for small hairpin RNA against PKCα (pALsh shPKCα) or nonsense control (pALsh shNS) were transfected into NIH/3T3 cells for 48 h (Fig. 1C). Protein analyses by Western blotting demonstrated effective PKCα knockdown using pALsh shPKCα (Fig. 1D). 3.2. Adenovirally delivered shRNA for PKCα RNAi in NRCM and functional consequences in circular engineered heart tissue (EHT) To obtain efficient gene silencing in NRCM, we generated recombinant AV encoding for shPKCα (Ad-EGFP-shPKCα) or

A. El-Armouche et al. / Journal of Molecular and Cellular Cardiology 43 (2007) 371–376

373

Fig. 1. Feasibility of RNAi in cardiac myocytes using siRNA transfection. (A) Neonatal rat cardiac myocytes (NRCM) were infected with a GFP-only adenovirus (MOI 3, 24 h) followed by Polyfect-mediated transfection of rhodamine-labeled siRNA targeting GFP. (B) NRCM and mouse NIH-3T3 cells were transfected with rhodamine-labeled siRNA and monitored by fluorescence microscopy. (C) Cloning scheme of shRNA target sequences into pALsh. (D) Western blot for PKCα and GAPDH from pALsh-shPKCα or -shNS transfected NIH3T3 cells.

shNS (Ad-EGFP-shNS) under H1 promoter control (Fig. 2A). Parallel GFP expression allowed evaluation of transfection efficiency (Fig. 2B). Efficient PKCα knockdown was achievable with an MOI ≥ 1 (Fig. 2C). Consequently, we used an MOI of 3 and 48 h incubation for all further experiments in NRCM monolayer cultures. Real-time PCR and Western blotting demonstrated 90% PKCα-mRNA knockdown (p b 0.01, Fig. 2D) resulting in markedly lower PKCα protein when compared to the nonsense control group (p b 0.01, Fig. 2E).

To address functional consequences of PKCα knockdown, we infected EHTs with Ad-EGFP-shPKCα and Ad-EGFPshNS at an MOI of 50 and analyzed force development after 4 days (Fig. 2F). Thereafter, each EHT was tested for efficient gene transfer by GFP epifluorescence and standard microscopy. In general, an MOI of 50 was sufficient to reach a high transfection efficiency (inset in Fig. 2F). PKCα knockdown increased basal twitch tension (TT: 0.39 ± 0.03 mN vs. 0.25 ± 0.04 mN in shNS, p b 0.05), max-

374

A. El-Armouche et al. / Journal of Molecular and Cellular Cardiology 43 (2007) 371–376

Fig. 2. Adenovirus-delivered short hairpin RNA targeting PKCα improves contractile function in circular engineered heart tissue (EHTs). (A) H1-shPKCa/NS were cloned into a pAdTrack vector. (B) Titration of recombinant adenoviruses encoding bicistronically GFP and shPKCα or shNS. (C) Western blot for PKCα and GAPDH after infection with AdshNS or AdshPKCα. (D) Real-time PCR for PKCα mRNA levels from AdshPKCα/NS infected NRCM. (E) Western blot for GAPDH and PKCα from AdshPKCα/NS infected NRCM. (F) Microscopic view of EHTs infected with AdshPKCa or with AdshNS (top). Basal isometric force of contraction as well as inotropic responses to Ca2+ (middle) and isoprenaline (bottom). (G) Real-time PCR for PKCα in EHTs (top). Correlation between PKCα mRNA levels and maximal force of contraction in EHTs (bottom). ⁎p b 0.05.

imal Ca2+ stimulated twitch tension (TT after 2.8 mM Ca2+: 0.67 ± 0.05 mN vs. 0.41 ± 0.07 mN in shNS, p b 0.05) and isoprenaline-induced inotropy (TT in the presence of 1 μM iso-

prenaline: 0.55± 0.04 mN vs. 0.36 ± 0.06 mN in shNS, p b 0.05). Real-time PCR from EHTs demonstrated 60% PKCα-mRNA knockdown (Fig. 2G). Notably, PKCα mRNA levels correlated

A. El-Armouche et al. / Journal of Molecular and Cellular Cardiology 43 (2007) 371–376

negatively with the maximal force of contraction in EHTs (Spearman r2 = 0.66, p b 0.01, Fig. 2G). 4. Discussion RNAi is the process of sequence-specific, posttranscriptional gene silencing mediated by double-stranded RNA [4,5]. Target validation using RNAi requires efficient siRNA introduction into cells. A limitation for the use of RNAi in cardiac cells is the low efficiency of liposome-mediated transfection of naked siRNA [11]. This limitation will be even more pronounced for any in vivo application [12,13]. In the present study, we also failed to effectively transfect cultured neonatal rat cardiac myocytes despite testing various conditions. In contrast, we demonstrate the feasibility of shRNA delivery by AV-mediated transfection in both cardiac myocyte monolayers and 3D-EHTs. In the latter, we could provide evidence for the functional consequence of PKCα knockdown in cardiomyocytes. Recent evidence suggests that PKCα acts as a negative regulator of contractility. Thus, (i) adenoviral overexpression of wild-type PKCα reduced contractility in isolated adult myocytes, whereas overexpression of a dominant negative form of PKCα enhanced it [3], (ii) hearts from PKCαdeficient mice were hypercontractile, whereas those of transgenic mice overexpressing PKCα were hypocontractile [3], (iii) pharmacological inhibition of PKCα augmented cardiac contractility in vitro and in vivo in wild-type, but not in PKCα deficient mice [2]. Collectively, these data provide strong evidence implicating PKCα as a pharmacological target for treating heart failure. Our data provide additional evidence supporting the therapeutic concept of PKCα knockdown. Moreover, we demonstrate that the recently described shRNA expression system [7] can be efficiently transfected into cardiac myocytes by an AV vector. Functional consequences of gene silencing of PKCα were determined in EHTs. EHTs represent threedimensional heart-tissue-like cardiac myocyte cultures that allow the determination of standard parameters of contractile function under isometric conditions [9]. They provide cells with a relatively physiological three-dimensional environment, are stable for weeks and are easy to infect with AV with high efficiency [6]. EHTs that had been infected with shPKCαencoding virus showed significantly higher contractile force than those infected with control virus throughout the Ca2+ and isoprenaline concentration–response curve. Intriguingly, there was a significant negative correlation between PKCα mRNA levels and force of contraction in EHTs, supporting the interpretation of a causal relationship. The variable magnitude of the PKCα knockdown is likely due to variations in the overall transfection efficiency of the AV which were in acceptable limits, but still high enough to explain the observed differences. In addition, it is difficult to decide unequivocally by standard epifluorescence microscopy of EHTs (diameter 1 mm) where exactly the green cells are. In this respect, the fraction of transfected cardiac myocytes inside the EHT may have varied from sample to sample.

375

4.1. Limitations This study did not assess the amounts of other PKC isozymes (e.g. PKCβ, PKCδ and PKCε). However, the target sequence for PKCα was carefully chosen not to overlap with other PKC isozymes and compensational changes are not expected in this short-term approach, since, even in PKCα knockout as well as in PKCα overexpressing mice, no compensatory changes in PKC isozymes were observed [3]. The mechanism for augmented contractility after PKCα knockdown in EHTs has not been re-examined since it has been already shown by two groups that PKCα regulates Ca2+ handling by altering the phosphorylation status of protein-phosphatase-inhibitor-1, which in turn suppresses phosphatase-1 activity, thus modulating phospholamban phosphorylation [3,14]. However, we cannot exclude that the increased force in the PKCα-knockdown EHTs both at baseline and over the entire concentration response curves for Ca2+ and isoprenaline involves other effects on the known PKCα targets in cardiac myocytes (e.g. TnI, [15– 17]). Moreover, it could also be due to a structural change of the cardiomyocytes or cardiac tissue formation in EHTs. 5. Conclusions The study demonstrates that AV-delivered shRNAs mediate efficient PKCα knockdown in cardiac myocytes. This was accompanied by improved contractility of cardiomyocytes in EHT supporting the notion that PKCα negatively regulates myocardial contractility and that PKCα knockdown may be a useful strategy to enhance contractile performance in failing hearts. Our data further demonstrate that EHT in conjunction with AV-delivered shRNA is a useful model for target validation. Acknowledgments These studies were supported by the Deutsche Forschungsgemeinschaft (FOR 604 to A.E.A. and T.E.), by the European Commission (EUGeneHeart to T. E.) and by the German Ministry for Education and Research (BMBF 01GN 0520 to W.H.Z.). References [1] Molkentin JD, Dorn II IG. Cytoplasmic signaling pathways that regulate cardiac hypertrophy. Annu Rev Physiol 2001;63:391–426. [2] Hambleton M, Hahn H, Pleger ST, Kuhn MC, Klevitsky R, Carr AN, et al. Pharmacological- and gene therapy-based inhibition of protein kinase Calpha/beta enhances cardiac contractility and attenuates heart failure. Circulation 2006;114:574–82. [3] Braz JC, Gregory K, Pathak A, Zhao W, Sahin B, Klevitsky R, et al. PKCalpha regulates cardiac contractility and propensity toward heart failure. Nat Med 2004;10:248–54. [4] Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 1998;391:806–11. [5] Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 2001;411:494–8. [6] El-Armouche A, Rau T, Zolk O, Ditz D, Pamminger T, Zimmermann WH, et al. Evidence for protein phosphatase inhibitor-1 playing an

376

[7]

[8]

[9]

[10]

[11]

A. El-Armouche et al. / Journal of Molecular and Cellular Cardiology 43 (2007) 371–376 amplifier role in beta-adrenergic signaling in cardiac myocytes. FASEB J 2003; 17:437–9. Laatsch A, Ragozin S, Grewal T, Beisiegel U, Joerg H. Differential RNA interference: replacement of endogenous with recombinant low density lipoprotein receptor-related protein (LRP). Eur J Cell Biol 2004;83:113–20. He TC, Zhou S, da Costa LT, Yu J, Kinzler KW, Vogelstein B. A simplified system for generating recombinant adenoviruses. Proc Natl Acad Sci U S A 1998;95:2509–14. Zimmermann WH, Schneiderbanger K, Schubert P, Didie M, Munzel F, Heubach JF. Tissue engineering of a differentiated cardiac muscle construct. Circ Res 2002;90:223–30. Zolk O, Marx M, Jackel E, El-Armouche A, Eschenhagen T. Betaadrenergic stimulation induces cardiac ankyrin repeat protein expression: involvement of protein kinase A and calmodulin-dependent kinase. Cardiovasc Res 2003;59:563–72. Kasahara H, Aoki H. Gene silencing using adenoviral RNAi vector in vascular smooth muscle cells and cardiomyocytes. Methods Mol Med 2005;112:155–72.

[12] Aigner A. Gene silencing through RNA interference (RNAi) in vivo: strategies based on the direct application of siRNAs. J Biotechnol 2006;124:12–25. [13] Lewis DL, Hagstrom JE, Loomis AG, Wolff JA, Herweijer H. Efficient delivery of siRNA for inhibition of gene expression in postnatal mice. Nat Genet 2002;32:107–8. [14] Rodriguez P, Mitton B, Waggoner JR, Kranias EG. Identification of a novel phosphorylation site in protein phosphatase inhibitor-1 as a negative regulator of cardiac function. J Biol Chem 2006;281:38599–608. [15] Sakthivel S, Finley NL, Rosevear PR, Lorenz JN, Gulick J, Kim S, et al. In vivo and in vitro analysis of cardiac troponin I phosphorylation. J Biol Chem 2005;280:703–14. [16] MacGowan GA, Evans C, Hu TC, Debrah D, Mullet S, Chen HH, et al. Troponin I protein kinase C phosphorylation sites and ventricular function. Cardiovasc Res 2004;63:245–55. [17] Roman BB, Goldspink PH, Spaite E, Urboniene D, McKinney R, Geenen DL. Inhibition of PKC phosphorylation of cTnI improves cardiac performance in vivo. Am J Physiol: Heart Circ Physiol 2004; 286:H2089–95.