Circ Arrhythm Electrophysiol 2014 Barana 861 8

861

Atrial fibrillation (AF) is the most prevalent arrhythmia and the main risk factor associated with myocardial-related

cerebrovascular events.1,2 Extremely fast atrial rates lead to progressive atrial structural and electrical changes (remod-eling) that favor arrhythmia recurrence and maintenance. Fibrosis is one of the most important structural changes asso-ciated with atrial remodeling.1,2 Electrical remodeling is char-acterized by a marked shortening of the atrial action potential duration (APD) and refractoriness because of changes in Ca2+ and K+ channel densities.1–5 Reduction of the density of the L-type Ca2+ current (I

Ca,L), which is generated by chan-

nels composed of α1c or Cav1.2 (encoded by CACNA1C), β2 (CACNB2), and α2δ (CACNA2D) subunits, is a hallmark

of the electrical remodeling.2,3 Importantly, the molecular mechanisms underlying I

Ca,L downregulation have not been

completely elucidated yet. Some authors proposed that it is because of a reduction of CACNA1C and CACNB2 mRNA expression,4,6 whereas others suggested post-transcriptional mechanisms, such as protein dephosphorylation.7

Clinical Perspective on p 868

MicroRNAs (miRNA) are a class of small noncoding RNAs (21–25 nucleotides) that regulate gene expression at the post-transcriptional level by binding to 3′ untranslated regions (3′UTR) of the target mRNA promoting its degra-dation or blocking translation.8 Recent evidence pointed to

© 2014 American Heart Association, Inc.

Circ Arrhythm Electrophysiol is available at http://circep.ahajournals.org DOI: 10.1161/CIRCEP.114.001709

Original Article

Background—Atrial fibrillation is characterized by progressive atrial structural and electrical changes (atrial remodeling) that favor arrhythmia recurrence and maintenance. Reduction of L-type Ca2+ current (I

Ca,L) density is a hallmark of

the electrical remodeling. Alterations in atrial microRNAs could contribute to the protein changes underlying atrial fibrillation–induced atrial electrical remodeling. This study was undertaken to compare miR-21 levels in isolated myocytes from atrial appendages obtained from patients in sinus rhythm and with chronic atrial fibrillation (CAF) and to determine whether L-type Ca2+ channel subunits are targets for miR-21.

Methods and Results—Quantitative polymerase chain reaction analysis showed that miR-21 was expressed in human atrial myocytes from patients in sinus rhythm and that its expression was significantly greater in CAF myocytes. There was an inverse correlation between miR-21 and the mRNA of the α1c subunit of the calcium channel (CACNA1C) expression and I

Ca,L density. Computational analyses predicted that CACNA1C and the mRNA of the β2 subunit of the calcium

channel (CACNB2) could be potential targets for miR-21. Luciferase reporter assays demonstrated that miR-21 produced a concentration-dependent decrease in the luciferase activity in Chinese Hamster Ovary cells transfected with CACNA1C and CACNB2 3′ untranslated region regions. miR-21 transfection in HL-1 cells produced changes in I

Ca,L properties

qualitatively similar to those produced by CAF (ie, a marked reduction of ICa,L

density and shift of the inactivation curves to more depolarized potentials).

Conclusions—Our results demonstrated that CAF increases miR-21 expression in enzymatically isolated human atrial myocytes. Moreover, it decreases I

Ca,L density by downregulating Ca2+ channel subunits expression. These results

suggested that this microRNA could participate in the CAF-induced ICa,L

downregulation and in the action potential duration shortening that maintains the arrhythmia. (Circ Arrhythm Electrophysiol. 2014;7:861-868.)

Key Words: atrial fibrillation ◼ calcium channel ◼ human atrial myocytes ◼ microRNA ◼ miR-21

Received October 11, 2013; accepted June 20, 2014.From the Department of Pharmacology (A.B., M.M., P.D.-G., M.P.-H., I.A., M.N., S.S., J.T., E.D., R.C.), and Instituto de Investigación Sanitaria

Gregorio Marañón (A.B., M.M., P.D.-G., M.P.-H., I.A., M.N., S.S., J.T., E.D., R.C.), School of Medicine, Universidad Complutense de Madrid, Madrid, Spain; and Cardiology and Cardiovascular Surgery Services, Hospital General Universitario Gregorio Marañón, Madrid, Spain (Á.P., Á.P., F.F.-A.)

The Data Supplement is available at http://circep.ahajournals.org/lookup/suppl/doi:10.1161/CIRCEP.114.001709/-/DC1.*Dr Barana and Mr Matamoros contributed equally to this work.Correspondence to Eva Delpón, BPharm, PhD, Department of Pharmacology, School of Medicine, Universidad Complutense de Madrid, 28040-Madrid,

Spain. E-mail [email protected]

Chronic Atrial Fibrillation Increases MicroRNA-21 in Human Atrial Myocytes Decreasing L-Type Calcium Current

Adriana Barana, BSci, PhD;* Marcos Matamoros, BSci;* Pablo Dolz-Gaitón, BSci; Marta Pérez-Hernández, BSci; Irene Amorós, BPharm, PhD; Mercedes Núñez, BPharm, PhD;

Sandra Sacristán; Álvaro Pedraz, MD; Ángel Pinto, MD, PhD; Francisco Fernández-Avilés, MD, PhD; Juan Tamargo, MD, PhD, FESC;

Eva Delpón, BPharm, PhD; Ricardo Caballero, BPharm, PhD

by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from

http://circep.ahajournals.org/lookup/suppl/doi:10.1161/CIRCEP.114.001709/-/DC1

mailto:[email protected]

http://circep.ahajournals.org/






















862 Circ Arrhythm Electrophysiol October 2014

the involvement of miRNAs in the AF-induced ICa,L

down-regulation. In this line, miR-1 and miR-328 bind 3′UTR of CACNA1C and decrease I

Ca,L.9,10 Expression of miR-21 is

increased in plasma samples from patients with chronic AF (CAF) compared with patients in sinus rhythm (SR).11 Moreover, Adam et al,12 showed a 2.5-fold upregulation of miR-21 expression in left atrial appendages from patients with CAF relative to patients in SR. They proposed that miR-21 could participate in atrial fibrosis formation and contribute to the AF-induced structural remodeling.12,13 Indeed, miR-21 is highly expressed in cardiac fibroblasts, where it is implicated in the activation of profibrotic pathways14 and atrial profibril-latory remodeling in experimental models of heart failure.15 However, expression of miR-21 in cardiac myocytes has not been consistently demonstrated, and its putative effects on the electrical remodeling are currently unknown. We hypoth-esized that miR-21 is expressed in human atrial myocytes and upregulated in myocytes from patients with CAF com-pared with those in SR, playing a role in the CAF-induced I

Ca,L decrease. Therefore, this study was undertaken to com-

pare the expression of miR-21 in isolated myocytes from atrial appendages obtained from patients in SR and with CAF and to analyze whether Ca2+ channel subunits are targets for miR-21.

MethodsThe study was approved by the Investigation Committee of the Hospital Universitario Gregorio Marañón (CNIC-13) and conforms to the principles outlined in the Declaration of Helsinki. Each patient gave written informed consent. Clinical data of the patients are in-cluded in Table I in the Data Supplement.

Analysis of the mRNA Expression in Human Atrial MyocytesReal-Time Quantitative Polymerase Chain ReactionRNA was extracted with miRNeasy Micro Kit (Qiagen, UK) from human atrial myocytes enzymatically isolated from right atrial ap-pendages obtained from patients in SR (n=10) and with CAF (n=10) as described.5,16 From each sample, 2 RNA fractions were obtained: a large RNA (>200 nucleotides) fraction and a separate miRNA-en-riched fraction. Quantitative polymerase chain reaction (qPCR) was performed using TaqMan gene and miRNA expression assays (Life Technologies, USA). The cycle to threshold (C

t) values were normal-

ized to 18S rRNA.

miRNA Target PredictionComputational prediction of putative targets for miR-21 was per-formed by using DIANA-microT3.1, TargetMiner, PicTar, mi-RANDA, TargetScan6.2, EIMMo, and RNA22-HSA miRNA target prediction algorithms.

Luciferase Gene Expression Reporter AssayLuciferase activity assays were performed in Chinese Hamster Ovary (CHO) cells transfected with the 3′UTR regions of CACNA1C, CACNB2, and CACNA2D cloned into pMirTarget vector (BlueHeron, USA) and contransfected or not with miR-21 mimic (hsa-miR-21-5p; MC10206), miRNA mimic negative control No. 1, or miR-21 inhibi-tor (MH10206; Life Technologies).

Western Blot AnalysisTo measure Cav1.2 protein expression, a Western blot analysis in HL-1 cells transfected or not with miR-21 mimic was performed by using anti-Cav1.2 antibody (1:1000; Neuromab, USA).

Calcium Current RecordingsCurrents were recorded in human atrial myocytes and HL-1 cells at room temperature using the whole-cell patch-clamp technique

(micropipette resistance <3.5 MΩ).5,16 Series resistance was com-pensated manually and usually ≥80% compensation was achieved. Under our experimental conditions, no significant voltage errors (<5 mV) caused by series resistance were expected with the micropipettes used.

Statistical AnalysisResults are expressed as mean±SEM. Unpaired t test or 1-way analy-sis of variance followed by Newman–Keuls test was used where ap-propriate. In small-size samples (n<15), statistical significance was confirmed by using nonparametric tests. Comparisons between cat-egorical variables were done using Fisher’s exact test. To take into account correlations between multiple levels of within-patient mea-surements, data were analyzed with multilevel mixed-effects models. A value of P<0.05 was considered significant.

Additional details are presented in the Data Supplement.

ResultsmiR-21 Expression Levels Are Increased in Myocytes Isolated From Patients With CAFWe first compared the expression of miR-21 in myocytes enzymatically isolated from atrial appendages obtained from patients in SR (n=10) and with CAF (n=10) by qPCR. Figure 1A shows the ΔC

t values corresponding to miR-21

levels measured in myocytes from patients in SR and with CAF. Importantly, miR-21 was expressed in human atrial myocytes from patients in SR, and as demonstrated by the comparison of the ΔC

t values, miR-21 expression was signifi-

cantly greater in CAF myocytes. Transformation of ΔCt to fold

differences demonstrated that miR-21 expression was ≈3.8× greater in CAF than in SR myocytes (Figure 1D). Because decreased I

Ca,L density is a hallmark of the CAF-induced elec-

trical remodeling,1,2 we measured CACNA1C and CACNB2 mRNA expression in isolated atrial myocytes from patients in SR and with CAF by qPCR. mRNA expression of the target genes was measured in the fractions of RNA >200 nucleo-tides obtained from the same human atrial myocyte samples used for miR-21 expression assay. The results demonstrated that CACNA1C and CACNB2 mRNA expression was ≈>60% and ≈>40%, respectively, in SR than in CAF myocytes (Fig-ure 1B–1D). Interestingly, association studies demonstrated an inverse correlation of miR-21 expression with that of CAC-NA1C in patients in SR and with CAF (Figure 1E) and with I

Ca,L density at +10 mV (Figure 1F), in such a way that the

greater miR-21 expression, the lower CACNA1C expression and I

Ca,L density. Therefore, we were interested in determining

whether miR-21 could modulate the expression of Ca2+ chan-nel subunits. A computational analysis using several algo-rithms revealed CACNA1C and CACNB2 as potential targets for miR-21. Indeed, CACNA1C and CACNB2 3′UTR regions contain a sequence, which is complementary to the seed site of miR-21 (nucleotides 2–8; Figure I in the Data Supplement). On the contrary, 3′UTR region of CACNA2D did not exhibit any complementary sequence to the seed site of miR-21.

miR-21 Directly Binds to the 3′UTR of CACNA1C and CACNB2To analyze whether miR-21 binds to the 3′UTR regions of the subunits forming the cardiac Ca2+ channel, CACNA1C, CACNB2, and CACNA2D 3′UTR regions were cloned into pMirTarget vector, which contains firefly luciferase as a

by guest on December 24, 2015http://circep.ahajournals.org/Downloaded from


Barana et al MicroRNA-21 and L-Type Ca2+ Currents 863

reporter. Cotransfection of miR-21 mimic (15–60 nmol/L) resulted in a concentration-dependent decrease of the lucif-erase activity in CHO cells transfected with CACNA1C and CACNB2 3′UTR regions compared with that measured in sham-transfected cells (Figure 2A and 2B). In both cases, miR-21 mimic at 30 nmol/L concentration reduced luciferase activity by >60%, and thus, this was the concentration selected for the rest of the experiments. The specificity of the miR-21 effect was supported by the absence of changes produced by the negative control miRNA (30 nmol/L). Furthermore, trans-fection of miR-21 inhibitor (antimiR-21; 30 nmol/L) signifi-cantly increased luciferase activity, demonstrating the relief of tonic repression by endogenous miR-21 in these cells.17 As shown in Figure 2C, transfection of miR-21 mimic did not sig-nificantly modify luciferase activity in cells transfected with the CACNA2D 3′UTR region in accordance with the absence of any complementary sequence to the seed site of miR-21.

miR-21 Reduces the Density of L-Type Calcium CurrentsqPCR experiments demonstrated that CAF myocytes dis-played greater miR-21 expression than SR myocytes and that there was a correlation between I

Ca,L density and miR-21

expression. We next compared properties of ICa,L

between SR and CAF by pooling recordings from 52 and 31 cells, respec-tively. Cell capacitance of CAF myocytes was greater than that of SR myocytes (77.9±5.6 versus 58.7±3.8 pF; P=0.005). Currents were recorded by applying 500-ms pulses from −80 mV to potentials between −40 and +50 mV in 5 mV incre-ments, with a prepulse to −30 mV to inactivate the sodium current. Figure 3A and 3B show traces for I

Ca,L recorded at +10

mV in SR and CAF myocytes and current density–voltage curves. As expected, I

Ca,L density was significantly lower in

CAF than in SR myocytes (−2.3±0.3 versus −3.3±0.3 pA/pF at +10 mV; P=0.016), whereas no differences in the activation

Figure 1. A–C, ΔCt values of miR-21 and the mRNA of the α1c subunit of the cal-cium channel (CACNA1C) and the mRNA of the β2 subunit of the calcium chan-nel (CACNB2) measured by quantitative polymerase chain reaction in right atrial myocytes obtained from patients in sinus rhythm (SR) and with chronic atrial fibrilla-tion (CAF). In A–C, Each bar represents the mean±SEM of n≥5. D, Relative expression levels of miR-21, CACNA1C, and CACNB2 in SR and CAF samples. Correlation between miR-21 and CACNA1C ΔCt (E) and ICa,L density at +10 mV (F) measured in right atrial myocytes from patients in SR (black circles) and with CAF (grey circles). In E and F, Each point corresponds to the mean ΔCt and ICa,L density obtained for each patient, and solid lines represent the linear regression to the data.

Figure 2. Normalized luciferase activity in cells expressing the pMir-target luciferase expression reporter vector carrying the 3′ untranslated region of the mRNA of the α1c subunit of the calcium channel (CAC-NA1C) (A), the mRNA of the β2 subunit of the calcium channel (CACNB2) (B), or the mRNA of the α2δ subunit of the calcium channel (CACNA2D) (C) cotransfected with an empty vector (sham-transfected), with increasing concentrations of miR-21 mimic, with a negative control microRNA (NC; 30 nmol/L), or with an miR-21 inhibitor (Anti; 30 nmol/L). Each bar represents the mean±SEM of 6 independent batches of cells for each group. * and ** P<0.05 and P<0.01 vs control, respectively.




and inactivation kinetics and voltage-dependent activation were observed. However, the midpoint of the inactivation curve was significantly shifted to more positive potentials (Figure 3C and 3D; Table).

We next analyzed whether miR-21 transfection in HL-1 cells reproduced the effects on the I

Ca,L induced by electrical

remodeling. HL-1 cells exhibit functional L- and T-type Ca2+ channels, and thus, both I

Ca,L and I

Ca,T can be recorded. HL-1

cells randomly patched were divided into 3 groups depending on the predominant voltage-gated Ca2+ current recorded (Figure II in the Data Supplement). In control conditions (n=54), around 24% of cells did not exhibit any Ca2+ current and ≈39% exhibited a large Ca2+ current, identified as I

Ca,T because it was

completely abolished by NiCl (50 μmol/L) and reached its maximum density at −30 mV (I

Ca,T-predominant cells; Figure

II in the Data Supplement). The rest of the cells (≈37%) exhib-ited a large I

Ca,L that was abolished by nifedipine (1 μmol/L)

and reached its maximum density at +20 mV (Figure II in the Data Supplement) and small or absent I

Ca,T (I

Ca,L-predominant

cells). Transfection of miR-21 mimic did not modify the

proportion of cell phenotypes (P>0.05, n=60; Figure II in the Data Supplement) or cell capacitance (22.9±3.2 pF; P=0.18).

Currents through L-type Ca2+ channels were recorded in I

Ca,L-predominant cells using Ba2+ as the charge carrier (I

Ba).

Figure 4A shows IBa

traces recorded by applying the protocol at the top transfected or not with miR-21 mimic (30 nmol/L). As can be observed in Figure 4B, current density reached its maxi-mum value at +20 mV (n=30) and transfection of miR-21 mimic significantly reduced I

Ba density (n=33; P=0.02) and slowed the

time course of inactivation (P=0.005), without modifying activa-tion kinetics (Figure 4A–4C; Table). This I

Ba density decrease

could be attributed to the reduction of Cav1.2 expression pro-duced by miR-21 as demonstrated by Western blot analysis (Figure 5A and 5B). I

Ba density was not modified by the nega-

tive control miRNA (30 nmol/L; n=15), but was significantly increased by antimiR-21 (30 nmol/L, n=15; Figure 4C), demon-strating a tonic inhibition of I

Ba by endogenous miR-21 in HL-1

cells.18 miR-21 did not modify activation curves (Figure 4D) and shifted the midpoint of the inactivation curve to more posi-tive potentials, without modifying the slope factor (Figure 4D;

Figure 3. A, ICa,L traces recorded by apply-ing the protocol shown at the top in human atrial myocytes from patients in sinus rhythm (SR) and with chronic atrial fibril-lation (CAF). Current density–voltage (B), activation (C), and inactivation (D) curves for ICa,L recorded in SR and CAF myocytes. Each point represents mean±SEM of >30 experiments. *P<0.05 vs SR.

Table. Time- and Voltage-Dependent Properties of ICa,L and IBa Recorded in Human Atrial and HL-1 Cells, Respectively

Maximum Density, pA/pF E

rev, mV

Time Course of Activation* (τ

act), ms

Time Course of Current Decay†

Voltage Dependence of Activation

Voltage Dependence of Inactivation

τf inact

, ms τs inact

, ms Vh, mV K V

h, mV k

ICa,L

SR (n=52) −3.3±0.3 56.1±1.5 1.1±0.1 21.3±6.5 189±20 −4.6±0.6 6.4±0.1 −22.2±1.6 7.2±0.5

CAF (n=31) −2.3±0.3‡ 54.8±2.1 1.2±0.1 21.8±2.3 163±11 −4.0±0.9 6.6±0.2 −17.6±1.1‡ 6.4±0.6

IBa

Control (n=30) −5.8±0.9 73.9±3.1 2.0±0.1 165±12 8.6±1.5 5.7±0.5 −35.7±3.6 8.4±1.4

+miR-21 (n=33) −3.4±0.6§ 74.8±4.0 1.9±0.1 300±48§ 7.5±1.9 5.7±0.3 −25.2±3.9§ 8.3±0.8

Erev

indicates reversal potential; τact

, time constant of activation; τf inact

and τs inact

, fast and slow time constants, respectively, of current decay; Vh and k,

midpoint and slope, respectively, of the conductance voltage and availability curves. CAF, chronic atrial fibrillation; and SR, sinus rhythm.*Activation kinetics was measured by fitting monoexponential functions to current traces recorded at +10 mV (human atrial myocytes) and +20 mV (HL-1). †Inactivation kinetics were measured by fitting biexponential and monoexponential functions to current traces recorded in human atrial myocytes (+10

mV) and HL-1 (+20 mV), respectively. ‡ and § P<0.05 vs SR and control values, respectively. Comparison between ICa,L

recorded from SR and CAF myocytes were made using multilevel mixed-effects models.




Table). Moreover, miR-21 did not significantly modify reactiva-tion kinetics (Figure III in the Data Supplement).

miR-21 Does Not Modify the Density of T-Type Calcium CurrentsTo determine whether the effects of miR-21 are specific to L-type Ca2+ channel subunits, the consequences of miR-21 mimic transfection on I

Ca,T were determined. In this group of

experiments, only ICa,T

-predominant HL-1 cells were included for the analysis. As can be observed, miR-21 did not signifi-cantly modify I

Ca,T density and time- and voltage-dependent

properties (Figure 6A and 6B).

DiscussionThe results presented here demonstrated that CAF upregulates miR-21 in human atrial myocytes, which, in turn, decreases I

Ca,L, but not I

Ca,T density.

miR-21 is Expressed in Human Atrial Myocytes and Is Upregulated by CAFOur qPCR results demonstrated for the first time that miR-21 was expressed in enzymatically isolated atrial myocytes obtained from patients in SR. Some authors suggested that miR-21 was expressed in mouse and rat cardiac myocytes, where it could play a role in cardiac hypertrophy and in the protective effect of ischemic preconditioning against isch-emia-induced cardiac myocyte damage.19 On the contrary, other authors described that miR-21 was not expressed in mouse cardiac cells,14 and thus, its role in cardiac remodel-ing produced in a pressure overload–induced mouse model was attributed to its presence in fibroblasts that affects cardiac myocytes in a paracrine fashion.14 In human atria, miR-21 was found in whole atrial biopsies.12,20,21 However, it is generally accepted that the presence of fibroblasts and nonmyocyte cells in whole biopsies complicates gene expression analysis. From

Figure 4. A, IBa traces recorded by applying the protocol shown at the top in sham-transfected and in miR-21–transfected HL-1 cells. B, Current density–voltage relationships for IBa measured in the absence and presence of miR-21. C, Peak current density for IBa measured in sham-transfected cells and in cells transfected with miR-21 mimic (30 nmol/L), with nega-tive control (NC; 30 nmol/L) or with miR-21 inhibitor (Anti; 30 nmol/L). D, Normalized steady state IBa activation and inactivation curves. In D, solid lines represent the fit of a Boltzmann function to the data. Each point/bar represents mean±SEM of >15 experiments. *P<0.05 vs control.

Figure 5. A, Representative Western blot images showing Cav1.2 expression in sham-transfected (left) or in miR-21–transfected (right) HL-1 cells. GAPDH (bottom) was used as a loading control. B, Densitometric analysis of the Western blots. Results are presented as mean±SEM of 6 inde-pendent batches of cells for each group. *P<0.05 vs control.




our results, we suggest that, besides in fibroblasts, miR-21 is expressed in human atrial myocytes.

Our results also demonstrated that miR-21 expression was increased by 3.8-fold in CAF relative to SR myocytes. Expression of several miRNAs is reduced (eg, miR-1, miR-26, miR-29, miR-30, miR-133, miR-208, and miR-590) or increased (eg, miR-21, miR-328, miR-499, and miR-155) in atrial tissue collected from patients with AF relative to patients in SR.22 Interestingly, CAF-induced increase of miR-21 expres-sion has been a consistent finding in several studies,12,20,21 being one of the most upregulated (3- to 4-fold) miRNAs.20,21 Adam et al12 demonstrated that miR-21 upregulation was cor-related positively with atrial collagen content and increased expression of profibrotic mediators. Therefore, miR-21 was considered to play a critical role in the structural remodeling associated with AF by means of an increased production of fibrosis. Indeed, atrial miR-21 knockdown suppressed atrial fibrosis and AF promotion in a rat model of myocardial infarction.15 Several mechanisms could underlie AF-induced miR-21 expression increase involving post-transcriptional, nontranscriptional, and mainly transcriptional mechanisms.23 In this regard, miR-21 gene is located on chromosome 17, in the 10th intron of the TMEM49 gene, although miR-21 is independently transcribed by its own promoter regions.23 Nuclear factor κB expression increases in response to several stimuli (hypertrophy, oxidative stress, tumor necrosis factor-α, transforming growth factor-β, and the renin–angiotensin sys-tem) involved in structural and electrical AF-induced remod-eling.24 Moreover, an increase in nuclear factor κB expression/activity under oxidative stress correlates with an increase in miR-21 expression, whose promoter region exhibits 5 nuclear factor κB binding sites.25 Therefore, it can be speculated that increase in the nuclear factor κB expression/activity could be responsible, at least partially, of the increase in miR-21 expression observed in CAF patients.

miR-21 Regulates Calcium Channel Subunits Expression and Decreases ICa,LDecrease in the expression of the channels generating I

Ca,L

critically contributes to the abbreviation of APD and atrial refractory period.2,3 Our results demonstrated that there is an inverse correlation between miR-21 and CACNA1C expres-sion and I

Ca,L density, suggesting that miR-21 could target

Ca2+ channel subunits. This hypothesis was supported by the results yielded by the bioinformatic analyses. Alignment of

CACNA1C and CACNB2 3′UTR sequences with miR-21 demonstrated a highly conserved 7mer interaction with the seed site of miR-21 and additional Watson–Crick base pairing matches near the seed site. It has been described that CAC-NA1C is a validated target for miR-1 and miR-328.9,10 Impor-tantly, both miRNAs are predicted to bind to 3′UTR sequence of CACNA1C with complementarity scores lower than those of miR-21. These predictions were confirmed by the concen-tration-dependent decrease in luciferase activity measured in cells transfected with the 3′UTR region of CACNA1C and CACNB2 produced by miR-21. The functional consequences of the interaction were demonstrated by the decreased expres-sion of Cav1.2 protein in HL-1 cells transfected with miR-21. Importantly, patch-clamp studies demonstrated that miR-21 produced changes in I

Ca,L properties qualitatively identical to

those produced by CAF (ie, a marked reduction of ICa,L

den-sity and shifts of the inactivation curves to more depolarized potentials). Current density reduction may be attributed to miR-21 effects on CACNA1C expression. However, voltage-dependent effects would not be expected from the decrease in the expression of the α-subunit and could be attributed to the repression of CACNB2. Our results confirmed previous data showing that miR-21 decreased luciferase activity generated by cells transfected with 3′UTR of CACNB2.26 Thus, consid-ering that coexpression of Cavβ2 with Cav1.2 subunits leads to a substantial increase in I

Ca,L density and a shift of the inac-

tivation curves to more hyperpolarized voltages,27 it is reason-able to propose that the voltage-dependent effects produced by miR-21 could be mediated by the decrease in CACNB2 expression. However, it cannot be ruled out that the presence of ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid in the internal solution can be partially responsible of these voltage-dependent effects, as was proposed previously,28 even when miR-21 effects on I

Ca,L were similar under ethylene

glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid-free conditions (Figure IV in the Data Supplement). To determine the specificity of the interaction, HL-1 cells were transfected with an miRNA mimic negative control, which is a random sequence miRNA mimic molecule with no predicted targets or with an miR-21 inhibitor that specifically binds to and inhibits endogenous miR-21. The absence of effects of miRNA mimic negative control confirmed that miR-21 effects were caused by its interaction with the 3′UTR of the channel subunits. On the contrary, miR-21 inhibitor increased I

Ca,L density, suggest-

ing that endogenous miR-21 was producing a tonic inhibition

Figure 6. Current traces (A) and current density–voltage relationships (B) for ICa,T recorded by applying the protocol shown at the top in sham-transfected or in miR-21–transfected HL-1 cells. In B, each point represents mean±SEM of >20 experiments.




of the current that was relieved by the inhibitor. Further sup-port came from the fact that miR-21 did not modify I

Ca,T, in

accordance with the absence of complementarity between miR-21 and the 3′UTR of CACNA1G, CACNA1H, and CAC-NA1I, which encode the 3 types of α-subunits responsible for I

Ca,T (Cav3.1, Cav3.2, and Cav3.3).

Reduction of ICa,L

density by CAF has been consistently described in patients,2,3,6,7,16 as well as in AF experimental mod-els,4 and it has been proposed to be triggered by Ca2+ overload secondary to the rapid atrial rates.29 The molecular mechanisms underlying I

Ca,L downregulation are complex and can include

impaired Cav1.2 protein trafficking induced by a zinc-binding protein, activation of the Ca2+/calmodulin/calcineurin/nuclear factor of activated T cells system causing transcriptional downregulation of the Cav1.2 α-subunit, Ca2+ channel dephos-phorylation because of serine/threonine protein phosphatase activation, enhanced Cav1.2 α-subunit S-nitrosylation, activa-tion of calpains, ankyrin-B dysfunction, or impaired src kinase activity.2,30,31 Besides all these mechanisms, it has been recently proposed that miRNAs could be involved in AF-induced I

Ca,L

downregulation.10,22 Among all the miRNAs that are upregu-lated in patients with CAF, only miR-328 has been described to regulate Ca2+ channel α1c and β subunits expression.10 miR-328 is increased in right atrial appendages from patients with CAF relative to those in SR. Moreover, overexpression of miR-328 through adenoviral infection in canine atrium and transgenic manipulation in mice enhanced AF susceptibility, decreased I

Ca,L density, and shortened atrial APD as a result of the repres-

sion of CACNA1C expression. It has been described that other miRNAs, such as miR-1 or miR-26 whose expression is decreased in AF, play a role in the AF-induced electrical remod-eling, mainly by targeting Kir2.1 channels.32,33 Both miRNAs are involved in the AF-induced increase of the inward rectifier current and, thus, in the shortening of the APD. Therefore, it can be proposed that all these miRNAs (miR-1, miR-21, miR-26, and miR-328) may contribute to AF-induced electrical remod-eling and eventually to the perpetuation of the arrhythmia.

Our study provides the first description of a putative involvement of miR-21 in the modulation of I

Ca,L and a poten-

tial mechanism linking structural and electrical remodel-ing processes. Recent evidence suggests that structural and electrical remodeling processes may not be independent and would share some common pathways.34,35 In fact, it has been demonstrated that activation of fibrotic signaling pathways via soluble cytokines, such as TGF-β1 and platelet-derived growth factor, modulate Na+, Ca,2+ and K+ currents involved in the control of atrial APD.34,35

Study LimitationsAll samples came from right atrial appendages, which could not be representative of the rest of the atria. Furthermore, miR-21 and Ca2+ channel subunits expression could be influ-enced by age, pharmacological treatment (it has been shown that statin treatment reduces miR-21 expression),12 sex, and underlying cardiac diseases of the patients. The proportion of patients with ischemic heart disease alone and combined with valvular cardiopathy or under statin therapy were equally dis-tributed in both groups, indicating that the changes described here could be attributed to CAF itself.

ConclusionsOur results demonstrated that miR-21 expression increases in myocytes isolated from patients with CAF. Moreover, it decreases I

Ca,L density by downregulating Ca2+ channel sub-

units expression. These results suggested that miR-21 could participate in the CAF-induced I

Ca,L downregulation and in the

APD shortening that maintains the arrhythmia.

AcknowledgmentsWe thank Dr Francisco Pérez-Vizcaíno for Luciferase Assay support and Paloma Vaquero for her technical assistance.

Sources of FundingThis work was supported by Ministerio de Ciencia e Innovación (SAF2011-30112 and SAF2011-30088), Centro Nacional de Investigaciones Cardiovasculares (CNIC-13 and CNIC-08–2009), Instituto de Salud Carlos III (Red HERACLES RD06/0009, Red Investigación Cardiovascular RD12/0042/0011 and PI11/01030), Comunidad Autónoma de Madrid (S2010/BMD-2374), and Sociedad Española de Cardiología grants.

DisclosuresNone.

References 1. Schotten U, Verheule S, Kirchhof P, Goette A. Pathophysiological

mechanisms of atrial fibrillation: a translational appraisal. Physiol Rev. 2011;91:265–325.

2. Wakili R, Voigt N, Kääb S, Dobrev D, Nattel S. Recent advances in the molecular pathophysiology of atrial fibrillation. J Clin Invest. 2011;121:2955–2968.

3. Van Wagoner DR, Pond AL, Lamorgese M, Rossie SS, McCarthy PM, Nerbonne JM. Atrial L-type Ca2+ currents and human atrial fibrillation. Circ Res. 1999;85:428–436.

4. Yue L, Melnyk P, Gaspo R, Wang Z, Nattel S. Molecular mechanisms underlying ionic remodeling in a dog model of atrial fibrillation. Circ Res. 1999;84:776–784.

5. Caballero R, de la Fuente MG, Gómez R, Barana A, Amorós I, Dolz-Gaitón P, Osuna L, Almendral J, Atienza F, Fernández-Avilés F, Pita A, Rodríguez-Roda J, Pinto A, Tamargo J, Delpón E. In humans, chronic atrial fibrillation decreases the transient outward current and ultrarapid component of the delayed rectifier current differentially on each atria and increases the slow component of the delayed rectifier current in both. J Am Coll Cardiol. 2010;55:2346–2354.

6. Brundel BJ, Van Gelder IC, Henning RH, Tieleman RG, Tuinenburg AE, Wietses M, Grandjean JG, Van Gilst WH, Crijns HJ. Ion channel remodeling is related to intraoperative atrial effective refractory periods in patients with paroxysmal and persistent atrial fibrillation. Circulation. 2001;103:684–690.

7. Christ T, Boknik P, Wöhrl S, Wettwer E, Graf EM, Bosch RF, Knaut M, Schmitz W, Ravens U, Dobrev D. L-type Ca2+ current downregulation in chronic human atrial fibrillation is associated with increased activity of protein phosphatases. Circulation. 2004;110:2651–2657.

8. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004;116:281–297.

9. Rau F, Freyermuth F, Fugier C, Villemin JP, Fischer MC, Jost B, Dembele D, Gourdon G, Nicole A, Duboc D, Wahbi K, Day JW, Fujimura H, Takahashi MP, Auboeuf D, Dreumont N, Furling D, Charlet-Berguerand N. Misregulation of miR-1 processing is associated with heart defects in myotonic dystrophy. Nat Struct Mol Biol. 2011;18:840–845.

10. Lu Y, Zhang Y, Wang N, Pan Z, Gao X, Zhang F, Zhang Y, Shan H, Luo X, Bai Y, Sun L, Song W, Xu C, Wang Z, Yang B. MicroRNA-328 con-tributes to adverse electrical remodeling in atrial fibrillation. Circulation. 2010;122:2378–2387.

11. Liu Z, Zhou C, Liu Y, Wang S, Ye P, Miao X, Xia J. The expression levels of plasma micoRNAs in atrial fibrillation patients. PLoS One. 2012;7:e44906.

12. Adam O, Löhfelm B, Thum T, Gupta SK, Puhl SL, Schäfers HJ, Böhm M, Laufs U. Role of miR-21 in the pathogenesis of atrial fibrosis. Basic Res Cardiol. 2012;107:278.




13. Dobrev D. Is altered atrial microRNA-ome a critical contributor to the pathophysiology of atrial fibrillation? Basic Res Cardiol. 2012;107:284.

14. Thum T, Gross C, Fiedler J, Fischer T, Kissler S, Bussen M, Galuppo P, Just S, Rottbauer W, Frantz S, Castoldi M, Soutschek J, Koteliansky V, Rosenwald A, Basson MA, Licht JD, Pena JT, Rouhanifard SH, Muckenthaler MU, Tuschl T, Martin GR, Bauersachs J, Engelhardt S. MicroRNA-21 contributes to myocardial disease by stimulating MAP ki-nase signalling in fibroblasts. Nature. 2008;456:980–984.

15. Cardin S, Guasch E, Luo X, Naud P, Le Quang K, Shi Y, Tardif JC, Comtois P, Nattel S. Role for MicroRNA-21 in atrial profibrillatory fibrot-ic remodeling associated with experimental postinfarction heart failure. Circ Arrhythm Electrophysiol. 2012;5:1027–1035.

16. González de la Fuente M, Barana A, Gómez R, Amorós I, Dolz-Gaitón P, Sacristán S, Atienza F, Pita A, Pinto Á, Fernández-Avilés F, Caballero R, Tamargo J, Delpón E. Chronic atrial fibrillation up-regulates β1-Adrenoceptors affecting repolarizing currents and action potential dura-tion. Cardiovasc Res. 2013;97:379–388.

17. Müller D, Katinger H, Grillari J. MicroRNAs as targets for engineering of CHO cell factories. Trends Biotechnol. 2008;26:359–365.

18. Humphreys DT, Hynes CJ, Patel HR, Wei GH, Cannon L, Fatkin D, Suter CM, Clancy JL, Preiss T. Complexity of murine cardiomyocyte miRNA biogenesis, sequence variant expression and function. PLoS One. 2012;7:e30933.

19. Cheng Y, Zhang C. MicroRNA-21 in cardiovascular disease. J Cardiovasc Transl Res. 2010;3:251–255.

20. Nishi H, Sakaguchi T, Miyagawa S, Yoshikawa Y, Fukushima S, Saito S, Ueno T, Kuratani T, Sawa Y. Impact of microRNA expression in human atrial tissue in patients with atrial fibrillation undergoing cardiac surgery. PLoS One. 2013;8:e73397.

21. Cooley N, Cowley MJ, Lin RC, Marasco S, Wong C, Kaye DM, Dart AM, Woodcock EA. Influence of atrial fibrillation on microRNA expression profiles in left and right atria from patients with valvular heart disease. Physiol Genomics. 2012;44:211–219.

22. Zhao Z, Liu T, Wang X, Li G. MicroRNAs as novel antiarrhythmic targets for atrial fibrillation. Int J Cardiol. 2013;168:e135–e137.

23. Kumarswamy R, Volkmann I, Thum T. Regulation and function of miR-NA-21 in health and disease. RNA Biol. 2011;8:706–713.

24. Gao G, Dudley SC Jr. Redox regulation, NF-kappaB, and atrial fibrilla-tion. Antioxid Redox Signal. 2009;11:2265–2277.

25. Wei C, Li L, Kim IK, Sun P, Gupta S. NF-κB mediated miR-21 regula-tion in cardiomyocytes apoptosis under oxidative stress. Free Radic Res. 2014;48:282–291.

26. Carrillo ED, Escobar Y, González G, Hernández A, Galindo JM, García MC, Sánchez JA. Posttranscriptional regulation of the β2-subunit of car-diac L-type Ca2+ channels by MicroRNAs during long-term exposure to isoproterenol in rats. J Cardiovasc Pharmacol. 2011;58:470–478.

27. Buraei Z, Yang J. The ß subunit of voltage-gated Ca2+ channels. Physiol Rev. 2010;90:1461–1506.

28. Wakili R, Yeh YH, Yan Qi X, Greiser M, Chartier D, Nishida K, Maguy A, Villeneuve LR, Boknik P, Voigt N, Krysiak J, Kääb S, Ravens U, Linke WA, Stienen GJ, Shi Y, Tardif JC, Schotten U, Dobrev D, Nattel S. Multiple potential molecular contributors to atrial hypocontractility caused by atrial tachycardia remodeling in dogs. Circ Arrhythm Electrophysiol. 2010;3:530–541.

29. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibril-lation. Time course and mechanisms. Circulation. 1996;94:2968–2974.

30. Cunha SR, Hund TJ, Hashemi S, Voigt N, Li N, Wright P, Koval O, Li J, Gudmundsson H, Gumina RJ, Karck M, Schott JJ, Probst V, Le Marec H, Anderson ME, Dobrev D, Wehrens XH, Mohler PJ. Defects in ankyrin-based membrane protein targeting pathways underlie atrial fibrillation. Circulation. 2011;124:1212–1222.

31. Greiser M, Halaszovich CR, Frechen D, Boknik P, Ravens U, Dobrev D, Lückhoff A, Schotten U. Pharmacological evidence for altered src kinase regulation of I (Ca,L) in patients with chronic atrial fibrillation. Naunyn Schmiedebergs Arch Pharmacol. 2007;375:383–392.

32. Girmatsion Z, Biliczki P, Bonauer A, Wimmer-Greinecker G, Scherer M, Moritz A, Bukowska A, Goette A, Nattel S, Hohnloser SH, Ehrlich JR. Changes in microRNA-1 expression and IK1 up-regulation in human atrial fibrillation. Heart Rhythm. 2009;6:1802–1809.

33. Luo X, Pan Z, Shan H, Xiao J, Sun X, Wang N, Lin H, Xiao L, Maguy A, Qi XY, Li Y, Gao X, Dong D, Zhang Y, Bai Y, Ai J, Sun L, Lu H, Luo XY, Wang Z, Lu Y, Yang B, Nattel S. MicroRNA-26 governs profibrilla-tory inward-rectifier potassium current changes in atrial fibrillation. J Clin Invest. 2013;123:1939–1951.

34. Kaur K, Zarzoso M, Ponce-Balbuena D, Guerrero-Serna G, Hou L, Musa H, Jalife J. TGF-β1, released by myofibroblasts, differentially regulates transcription and function of sodium and potassium channels in adult rat ventricular myocytes. PLoS One. 2013;8:e55391.

35. Musa H, Kaur K, O’Connell R, Klos M, Guerrero-Serna G, Avula UM, Herron TJ, Kalifa J, Anumonwo JM, Jalife J. Inhibition of platelet-de-rived growth factor-AB signaling prevents electromechanical remodel-ing of adult atrial myocytes that contact myofibroblasts. Heart Rhythm. 2013;10:1044–1051.

CLINICAL PERSPECTIVEChronic atrial fibrillation is associated with atrial structural and electrical changes (atrial remodeling) that favor arrhythmia recurrence and maintenance. Electrical remodeling is characterized by a marked shortening of the atrial action potential duration mainly because of the reduction of L-type Ca2+ current (I

Ca,L) density. Importantly, molecular mechanisms underly-

ing ICa,L

downregulation are not completely understood. MicroRNAs are small noncoding RNAs that regulate expression of the target mRNA, promoting its degradation or blocking translation. MiR-21 is highly expressed in cardiac fibroblasts, where it is implicated in the activation of profibrotic pathways. For this reason, it was proposed to participate in atrial fibrosis formation and contribute to the AF-induced structural remodeling. However, expression of miR-21 in human atrial myocytes has not been explored, and its putative involvement in the electrical remodeling is currently unknown. Our results demonstrated that miR-21 was expressed in human atrial myocytes from patients in sinus rhythm, and its expression was 3.8-fold greater in chronic atrial fibrillation myocytes. Interestingly, miR-21 expression and I

Ca,L density in human atrial

myocytes were inversely correlated. Furthermore, miR-21 transfection in cultured atrial cells produced changes in ICa,L

properties qualitatively similar to those produced by chronic atrial fibrillation (ie, a marked reduction of I

Ca,L density and

shift of the inactivation curves to more depolarized potentials). These results demonstrate that, besides in fibroblasts, miR-21 is expressed in human atrial myocytes and would imply that miR-21 participates in the chronic atrial fibrillation–induced electrical remodeling by contributing to I

Ca,L downregulation.



1

SUPPLEMENTAL MATERIAL

Supplemental Methods

1. Human atrial myocyte isolation

The study was approved by the Investigation Committee of the Hospital Universitario

Gregorio Marañón (CNIC-13) and conforms to the principles outlined in the

Declaration of Helsinki. Each patient gave written informed consent. Data regarding

age, sex, type of surgery, and pharmacological treatment of the patients are included in

Supplemental Table 1. Human right atrial samples were obtained from patients in sinus

rhythm (SR) (n=10) and with chronic atrial fibrillation (CAF) (n=10) that underwent

cardiac surgery. Atrial myocytes were enzimatically isolated following previously

described methods.1,2 Just after excision, atrial appendages were submerged in RNA

later (Qiagen, United Kingdom) to stabilize RNA and immediately placed into chilled

Ca2+-free Tyrode’s solution containing (mmol/L): NaCl 100, KCl 10, KH2PO4 1.2,

MgSO4 5, taurine 50, MOPS 5, 2,3-butanedione monoxime (BDM, 30 mmol/L), and

glucose 20 (pH 7.0 with NaOH), chopped into small pieces (≈1 mm3), and washed 3

times for 3 minutes with Ca2+-free Tyrode’s solution. Tissue pieces were then changed

to Ca2+-free solution containing 254 U/mL collagenase type I (Worthington, USA) and

0.5 mg/mL protease type XXIV (Sigma Chemical Co., United Kingdom) and gently

stirred for 15 minutes. Afterwards, the Ca2+ concentration was raised to 0.2 mmol/L,

and the tissue was stirred for 30 minutes more. Stirring was continued with Tyrode’s

solution (0.2 mmol/L Ca2+) containing only collagenase until rod-shaped striated

myocytes were seen (≈35 minutes). During all these steps, the solutions were

continuously oxygenated with 100% O2 at 37°C. Myocytes were kept in a storage

solution containing (mmol/L): KCl 20, KH2PO4 10, glucose 10, K-glutamate 70, β-

hidroxybutyrate 10, taurine 10, EGTA 10, and albumin 1% (pH 7.4 with KOH)

2

supplemented with RNAprotect Cell Reagent (Qiagen) to avoid RNA degradation in

isolated cells until use for mRNA/microRNA expression assays or electrophysiological

experiments.

2. Analysis of the mRNA expression

Tissue and RNA preparation

The suspension of isolated human atrial myocytes obtained was centrifuged at 500 g for

30 sec to remove non-myocytes (including fibroblasts) and dead or hypercontracted

myocytes.3 Supernatant was discarded and the pellet with the isolated myocytes was

used for subsequent RNA extraction with miRNeasy Micro Kit (Qiagen) according to

the manufacturer’s instructions. This kit enabled to obtain two fractions of RNA from

each sample: a large RNA (>200 nucleotides) fraction and a separate miRNA-enriched

fraction. Large RNA was quantified by spectrophotometry (= 260 nm) and purity of

the samples was verified by the 260/280 ratio with a Nano-Drop 2000 (Thermo

Scientific, USA).

Real-time quantitative polymerase chain reaction

RT-PCR was performed using the High Capacity cDNA Reverse Transcription kit and

the TaqMan MicroRNA Reverse Transcription Kit for large RNA and miRNA-enriched

fractions, respectively, following manufacturer instructions (Life Technologies, USA).

In both cases, the resulting cDNA template was subjected to real-time quantitative PCR

(qPCR) using Taqman-based gene and microRNA expression assays, TaqMan Fast

Universal PCR Master Mix (Life Technologies) and the 7900HT Fast Real-Time PCR

System (Life Technologies). Gene expression analysis was performed using the

TaqMan Gene Expression Assays Hs00167681_m1 and Hs00167861_m1 for target

genes 1c (CACNA1C) and 2 (CACNB2) subunits of the L-type Ca2+ channel,

3

respectively and the TaqMan MicroRNA Assay hsa-miR-21-5p for target miR-21 (Life

Technologies). The eukaryotic 18S rRNA endogenous control was used as the

normalization gene. Each sample was run in triplicates and non-template control to test

for contamination of assay reagents was also included in the plate. Moreover, three

different controls aimed at detecting genomic DNA contamination in the RNA sample

or during the RT or qPCR reactions were always included: a RT mixture without

reverse transcriptase, a RT mixture including the enzyme but no RNA, negative control

(reaction mixture without cDNA template). The data were collected and analyzed using

One-Step Software (Life Technologies). The obtained cycle to threshold (Ct) values

were normalized to the 18S rRNA. The Ct values are based on a log scale and were

transformed to delta Ct (∆Ct) values by subtracting the value corresponding to the gene

of interest from that of 18S. To compare CAF vs SR expression differences, the

respective data were transformed from ∆Ct values to equivalent fold differences using

the following equation, previously used for the same purposes: Fold Difference (mean

∆CtSR – mean ∆CtCAF) = 2(mean ∆CtSR – mean ∆CtCAF). The standard errors were omitted due

to transformation of the data and the subsequent loss of meaning.2

3. MicroRNA target prediction Computational prediction of putative targets for miR-21 was performed by using seven

established miRNA target prediction algorithms: DIANA-microT3.1, TargetMiner,

PicTar, miRANDA, TargetScan6.2, EIMMo, and RNA22-HSA. Most of the algorithms

(6 out 7) predicted CACNA1C and CACNB2 as putative targets for miR-21.

4. Luciferase gene expression reporter assay For luciferase reporter assays, CHO cells were seeded in 96-well plates and cultured in

Ham-F12 medium supplemented with 10% fetal bovine serum, 100 U/mL penicillin,

4

and 100 µg/mL streptomycin at 37ºC and 5% CO2.4 Cells were transfected with the 3’

UTR regions of the subunits forming the human cardiac L-type Ca2+ channel,

CACNA1C, CACNB2, and CACNA2D cloned into pMirTarget vector (BlueHeron,

USA), which contains firefly luciferase as a reporter. Since 3’ UTR of CACNA1C is

extremely large (6750 nt), a 1500-nt fragment comprising the region predicted to bind

miR-21 was cloned. Cotransfections were performed with 15–60 nmol/L miR-21 mimic

(hsa-miR-21-5p; MC10206; Life Technologies), 30 nmol/L miRNA mimic negative

control #1 (Life Technologies) or miR-21 inhibitor (MH10206; Life Technologies).

These mimics are chemically modified to avoid off-target effects induced by the

passenger strand. All transfections were performed by using Lipofectamine 2000

(Invitrogen, USA) according to the manufacturer’s instructions. Luciferase activity

assays were performed 48 hours after transfection using the Steady-Glo Luciferase

Assay System (Promega, USA) and a Fluoroskan Ascent Microplate Fluorometer

(Thermo Scientific). Luciferase activity was normalized to sample protein

concentration. All reporter assays were performed in triplicate.

5. HL-1 cell culture and transfection HL-1 cells were cultured in gelatin/fibronectin-coated dishes at 37ºC in an atmosphere

of 5% in air, with humidity of ≈95%, as previously described.5 Cells were grown in

Claycomb medium (Sigma) supplemented with 10% fetal bovine serum, 2 mmol/L L-

glutamine, 100 mol/L norepinephrine, 100 U/mL penicillin, and 100 µg/mL

streptomycin. Medium was changed every 24-48 h and when cells reached high

confluency at 3-4 days, the cultures were split 1 to 3 using a brief trypsin treatment. HL-

1 cells were transfected with miR-21 mimic, miRNA mimic negative control #1 or miR-

21 inhibitor (all at 30 nmol/L) by using Lipofectamine 2000, according to manufacturer

instructions. In a different group of experiments FAM dye-labeled synthetic miRNA

5

mimic (Life Technologies) was transfected to allow evaluation of the transfection

efficiency. In all cases, transfection efficiencies were greater than 85%. For culture and

transfections, 60 mm culture dishes were used. Preliminary qPCR experiments

demonstrated that transfection of miR-21 mimic increased more than 10-fold miR-21

levels in HL-1 cells. Twenty four-forty eight h after transfection, cells were removed

from the dish by trypsinization (1%, 37ºC for 5 min).

6. Western-blot analysis Detection of Cav1.2 proteins was performed in HL-1 cells transfected or not with miR-

21 mimic (30 nmol/L) by Western blot following previously described procedures.6 HL-

1 cells were homogenized in a non-denaturing solution (in mmoL/L): Tris-HCl 50,

NaCl 500, and 1% NP-40, 0.5% sodium-deoxicolate and protease inhibitor cocktail

(Sigma). Nuclei and cell debris were removed by centrifugation at 10000 g for 20 min

at 4ºC. The total protein amount of the extracts was calculated with the bicinchoninic

acid method (BCA, Pierce, USA) and each extract was then adjusted to 0.6 mg/mL of

protein. Afterwards, samples were separated on denaturing SDS polyacrylamide 8%

gels, transferred to nitrocellulose membranes, blocked with 5% nonfat dried milk in

PBS-Tween, and incubated with anti-Cav1.2 antibody (1:1000; Neuromab, USA)

overnight at 4ºC and then for 1 h with a peroxidase-conjugated goat anti-mouse

secondary antibody (1:5000; Jackson Immunoresearch, USA). Membranes were washed

three times with PBS-Tween before adding primary and secondary antibodies. Protein

expression was detected by chemiluminescence (ECL, General Electric Healthcare,

USA). To ensure equal protein loading, GADPH expression was determined by using

GAPDH antibody (1:1000; Sigma).

7. Calcium current recordings

6

Recording techniques

A small aliquot of a suspension containing human atrial myocytes or HL-1 cells was

placed in a 0.5 mL chamber mounted on the stage of an inverted microscope (Nikon

TMS, Nikon Co., Japan). After settling to the bottom of the chamber, cells were

perfused at 1 mL/min with external solution (see composition below). Currents were

recorded at room temperature (21-23ºC) using the whole cell patch-clamp technique

using an Axopatch-200B patch clamp amplifier (Molecular Devices, USA).1,2,4

Recording pipettes were pulled from 1.0 mm o.d. borosilicate capillary tubes (GD1,

Narishige Co., Ltd, Japan) using a programmable patch micropipette puller (Model P-

2000 Brown-Flaming, Sutter Instruments Co., USA) and were heat-polished with a

microforge (Model MF-83, Narishige). Micropipette resistance was kept below 3.5 MΩ

when filled with the internal solution and immersed in the external solution. The

capacitive transients elicited by symmetrical 10 mV steps from 0 mV were recorded at

50 kHz (filtered at 10 kHz) for subsequent calculation of capacitative surface area,

access resistance and input impedance. In all the experiments, series resistance was

compensated manually by using the series resistance compensation unit of the Axopatch

amplifier, and ≥80% compensation was achieved. In myocytes from SR patients, mean

maximum Ca2+ current (ICa,L) amplitude at +10 mV, uncompensated access resistance,

and capacitance averaged -190±17 pA, 3.3±0.4 MΩ, and 58.7±3.8 pF (n=52),

respectively. In HL-1 cells, mean maximum ICa,L amplitude at +20 mV, uncompensated

access resistance, and capacitance averaged -88.2±18.7 pA, 3.1±0.7 MΩ, and 20.6±1.8

pF (n=20), respectively. Thus, under our experimental conditions no significant voltage

errors (<5 mV) due to series resistance were expected with the micropipettes used. In all

cases currents were filtered at half the sampling frequency and stored on the hard disk

of a computer for subsequent analysis. To record ICa,L in human atrial myocytes and T-

7

type Ca2+ current (ICa,T) and ICa,L in HL-1 cells, the external solution contained

(mmol/L): Tetraethylammonium (TEA)-Cl 137, CaCl2 1, MgCl2 0.5, HEPES 10, and

glucose 10, (pH 7.4 with CsOH). Recording pipettes were filled with an internal

solution containing (mmol/L): CsCl 125, TEA-Cl 20, MgATP 5, phosphocreatine 3.6,

HEPES 10, and EGTA 10 (pH 7.2 with CsOH).2,7 To increase current amplitude and

eliminate Ca2+-induced Ca2+ inactivation, Ba2+ was used as charge carrier (IBa).

Therefore, for these experiments the external solution contained (mmol/L): N-Methyl-

D-glucamin (NMDG) 137, CsCl 12, BaCl2 20, MgCl2 1, HEPES 10, and glucose 10 (pH

7.4 with HCl).6 Under these conditions current amplitudes were stable during the time

of recordings. In some experiments, IBa was recorded in the absence of EGTA in the

internal solution.

Pulse protocols and analysis

To record ICa,L in human atrial myocytes, the holding potential was maintained at -80

mV and the cycle time for any protocol was 10 s. The protocol to obtain current-voltage

relationships consisted of 500-ms pulses that were imposed in 5 mV increments

between -40 and +50 mV. The INa was inactivated by the application of a 50-ms

prepulse to -30 mV. The protocol to obtain current-voltage relationships for ICa,T and

ICa,L consisted of 500-ms pulses in 10 mV increments from -80 mV to potentials ranging

-70 and +70 mV. However, the protocol to obtain current-voltage relationships for IBa

consisted of 500-ms pulses in 10 mV increments from -30 mV to potentials ranging -40

and +70 mV.

In each experiment, current amplitude (measured as the difference between peak current

and zero current level) was normalized to membrane capacitance to obtain current

density. Activation curves for ICa,L and IBa were constructed plotting the normalized

8

conductance as a function of the membrane potential. The conductance was estimated

by the equation:

G=I/(Vm-Erev)

where G is the conductance at the test potential Vm, I represents the current amplitude at

Vm, and Erev is the reversal potential. To determine the Erev, the current density-voltage

relationships were fitted to a function of the form:

I=(Vm-Erev)*Gmax*(1+exp[Vm-Vh]/k)-1

where I is the current amplitude elicited at the test potential Vm, Gmax is the maximum

conductance, and k is the slope factor. The fit of a Boltzmann function to the data

yielded the midpoint (Vh) and the slope (k) of the activation curve.

To obtain the inactivation curves for human atrial ICa,L and IBa recorded in HL-1 cells, a

two-step protocol was used. For ICa,L the first 500-ms conditioning pulse from –80 mV

to potentials between –90 and +50 mV was followed by a 500-ms test pulse to +10 mV.

For IBa recorded in HL-1 cells the first 500-ms conditioning pulse from –30 mV to

potentials between –50 and +20 mV was followed by a 500-ms test pulse to +20 mV. In

both cases, inactivation curves were constructed by plotting the current amplitude

obtained with the test pulse normalized to the largest current, as a function of the

voltage command of the conditioning pulse. Thereafter, a Boltzmann function was fitted

to the data to obtain Vh and k values of the inactivation curve. In order to describe the

time course of current activation and decay, an exponential analysis was used as an

operational approach, fitting current traces with an equation of the form:

y=C+A*exp(-t/τ)

where τ is the time constant, whereas A is the amplitude of the exponential, and C is the

baseline value. To analyze the recovery from inactivation for IBa two 500-ms pulses (P1

and P2) from -30 to +20 mV were applied at increasing coupling intervals (5-4000 ms).

9

Afterwards, a monoexponential function was fitted to the data to obtain the time

constant of recovery from inactivation.

8. Statistical analysis

Results are expressed as mean±SEM. Unpaired t-test or one-way ANOVA followed by

Newman-Keuls test were used where appropriate. In small-size samples, statistical

significance was confirmed by using nonparametric tests. Comparisons between

categorical variables were done using Fisher´s exact test. To take into account

correlations between multiple levels of within-patient measurements, data were

analyzed with multilevel mixed-effects models. A value of P<0.05 was considered

significant.

10

Supplemental Table 1. Characteristics of the patients whose samples were used for

qPCR experiments and ICa,L recordings.

SR CAF

Patients (n) 10 10 Mean age (years) 59±2 69±4* Male/Female (n) 8 / 2 5 / 5 Surgery

Valve surgery (n) 3 5 CABG surgery (n) 6 0 Combined (n) 1 5

Cardiopathy Valvular (n) 3 5 Ischemic (n) 6 0 Combined (n) 1 5 Hypertension (n) 10 6 Diabetes mellitus (n) 4 3 Dyslipidemia (n) 7 4 Left atrial diameter (cm) 3.4±0.1 5.0±0.1* Ejection fraction (%) 53.7±2.2 56.4±2.4 NYHA functional class I (n) 3 2 II (n) 7 7 III (n) 0 1 IV (n) 0 0 Creatinine (mg/dL) 0.88±0.05 1.05±0.1 Treatment

Beta blockers (n) 7 8 ACE inhibitors/ARBs (n) 9 2 Statins (n) 4 3 Acetylsalicylic acid (n) 3 1 Digoxin (n) 0 4 Calcium antagonists (n) 0 3 Diuretics (n) 0 4 Spironolactone (n) 1 0 Oral anticoagulants (n) 0 10 Class I Antiarrhythmic drugs (n) 0 0 Class III Antiarrhythmic drugs (n) 0 1

ACE, angiotensin converting enzyme; ARBs, angiotensin II type 1 receptor blockers;

CABG, coronary artery bypass grafting. * P<0.05 vs SR patients.

11

Expanded results

Supplemental Figures

Supplemental Figure 1. Prediction of miR-21 binding to 3’UTR of human CACNA1C

(top) or CACNB2 (bottom) performed by using miRANDA prediction algorithm.

Computational analysis using several algorithms revealed CACNA1C and CACNB2 as

potential targets for miR-21. Indeed, CACNA1C and CACNB2 3’UTR regions contain

a sequence, which is complementary to the seed site of miR-21 (nucleotides 2-8).

Perfect Watson-Crick-base pair complementarity between nucleotides 2673 to 2678 of

CACNA1C and 1178 to 1183 CACNB2 and nucleotides 2 to 7 of miR-21 were

established together with GU wobbles with nucleotide 8 in miR-21 (7mer binding).

Furthermore, in both cases additional Watson-Crick-base pair matches elsewhere were

also established that reinforce complementarity. In addition, eight 6mer interaction sites

between CACNA1C 3’UTR and miR-21 were also found. On the contrary, 3’UTR

region of CACNA2D did not exhibit any complementary sequence to the seed site of

miR-21.

Human CACNA1C 3’UTR0 bp 6750 bp

CACNA1C 3’UTR 2659: 5’ ggcACAU-AUAUAAGUAAGCUa 3’|||| | : :||||||

miR-21 3’ aguUGUAGUCAGACUAUUCGAu 5’

2659 bp

Human CACNB2 3’UTR0 bp 2035 bp

CACNB2 3’UTR 1157: 5’ ugAGCAGUCAGGGAAUGUGAGUAAGCUu 3’|:|| |||| | || :||||||

miR-21 3’ agUUGU-AGUC----AGAC-UAUUCGAu 5’

1157 bp

Supplemental Figure 1

Seed site

Seed site

12

Supplemental Figure 2. (A and B) ICa,T (A) and ICa,L (B) traces obtained by applying

the protocol shown at the top in HL-1 cells perfused or not with 50 mol/L NiCl (A) or

1 mol/L nifedipine (B). (C) Percentage of HL-1 cells that exhibited ICaT- or ICaL-

predominant patterns and that of cells that did not exhibit any Ca2+ current in sham-

(black) or in miR-21- transfected cells (white). (D) Current density-voltage relationships

for ICa,T and ICa,L. In D, each point represents mean±SEM of ≥20 experiments.

Currents were recorded in randomly selected HL-1 cells by applying 500-ms pulses

from -80 mV to potentials ranging -70 and +70 mV. In control conditions (n=54),

around 24% of cells did not exhibit any Ca2+ current and 39% exhibited a large Ca2+

current, which was identified as ICa,T since it was completely abolished by NiCl (50

mol/L) and reached its maximum density (-7.9±1.5 pA/pF) at -30 mV (ICa,T-

predominant cells) (Panels A and D). The rest of the cells (≈37%) exhibited a large ICa,L

that was completely inhibited by nifedipine (1 mol/L) and reached its maximum

density (-4.6±1.2 pA/pF) at +20 mV (Panels B and D) and small or absent ICa,T (ICa,L-

predominant cells). These percentages are in reasonable agreement with those

previously described for these cells7 and transfection of miR-21 mimic did not modify

them (P>0.05, n=60) (Panel C).

B

+70 mV

-70 mV-80 mV

-80 -40 40 80

-10

-6

-2

2

ICa d

en

sity (pA

/pF

)

ICa,L

ICa,T

Membrane potential (mV)D

Control NiCl50 μmol/L

200

pA

100 ms

C

Control Nifedipine1 μmol/L

100

pA

100 ms


0

10

20

30

40

50

ICa,L ICa,T None

Cel

ls (

%)

A

13

Supplemental Figure 3. Recovery from inactivation data for IBa recorded by applying

the protocol shown in the inset in sham- or in miR-21 transfected HL-1 cells. The solid

lines represent the best fit of a monoexponential function to the data points. Each point

represents the mean±SEM of >30 experiments.

In this group of experiments, the effects of miR-21 transfection on the IBa reactivation

were analyzed. As can be observed, the time course of current reactivation in the

presence of miR-21 was not different from that measured in control conditions

(re=450.1±20.6 vs. 436.6±51.6 ms; P>0.05).

0 1000 2000 3000 40000.0

0.2

0.4

0.6

0.8

1.0Control miR-21

Coupling interval (ms)

P2/

P1

+20 mV

-30 mV∆t

P1 P2


14

Supplemental Figure 4. (A) Current density-voltage relationships for IBa measured

under EGTA free conditions in the absence and presence of miR-21. (B) Normalized

steady-state IBa inactivation curves. In this panel, continuous lines represent the fit of a

Boltzmann equation to the data. Each point represents mean±SEM of >8 experiments.

*P<0.05 vs Control.

This group of experiments was conducted to determine the effects of miR-21

transfection on IBa recorded in the absence of EGTA in the internal solution. Under

these conditions, current-density reached its maximum value at +20 mV (-3.7±0.6

pA/pF, n=10) and transfection of miR-21 mimic significanly reduced IBa density (-

1.7±0.4 pA/pF at +20 mV) (Panel A) and slowed time course of current decay (inact

MIR21=143.8±8.6 ms vs. inact CON=81.5±18.8 ms) (P<0.05, n=8). The analysis of the

voltage dependence of activation and inactivation (Panel B) revealed that miR-21 did

not modify conductance curves (Vh= 10.3±1.8 mV and k= 5.4±0.5). However, it shifted

the midpoint of the inactivation curves from -20.1±2.8 to -11.6±3.2 mV, without

modifying the slope of the curves (11.3±1.0). These results indicated that under EGTA-

free conditions, miR-21 effects on ICa,L are identical to those observed in the presence of

EGTA.

-40 40 80

-5

-3

-1

1

miR-21Control

** *

*

IBa d

en

sity (pA

/pF

)

Membrane potential (mV)

-60 -20 20 600.0

0.2

0.4

0.6

0.8

1.0

ControlmiR-21

Membrane potential (mV)

Nor

mal

ized

I Ba


BA

15

Suplemental references

1. Caballero R, de la Fuente MG, Gómez R, Barana A, Amorós I, Dolz-Gaitón P,

Osuna L, Almendral J, Atienza F, Fernández-Avilés F, Pita A, Rodríguez-Roda J,

Pinto A, Tamargo J, Delpón E. In humans, chronic atrial fibrillation decreases the

transient outward current and ultrarapid component of the delayed rectifier current

differentially on each atria and increases the slow component of the delayed rectifier

current in both. J Am Coll Cardiol. 2010;55:2346-2354.

2. González de la Fuente M, Barana A, Gómez R, Amorós I, Dolz-Gaitón P, Sacristán

S, Atienza F, Pita A, Pinto Á, Fernández-Avilés F, Caballero R, Tamargo J, Delpón

E. Chronic atrial fibrillation up-regulates β1-Adrenoceptors affecting repolarizing

currents and action potential duration. Cardiovasc Res. 2013;97:379-388.

3. Kaur K, Zarzoso M, Ponce-Balbuena D, Guerrero-Serna G, Hou L, Musa H, Jalife J.

TGF-β1, released by myofibroblasts, differentially regulates transcription and

function of sodium and potassium channels in adult rat ventricular myocytes. PLoS

One. 2013;8:e55391.

4. Amorós I, Dolz-Gaitón P, Gómez R, Matamoros M, Barana A, González de la

Fuente M, Núñez M, Pérez-Hernández M, Moraleda I, Gálvez E, Iriepa I, Tamargo

J, Caballero R, Delpón E. Propafenone blocks human cardiac Kir2.x channels by

decreasing the negative electrostatic charge in the cytoplasmic pore. Biochem

Pharmacol. 2013;86:267-278.

5. Claycomb WC, Lanson NA Jr, Stallworth BS, Egeland DB, Delcarpio JB, Bahinski

A, Izzo NJ Jr. HL-1 cells: a cardiac muscle cell line that contracts and retains

phenotypic characteristics of the adult cardiomyocyte. Proc Natl Acad Sci U S A.

1998;95:2979-2984.

16

6. Tsai CT, Wang DL, Chen WP, Hwang JJ, Hsieh CS, Hsu KL, Tseng CD, Lai LP,

Tseng YZ, Chiang FT, Lin JL. Angiotensin II increases expression of alpha1C

subunit of L-type calcium channel through a reactive oxygen species and cAMP

response element-binding protein-dependent pathway in HL-1 myocytes. Circ Res.

2007;100:1476-1485.

7. Xia M, Salata JJ, Figueroa DJ, Lawlor AM, Liang HA, Liu Y, Connolly TM.

Functional expression of L- and T-type Ca2+ channels in murine HL-1 cells. J Mol

Cell Cardiol. 2004;36:111-119.

Fernández-Avilés, Juan Tamargo, Eva Delpón and Ricardo CaballeroAmorós, Mercedes Núñez, Sandra Sacristán, Álvaro Pedraz, Ángel Pinto, Francisco

Adriana Barana, Marcos Matamoros, Pablo Dolz-Gaitón, Marta Pérez-Hernández, IreneL-Type Calcium Current

Chronic Atrial Fibrillation Increases MicroRNA-21 in Human Atrial Myocytes Decreasing

Print ISSN: 1941-3149. Online ISSN: 1941-3084 Copyright © 2014 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleCirculation: Arrhythmia and Electrophysiology

doi: 10.1161/CIRCEP.114.0017092014;7:861-868; originally published online August 8, 2014;Circ Arrhythm Electrophysiol.

http://circep.ahajournals.org/content/7/5/861World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circep.ahajournals.org/content/suppl/2014/08/08/CIRCEP.114.001709.DC1.htmlData Supplement (unedited) at:

http://circep.ahajournals.org//subscriptions/

is online at: Circulation: Arrhythmia and Electrophysiology Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Answer

Permissions and Rights Question andunder Services. Further information about this process is available in thepermission is being requested is located, click Request Permissions in the middle column of the Web pageClearance Center, not the Editorial Office. Once the online version of the published article for which

can be obtained via RightsLink, a service of the CopyrightCirculation: Arrhythmia and Electrophysiologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


http://circep.ahajournals.org/content/7/5/861

http://circep.ahajournals.org/content/suppl/2014/08/08/CIRCEP.114.001709.DC1.html

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circep.ahajournals.org//subscriptions/


Documents

Circ Arrhythm Electrophysiol 2014 Barana 861 8