MiR-486-3p and MiR-938—Important Inhibitors of Pacemaking Ion Channels and/or Markers of Immune Cells

The sinus node (SN) is the heart’s primary pacemaker and has a unique expression of pacemaking ion channels and immune cell markers. The role of microribonucleic acids (miRNAs) in control of ion channels and immune function of the sinus node is not well understood. We have recently shown that hsa-miR-486-3p downregulates the main pacemaking channel HCN4 in the SN. In addition, we recently demonstrated that immune cells are significantly more abundant in the SN compared to the right atrium. The aim of this study was to validate the previously predicted interactions between miRNAs and mRNAs of key Ca2+ ion channels (involved in peacemaking) and mRNA of TPSAB1—(a mast cells marker) using luciferase assay. We now show that miR-486 significantly downregulates Cav1.3, Cav3.1, and TPSAB1-mediated luciferase activity, while miR-938 significantly downregulates only TPSAB1-mediated luciferase activity. This makes miR-486-3p a potential therapeutic target in the treatment of SN dysfunctions.


Introduction
As part of the cardiac conduction system (CCS), the sinus node (SN) is the primary pacemaker of the heart and is located at the junction of the superior vena cava and right atrium. The SN is a crescent-shaped structure that extends along the crista terminalis [1,2]. Compared to the surrounding working myocardium, the SN has distinctive molecular and functional properties-owing to its unique expression of ion channels and Ca 2+handling proteins, both responsible for the membrane voltage and Ca 2+ clocks-as we have previously described [1].
Hyperpolarization-activated cyclic nucleotide-gated channels 1 and 4 (HCN1, HCN4) are the key ion channels responsible for the SN's myocytes' diastolic depolarization also known as the pacemaker potential, with HCN4 being the main isoform that is highly expressed in human SN [3,4]. The Ca v 3.1 ion channel carries the T-type calcium current (I Ca,T ) at the late diastole phase of the pacemaker potential, and Ca v 1.3 carries the L-type calcium current (I Ca,L ) at the upstroke phase of the action potential [3]. We have previously shown that these key ion channels are highly expressed in the human SN, compared to the atrial muscle [1,4]. The SN dysfunction of these ion channels contribute to the development of SN-related rhythm abnormalities such as bradycardia, tachycardia, sinus arrest, etc. [3,5].
Small noncoding RNA molecules (miRNAs) have been extensively studied and are known to regulate the post-transcriptional expression of protein-coding genes, through inhibition [6] by binding to complimentary 3 prime untranslated region (3 UTR) of their target messenger-ribonucleic acids (mRNAs). This regulation plays key roles in the development of cardiac diseases such as cardiac arrhythmia, hypertrophy, fibrosis, and heart failure [7][8][9].
To further our understanding of the SN's pacemaking and immune response function, the aim of this work was to validate the previously predicted interactions between six miRNAs and key ion channels, transcription factors, immune cell markers, and collagen using luciferase assay experiments. A reduction in bioluminescence indicated that the inhibition of miRNA's target is taking place. Renilla reporter gene is used to normalize values produced by the luciferase reporter gene. We found that out of the seven microRNAs that are predicted to inhibit key pacemaking channels and mast cell marker, some microRNAs did or did not downregulate their predicted targets.

Materials and Methods
Main methods are summarized in Figure 1.

MiRNAs
MiRNA mimics were ordered from the Dharmacon Cherry Pick library (Horizon Discovery) and arrived in a powder form in a 96-well plate. This was then resuspended in 1X siRNA buffer (diluted from 5X buffer Dharmacon using RNAse-free water). All miRNAs were diluted to a final concentration of 5 µM. Plates were spun down, at a speed of 2000 rpm for 30 s, before adding 20 µL of 1X siRNA buffer to each well and storing at −20 • C. As one of the control experiments, scrambled miRNA (a nonfunctional miRNA) was used.  Figure 2. Further information regarding predicted interactions between miRNAs and mRNA is shown in in Table 1. values produced by the luciferase reporter gene. We found that out of the seven mi-croRNAs that are predicted to inhibit key pacemaking channels and mast cell marker, some microRNAs did or did not downregulate their predicted targets.

Materials and Methods
Main methods are summarized in Figure 1.  MiRNA mimics were ordered from the Dharmacon Cherry Pick lib covery) and arrived in a powder form in a 96-well plate. This was then siRNA buffer (diluted from 5X buffer Dharmacon using RNAse-free w were diluted to a final concentration of 5 µM. Plates were spun down, rpm for 30 s, before adding 20 µL of 1X siRNA buffer to each well and As one of the control experiments, scrambled miRNA (a nonfunctional    [4,10]. Significant change in the expression of mRNA and mi p ≤ 0.05. Red represents significantly more expressed mRNA in the SN vs. righ

Plasmid Amplification
In total, 1 µL of plasmid was taken up in 20 µL of supercompetent cells (XL1-Blue, Agilent Technologies) and placed on ice for 3 min before heat-shocking for 30 s. The solution was placed on ice again before spreading 21 µL solution on an agar gel plate (containing carbenicillin) and incubating at 37 • C overnight. Then, a single bacterial colony transfected with the plasmid was added to a solution containing 3 mL sterile LB medium and 3 µL carbenicillin (100 µg/µL) overnight at 37 • C, shaking at a speed of 120 rpm. Following this, 1 mL of the resulting culture was mixed with a solution containing 200 mL LB medium and 200 µL carbenicillin (100 µg/µL) overnight at 37 • C, shaking at a speed of 120 rpm. The remaining 2 mL of culture was added to 200 µL 50% glycerol and stored at −80 • C for future use. The next day, the plasmid DNA was extracted and purified using Purelink Plasmid Kit (Thermo Fisher Scientific, Altrincham, UK), according to manufacturer's protocol. Concentration of plasmids was determined using a NanoDrop ND-1000 spectrophotometer (ThermoScientific, Altrincham, UK). The plasmid concentrations ranged between 0.3 and 2.3 µg/ µL. Table 1. Predicted microRNA-mRNA interactions. MicroRNA expression data are based on our previous studies [4,10]. Significant change in the expression of mRNA and microRNA is classed as p ≤ 0.05. Red represents significantly more expressed mRNA in the SN vs. right atrium. Green represents significantly less expressed microRNA in the SN vs. right atrium. HCN1, HCN4, Ca v 1.3, and Ca v 3.1 are pacemaking ion channels (see text for details); LZTS1, LBH, TBX3, and TBX18 are transcription factors; COL1A1 is a collagen marker; TPSAB1 is a mast cell marker; HLA-DRA is histocompatibility complex marker; and CD209 is a macrophage cell marker. N = no; Y = yes; Hsa = homosapien.

Luciferase Reporter Gene Assay
Rat cardiac H9C2 cells (cell line purchased from ATCC, LGC Limited, Middlesex, UK) were maintained in DMEM (Life Technologies, Inc., Gaithersburg, MD, USA) containing 10% feline bovine serum (Life Technologies, Inc., Gaithersburg, MD), 1% penicillinstreptomycin (Life Technologies, Inc., Gaithersburg, MD), and 1% nonessential amino acids (Thermo Fisher Scientific, Altrincham, UK). At day 1, cells were plated in 48-well plates at a density of 50,000 cells in a volume of 250 µL of media for 24 h. Following this, the medium was removed and replaced with 225 µL of fresh medium. Cells were transfected with 0.25 µg plasmid (from a 1 µg/µL stock solution) with 1.25 µL miRNA (from a 5 µM stock) in 11 µL of optiMEM (Life Technologies, Inc., Gaithersburg, MD). Lipofectamine 2000 (ThermoScientific, Altrincham, UK) was diluted in optiMEM (0.75 µL lipofectamine 2000 with 11.75 µL optiMEM). Then, the diluted lipofectamine 2000, diluted miRNA, and diluted plasmid were mixed (see Figure 1 for details) and allowed to incubate for 5 min at room temperature before adding to the cells in each well. The cells were then incubated at 37 • C and 5% CO 2 for 24 h before luciferase activity was measured.
Following this, renilla and luciferase substrate and buffers (Promega) were thawed at room temperature. Here, 1X cell culture lysis was prepared from 5X cell culture lysis reagent (Promega), using milliQ water to dilute. Medium was aspirated from the cells before washing with PBS, and then 50 µL lysis buffer was added and placed on a rocker for 20-25 min. In addition, 10 µL of each lysate was added into a 96-well plate in duplicate for both renilla and luciferase assay readings. A GloMax Luminometer (Promega, Southampton, UK) was set up to inject 50 µL of either luciferase (from injector pump 1) or renilla (from injector pump 2) per well. Luciferase assay activity was normalized to renilla assay activity to obtain a ratio than was then inputted in GraphPad Prism 8. Batches of experiments ranged between 3 and 6.

Statistical Analysis
In Tables 2 and 3 and Figures 3-5 mean ± SEM values are shown. All statistical analysis was carried out using GraphPad Prism 8. Significant differences were identified using unpaired t test. Significant difference was assumed at p ≤ 0.05.

Binding Sites Prediction
Genome (https://genome.ucsc.edu accessed on 25 November 2021) was used to obtain the 3′UTR sequence for the target mRNAs. MirBase (https://www.mirbase.org accessed on 25 November 2021) was used to obtain the sequence for the microRNAs. The microRNA sequence and the 3′UTR sequence for their target mRNA were uploaded into RNA22 v2 (https://cm.jefferson.edu/rna22/interactive/ accessed on 25 November 2021) or TargetScan Human (http://www.targetscan.org/vert_72/ accessed on 25 November 2021) in order to predict the number of binding sites on the mRNAs for their corresponding microRNAs and identify predicted binding sites (Table 1).

Results
Out of the 12 hsa-mRNAs that had been predicted to be inhibited by their respective has-miRNAs, we did not observe a significant decrease in luciferase activity with eight hsa-mRNAs and their respective hsa-miRNAs (Table 2). However, we observed a significant decrease in luciferase activity with four hsa-mRNAs and their respective hsa-miR-NAs (Table 3, Figures 3-5), suggesting the presence of interaction between the hsa-mRNAs and hsa-miRNAs. Therefore, we can assume a reduction in the expression of the genes and proteins encoded by these genes, based on these observed interactions.

Binding Sites Prediction
Genome (https://genome.ucsc.edu accessed on 25 November 2021) was used to obtain the 3 UTR sequence for the target mRNAs. MirBase (https://www.mirbase.org accessed on 25 November 2021) was used to obtain the sequence for the microRNAs. The microRNA sequence and the 3 UTR sequence for their target mRNA were uploaded into RNA22 v2 (https://cm.jefferson.edu/rna22/interactive/ accessed on 25 November 2021) or TargetScan Human (http://www.targetscan.org/vert_72/ accessed on 25 November 2021) in order to predict the number of binding sites on the mRNAs for their corresponding microRNAs and identify predicted binding sites (Table 1).

Results
Out of the 12 hsa-mRNAs that had been predicted to be inhibited by their respective has-miRNAs, we did not observe a significant decrease in luciferase activity with eight hsa-mRNAs and their respective hsa-miRNAs (Table 2). However, we observed a significant decrease in luciferase activity with four hsa-mRNAs and their respective hsa-miRNAs (Table 3, Figures 3-5), suggesting the presence of interaction between the hsa-mRNAs and hsa-miRNAs. Therefore, we can assume a reduction in the expression of the genes and proteins encoded by these genes, based on these observed interactions.

Discussion
Following our previously predicted interactions between key hsa-miRNAs and key hsa-mRNAs [4,10], it was important for us to experimentally verify these predictions. It should be noted that the experimental validation of miRNA-mRNA interaction is still a challenging feat. The reason why some predicted microRNA-mRNA interactions have been confirmed (through observations of significantly reduced bioluminescence/luciferase activity) and some interactions could not be confirmed could be due to several factors such as: thermodynamic stability and binding site accessibility-as recently reviewed by Riolo et. al., 2021 [11].
The current study focuses on two microRNAs (hsa-miR-486-3p and hsa-miR-938) that significantly reduces the luciferase of the mRNAs encoding for Ca 2+ ion channels (Ca v 1.3 and Ca v 3.1) and a mast cell marker (TPSAB1) in H9C2 cells (Table 3). This suggests that an interaction between these mRNAs and their respective microRNAs is taking place; therefore, this could result in a reduction in expression of Ca v 1.3, Ca v 3.1, and TPSAB1.
In our previous study, we have shown that the direct binding of hsa-miR-486-3p to hsa-HCN4 significantly reduces HCN4-mediated luciferase activity and reduces the beating rate in the rat SN preparations [4]. We now provide evidence that hsa-miR-486-3p also significantly downregulates the luciferase activity of mRNAs encoding hsa-Ca v 1.3, hsa-Ca v 3.1, and hsa-TPSAB1 (Figures 3-5).
Based on our studies and many other studies described above, hsa-miR-486-3p can be a principal therapeutic target in the treatment of cardiac dysfunction.

Hsa-Ca v 1.3 and Hsa-miR-486-3p
Ca v 1.3 is a voltage-gated L-type Ca 2+ channel that is important in the regulation of the SN's pacemaking function. When this channel is deactivated in mice, bradycardia and sinoatrial arrhythmia are observed [21,22]. We have recently shown that hsa-miR-486-3p is predicted to indirectly inhibit hsa-Ca v 1.3-with hsa-Ca v 1.3 being predicted to have six binding sites for hsa-miR-486-3p [4]. Our luciferase assay experiment confirms this prediction because the binding of hsa-miR-486-3p significantly reduces hsa-Ca v 1.3mediated bioluminescence (Figure 3).

Hsa-Ca v 3.1 and Hsa-miR-486-3p
Ca v 3.1 is a voltage-gated T-type Ca 2+ channel that also plays a key role in regulating the pacemaking function of the SN. Mangoni et al. (2006) demonstrated that the knockout of Ca v 3.1 in mice led to the lack of transitory I Ca,T current and slowed down the SN pacemaker activity and AV conduction and reduced their heart rate [23]. We have recently shown that hsa-miR-486-3p is predicted to indirectly inhibit hsa-Ca v 3.1-with Ca v 3.1 being predicted to have one binding site for hsa-miR-486-3p [4]. Our luciferase assay experiment confirms this prediction because the binding of hsa-miR-486-3p significantly reduces hsa-Ca v 3.1-mediated bioluminescence ( Figure 3).

Hsa-TPSAB1 and Hsa-miR-486-3p
In the early phase of myocardial infarction, inflammatory cells migrate to the infarction in order to aid cardiac repair [24]. It has been shown that a dysfunction of the immune and inflammatory pathways may contribute to the development of various cardiovascular diseases, including heart failure and arrhythmia [25,26]. Mast cells are involved in the immune response pathways, and TPSAB1-the main tryptase isoenzyme that is expressed in mast cells-plays a role in immunity [27].
Tryptase is the enzymatic activator of membrane protease-activated receptor 2 (PAR-2)-a receptor that is expressed by cardiomyocytes [28]. These PAR-2 receptors are connected to ERF kinases that mediate survival and hypertrophic signals. It could be assumed that hsa-miR-486-3p suppresses the tryptase/PAR-2/ERK hypertrophy pathway in the SN, preventing myocyte hypertrophy and therefore supporting the SN's spindle-like neonatallike morphology. In addition, tryptase belong to the serine proteases family [29], and they all have very conservative 3-UTRs with a unique secondary structure that is crucial to mRNA stability.
We recently reported that hsa-TPSAB1 is significantly expressed in the adult human SN and is predicted to be directly inhibited by hsa-miR-486-3p [10]. The function of TPSAB1 in the SN is unknown, but it is interesting to see its high expression in the SN. The binding of hsa-miR-486-3p significantly reduces hsa-TPSAB1-mediated bioluminescence ( Figure 5). It is possible that the immune system is important for maintaining the pacemaking function of the human SN. A high expression of hsa-miR-486-3p in the blood plasma of COVID-19 patients [20] could result in the reduced expression of hsa-TPSAB1 and abnormal functioning of immune cells in the SN. Based on this information, hsa-miR-486-3p could be a promising target in the treatment of bradycardia that is experienced by COVID-19 patients [18,19].
The high expression of immune cells in the SN could slow down the development of effective biological pacemakers. It has been supposed that immune cells contribute to SN functioning. For example, Hu et al. (2014) reported that when an adenoviral vector of TBX18 was delivered to a swine model of total heart block, it affected not only the heart rhythm but also caused an immune response [30].

Hsa-miR-938
We have previously reported that hsa-miR-938 is one of the novel microRNAs that is uniquely expressed in the human right atrium and is predicted to inhibit Ca v 1.3 [4]. Now, following our luciferase assay experiments, we observed a significant decrease in the luciferase activity of the mRNA encoding for Ca v 1.3 when hsa-miR-938 is added (Figure 4). This observation suggests there is interaction between hsa-Ca v 1.3 and hsa-miR-938. The function of this microRNA in the cardiovascular system is unknown, but this microRNA is known to promote cell proliferation [31] and is linked to the development of gastric [32] and pancreatic cancer [33].

Conclusions
With the growing occurrence of cardiovascular disease, particularly sinus node dysfunction (SND), as the global ageing population increases, it is important for us to identify new therapeutic targets and to further understand the complex mechanisms that regulate the SN's pacemaking function. We hereby provide novel insights into the interacting between key microRNAs and ion channels and immune cell markers in H9C2 cells by further exploring our previous predicted interactions (as described in Petkova et al. (2020) [4] and Aminu et al. (2021) [10]). Using luciferase assay experiments, our observation of significant reduction of luciferase activity suggests there is interaction between miR-486-3p and three mRNAs (Ca v 1.3, Ca v 3.1, and TPSAB1) and between miR-938 and Ca v 1.3 mRNA. These observations allow room for further explorations of how these microRNAs can be used for ex vivo and in vivo experiments to study their effect on cardiac function/dysfunction, as previously shown by Yanni et al., who injected an antimiR to miR-370 into heart failure mice and restored HCN4 mRNA and protein expression in the SN, thus increasing the beating rate [34], and recently by Petkova et al., who injected miR-486-3p into the rat SN tissue and observed a reduced heart rate [4].