Antisense-mediated regulation of exon usage in the elastic spring region of Titin modulates sarcomere function

Selvi Celik; Ludvig Hyrefelt; Tomasz Czuba; Yuan Li; Juliana Assis; Julia Martinez; Markus Johansson; Oscar André; Jane Synnergren; Joakim Sandstedt; Pontus Nordenfelt; Kristina Vukusic; J Gustav Smith; Olof Gidlöf

doi:10.1093/cvr/cvaf037

. 2025 Mar 5;121(4):629–642. doi: 10.1093/cvr/cvaf037

Antisense-mediated regulation of exon usage in the elastic spring region of Titin modulates sarcomere function

Selvi Celik ^1,², Ludvig Hyrefelt ³, Tomasz Czuba ^4,^5,^6,⁷, Yuan Li ^8,⁹, Juliana Assis ¹⁰, Julia Martinez ^11,¹², Markus Johansson ¹³, Oscar André ¹⁴, Jane Synnergren ^15,¹⁶, Joakim Sandstedt ¹⁷, Pontus Nordenfelt ¹⁸, Kristina Vukusic ¹⁹, J Gustav Smith ^20,^21,^22,²³, Olof Gidlöf ^24,^25,^✉,^b

¹ Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden

² Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden

³ Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden

⁴ Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden

⁵ Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden

⁶ The Wallenberg Laboratory/Department of Molecular and Clinical Medicine, Institute of Medicine, Gothenburg University and The Department of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden

⁷ Science for Life Laboratory, Gothenburg University, Gothenburg, Sweden

⁸ National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Lund University, Lund, Sweden

⁹ Department of Immunotechnology, Lund University, Lund, Sweden

¹⁰ National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Lund University, Lund, Sweden

¹¹ Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden

¹² Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden

¹³ Systems Biology Research Center, School of Bioscience, University of Skövde, Skövde, Sweden

¹⁴ Division of Infection Medicine, Department of Clinical Sciences, Lund University, Lund, Sweden

¹⁵ Systems Biology Research Center, School of Bioscience, University of Skövde, Skövde, Sweden

¹⁶ Department of Molecular and Clinical Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

¹⁷ Department of Laboratory Medicine, Institute of Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

¹⁸ Division of Infection Medicine, Department of Clinical Sciences, Lund University, Lund, Sweden

¹⁹ Department of Laboratory Medicine, Institute of Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

²⁰ Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden

²¹ Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden

²² The Wallenberg Laboratory/Department of Molecular and Clinical Medicine, Institute of Medicine, Gothenburg University and The Department of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden

²³ Science for Life Laboratory, Gothenburg University, Gothenburg, Sweden

²⁴ Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden

²⁵ Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden

^✉

Corresponding author. Tel: +46 (0)462224707, E-mail: olof.gidlof@med.lu.se

Conflict of interest: none declared.

PMCID: PMC12054628 PMID: 40042822

Abstract

Aims

Alternative splicing of Titin (TTN) I-band exons produce protein isoforms with variable size and elasticity, but the mechanisms whereby TTN splice factors regulate exon usage and thereby determining cardiomyocyte passive stiffness and diastolic function, is not well understood. Non-coding RNA transcripts from the antisense strand of protein-coding genes have been shown to regulate alternative splicing of the sense gene. The TTN gene locus harbours >80 natural antisense transcripts (NATs) with unknown function in the human heart. The aim of this study was to determine if TTN antisense transcripts play a role in alternative splicing of TTN.

Methods and results

RNA-sequencing and RNA in situ hybridization (ISH) of cardiac tissue from heart failure (HF) patients, unused donor hearts, and human iPS-derived cardiomyocytes (iPS-CMs) were used to determine the expression and localization of TTN NATs. Live cell imaging was used to analyse the effect of NATs on sarcomere properties. RNA ISH and immunofluorescence was performed in iPS-CMs to study the interaction between NATs, TTN mRNA, and splice factor protein RBM20. We found that TTN-AS1-276 was the predominant TTN NAT in the human heart and that it was up-regulated in HF. Knockdown of TTN-AS1-276 in human iPS-CMs resulted in decreased interaction between RBM20 and TTN pre-mRNA, decreased TTN I-band exon skipping, and markedly lower expression of the less compliant TTN isoform N2B. The effect on TTN exon usage was independent of sense–antisense exon overlap and polymerase II elongation rate. Furthermore, knockdown resulted in longer sarcomeres with preserved alignment, improved fractional shortening, and relaxation times.

Conclusions

We demonstrate a role for TTN-AS1-276 in facilitating alternative splicing of TTN and regulating sarcomere properties. This transcript could constitute a target for improving cardiac passive stiffness and diastolic function in conditions such as heart failure with preserved ejection fraction.

Keywords: Sarcomere function, Splicing, Non-coding RNA, Titin

Graphical Abstract

Time of primary review: 39 days

1. Introduction

Titin (TTN) is the largest protein in the human body (∼3 MDa) and constitutes a central component of the sarcomere.¹ It is anchored in the Z-disc and extends to the M-band, aligned along the length of an entire half-sarcomere (physiological range in adult human cardiomyocytes: 1.8–2 µm).² The presence of extensible PEVK-repeat and immunoglobulin-domains in the I-band region gives TTN elastic properties, acting as a biomechanical spring that contributes to sarcomere passive tension and provides elastic force in response to stretch.^3,4 In the heart, TTN is a major determinant of myocardial passive stiffness^1,5 and plays important roles in regulating diastolic function.^6–8 Moreover, truncating genetic variants in TTN are the most common genetic lesion in patients with dilated cardiomyopathy (DCM),⁹ underscoring the importance of TTN in regulating contractile function. Elucidation of molecular mechanisms regulating TTN compliance could thus be of therapeutic importance for heart diseases characterized either by excessive myocardial stiffness and diastolic dysfunction, such as heart failure with preserved ejection fraction (HFpEF), or by systolic dysfunction, such as DCM.

The mechanical properties of TTN are primarily regulated at the post-transcriptional level, where a series of exon-skipping events among I-band exons produces an array of transcript isoforms differing in the number of extensible domains.¹⁰ Alternative splicing of the enormous 363 exon TTN gene is extraordinarily complex and presumably orchestrated through a highly regulated process involving a number of splicing factors. However, RNA binding motif protein 20 (RBM20) is currently the only protein known to promote alternative splicing of TTN.¹¹ Consequently, pathogenic variants in RBM20 have been implicated in 2–6% of DCM cases.¹² RBM20 interacts with intronic motifs in pre-mRNA¹³ and prevents inclusion of up- and downstream exons through interference with U1 snRNP splice site recognition.¹⁴ However, mechanisms regulating recruitment of RBM20 to TTN pre-mRNA or additional factors affecting alternative splicing of TTN have not been elucidated.

Antisense transcription, i.e. transcription of non-coding RNA from the opposite strand of a coding gene, is pervasive throughout the human genome¹⁵ and can influence expression and splicing of the sense transcript. Morrissy et al.¹⁶ found a striking association between the presence of antisense transcription and alternative splicing of the sense gene across the human genome and postulated that slowing of the RNA polymerase II elongation rate at overlapping sense–antisense exons promotes alternative splicing of the sense transcript. Others have reported that antisense transcripts can mask splice sites in the sense transcript by forming an RNA-duplex with complementary sequences in the coding transcript¹⁷ or recruit specific components of the spliceosome to the sense gene pre-mRNA.^18,19 The TTN locus spans >280 kb on chromosome 2q31 and harbours >80 annotated antisense transcripts, the function of which have not been studied in the heart previously.

The aims of this study were to map TTN antisense transcription in the human heart, to elucidate its possible role in regulation of TTN splicing, and to study the potential downstream implications of targeting specific TTN antisense transcripts on sarcomere organization and function.

2. Methods

A full description of methods can be found in Supplementary material online, Methods.

2.1. Human samples

Left ventricular biopsies from unused organ donor hearts (n = 7) and explanted cardiac tissue from heart failure patients (n = 100) were collected at transplantation centres at Lund University Hospital, Lund, Sweden and Sahlgrenska University Hospital, Gothenburg, Sweden, and stored at −80°C (patient characteristics are provided in Table 1). Informed consent was provided by patients or in the case of unused organ donor hearts, a close relative. The study was approved by the Ethics Board at Lund and Gothenburg University, respectively. The study was conducted in concordance with the Declaration of Helsinki. The methodology, conduct, and reporting of this study were in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for observational studies. STROBE recommendations for reporting observational studies are available as Supplementary material online, Table S1.

Table 1.

Patient characteristics

	Non-failing controls	Heart failure patients
Sample source	Unused donor hearts	Explanted hearts
Number of individuals	7	100
Age, mean (SD), y	55.7 (12.3)	50.3 (14.1)
Sex
Female (n)	3	25
Male (n)	4	75
Aetiology
DCM, n		47
ICM, n		19
HCM, n		10
Other, n		24
Comorbidities
Hypertension, n		17
Diabetes, n		9
Coronary artery disease, n		26
Stroke, n		14
Smoking
Yes		0
Ex		38
No		62

Open in a new tab

2.2. Human heart muscle cells

Human iCell iPS-derived cardiomyocytes (iPS-CMs) were sourced from FujiFilm Cellular Dynamics Inc. (Madison, WI, USA) and Takara Bio (Takara Bio Europe, Saint-Germain-en-Laye, France). Cells were cultured as recommended by the manufacturers.

2.3. RNA isolation

Frozen tissue was cut into small pieces and homogenized in 700 μL of QIAzol (Qiagen, Hilden, Germany) using an Omni TH rotor-stator homogenizer. For isolation of RNA from cells, 700 μL of QIAzol was added directly to cell culture plates. For isolation of chromatin-enriched and soluble nuclear RNA fractions from cells, the protocol described by Werner et al.²⁰ was used. Total RNA was isolated using the miRNeasy mini kit (Qiagen) according to the manufacturer’s instructions. The quantity and quality of isolated RNA was assessed with Qubit Flex (ThermoFisher) using the QuantIT RNA HS Assay Kit (ThermoFisher) and Agilent 4200 TapeStation (Agilent Technologies, Santa Clara, CA, USA) using the RNA ScreenTape Analysis Kit (Agilent Technologies).

2.4. siRNA and plasmid DNA transfection

For knockdown experiments, cells were transfected with Silencer Select siRNA (ThermoFisher) directed towards exon 1 (si276-Ex1, #n294437) or exon 12 (si276-Ex12, custom design ID #ABRSBMG) of TTN-AS1-276, towards RBM20 (ENST00000369519.4, #s49081) or with a scrambled negative control siRNA sequence (#4390843). For visualization and tracking of sarcomeres, cells were transfected with a plasmid expressing ACTN2 (NM_001103.4) with a eGFP tag.²¹ For immunoprecipitation of RBM20, cells were transfected with a pEZ-M03 vector (GeneCopoeia, Rockville, MD, USA) expressing RBM20 (ENST00000369519.4) with an eGFP tag. An in-frame deletion of a region spanning from exon 5 to exon 9 of the RBM20 open reading frame was made using site-directed mutagenesis (SDM) in order to produce a fusion protein lacking the RBM20 RNA recognition motif (RRM). SDM was carried out using the Phusion Site-Directed Mutagenesis Kit (ThermoFisher) according to the manufacturer’s instructions with forward primer 5′-GAGCCCAAAGCCAAGTCGGACAAGTAT-3′ and reverse primer 5′-CCTTGCTGGAATGGGCACGTATGATGTT-3′. Confirmation of the deletion was performed with PCR using the forward primer 5′-ATAACCCTGCTGGGAATGAAG-3′ and reverse primer 5′-CCACTGATTGAGGGCTTTCT-3′. Transfections were performed using Lipofectamine 3000 (ThermoFisher) according to the manufacturer’s instructions.

2.5. RNA in situ hybridization of human cardiac tissue

Human cardiac cryosections were fixed, dehydrated and pre-treated with Protease IV for the RNAScope Fluorescent Multiplex Assay (Advanced Cell Diagnostics, Hayward, CA, USA) according to the manufacturer’s recommendations. The RNAScope Multiplex Fluorescent Assay was performed using probes Hs-TTN-C1 (#550361, Advanced Cellular Diagnostics) and Hs-TTN-AS1-C2 (#1115141-C2, Advanced Cellular Diagnostics), according to the manufacturer’s recommendations. Before mounting, Wheat Germ Agglutinin-AlexaFluor488 (ThermoFisher) was added. Sections were then washed, counterstained with DAPI and mounted. Sections were imaged using a Nikon TiE TIRF microscope (Nikon Corporation, Tokyo, Japan) equipped with a Photometrics Prime95B sCMOS camera (Teledyne Photometrics, Tucson, AZ, USA).

2.6. Combined RNA in situ hybridization and immunofluorescence

iPS-CM was seeded in Lab-Tek 4-well chamber slides (Sigma-Aldrich) and fixed, de- and rehydrated and treated with Protease III in preparation for RNAScope Multiplex Fluorescent Assay (Advanced Cellular Diagnostics) according to the manufacturer’s instructions. The RNAScope Multiplex Fluorescent Assay was performed according to the manufacturer’s recommendations. Before mounting, immunofluorescent staining was performed with a rabbit anti-RBM20 antibody (Abcam, #ab233147) at 10 μg/mL for 1 h. An AlexaFluor488 anti-rabbit IgG secondary antibody (#4412 Cell Signaling, Danvers, MA, USA) at 1:1000 dilution was then added and incubated for 30 min. Slides were counterstained with DAPI and mounted. Slides were imaged using Operetta CLS high content screening instrument (PerkinElmer, Waltham, MA, USA), and the number and localization of TTN (Atto 550), TTN-AS1 (Atto 647), and RBM20 (AlexaFluor488) fluorescent foci were analysed using Harmony 5.2 software.

2.7. Protein gel electrophoresis

Analysis of TTN protein isoforms was performed with agarose gel electrophoresis according to a previously established protocol.²² Protein bands representing TTN isoforms were visualized on ChemiDoc MP imaging system (Bio-Rad) and quantified using densiometric measurements in Image Lab 6.1 (Bio-Rad) and normalized to the myosin heavy chain 7 (MYH7) band.

2.8. RNA immunoprecipitation

Human iPS-CM was transfected with pCMV-RBM20-GFP plasmid DNA and siRNA as described elsewhere. RNA immunoprecipitation (RIP) was performed on 100 μL cell lysate per sample using the Magna RIP Kit (Merck) according to the manufacturer’s instructions with rabbit polyclonal anti-GFP antibody (#ab290, Abcam) or rabbit IgG antibody.

2.9. Chromatin immunoprecipitation

Preparation of cross-linked chromatin and chromatin immunoprecipitation was performed using the SimpleChIP Kit (Cell Signaling) according to the manufacturer’s instructions with ChIPAb+ anti-RNA Pol II mouse monoclonal antibody or mouse IgG antibody.

2.10. Sarcomere tracking

iPS-CM was transfected with pACTN2-GFP to visualize the sarcomere Z-discs. Live cell imaging of contracting iPS-CM was performed with wide-field epifluorescence microscopy using an ECLIPSE Ti2-E microscope (Nikon). Videos of contracting cells, capturing at least two contractions, were recorded at 30 frames per second using a Nikon DS-Qi2 CMOS camera. Segmentation of Z-discs and sarcomere tracking was then performed on a total of 32 video files in the SarcGraph software.²³

2.11. qPCR and RT–PCR

cDNA was synthesized using the RevertAid First Strand cDNA Synthesis Kit (ThermoFisher) according to the manufacturer’s instructions and used in qPCR reactions with 2 × Universal TaqMan Master Mix (ThermoFisher) or in RT–PCR reactions with 2 × PCR Master Mix (ThermoFisher). The expression of TTN, RBM20, GAPDH, CAMK2D, CACNA1C, LMO7, PRKRA, CCDC141, PLEKHA3, FKBP7, and DFNB59 was assessed with TaqMan Gene Expression Assays (ThermoFisher). For the quantification of specific splice products or exons from TTN, TTN-AS1, CAMK2D, CACNA1C, and LMO7, custom PrimeTime qPCR Probe Assays spanning specific exon–exon junctions or within exons were designed using the PrimerQuest Tool (Integrated DNA Technologies, Coralville, IA, USA). See Supplementary material online, Table S2 for primer and probe sequences. All qPCR reactions were run on a Bio-Rad CFX 96 instrument (Bio-Rad, Hercules, CA, USA). For gene expression analysis, Ct-values were normalized to the reference gene GAPDH and expressed relative to the mean of the control group. For quantification of splicing products/isoforms, Ct-values were normalized to a qPCR assay designed to measure all transcripts from the corresponding gene and expressed relative to the mean of the control group. For quantification of immunoprecipitated RNA, Ct-values were transformed to ‘% of input’ using the Ct-value of the 2% Input sample with the formula 2^{−(CtRIP − (CtInput − 3.32))}. RT–PCR reactions were run on an agarose gel with GelRed (Biotium, Fremont, CA, USA), and amplicons were visualized using a ChemiDoc MP imaging system with the ‘UV Trans’ application. Bands corresponding to the PCR amplicons were quantified using densiometric measurements in Image Lab 6.1 and normalized to the input samples.

2.12. RNA-sequencing

For human cardiac tissue samples, 800 ng of RNA was used as input for library preparation using the TruSeq Stranded Total RNA Library Prep Gold kit (Illumina, San Diego, CA, USA) with rRNA Removal Mix. Sequencing was performed on a NovaSeq 6000 system using the NovaSeq 6000 S4 reagent kit (Illumina) with 101 bp paired-end reads (see Supplementary material online, Figure S1).

For human iPS-CMs, 10 ng of RNA was used as input for cDNA synthesis using the SMART-Seq v4 Ultra Low Input RNA Kit for Sequencing (Takara Bio, San Jose, CA, USA). Library preparation was performed using the Nextera XT DNA Library Preparation Kit (Illumina). Sequencing was performed using the NovaSeq 6000 SP Reagent Kit v 1.0 (Illumina) with 101 bp paired-end reads. All mapped reads with mapping quality > 10 were counted analysed with DEXSeq 1.44.0.^24,25 Per cent spliced in (PSI) was estimated using the Calculate-PSI package (https://github.com/jalwillcox/Calculate-PSI).

2.13. Statistical analysis

All data are shown with mean and standard deviation. Differences between experimental groups were assessed with Student’s t-test or ANOVA with Dunnett’s multiple comparisons test as specified in the figure legends. Correlation between continuous variables was assessed with linear regression. Associations between binary outcomes and continuous variables were assessed with binary logistic regression. All statistical analyses were performed in GraphPad Prism v. 10.

3. Results

3.1. Cardiac antisense transcription in the TTN locus

Antisense transcription in the TTN gene locus is extensive, with 81 TTN-AS1 transcripts annotated in GENCODE v. 44 (see Supplementary material online, Figure S2A). To obtain an overview of TTN-AS1 transcript structure and expression in the human heart, we determined exon usage across all TTN-AS1 isoforms by applying the DEXSeq analysis pipeline on RNA-sequencing data from left ventricular tissue without heart disease (n = 7, Figure 1A). Exon usage varied substantially within and between transcripts, but the high-expression (75th percentile) exons all belonged to a group of core TTN-AS1 transcripts (Figure 1B). We quantified the four TTN-AS1 transcripts with the highest expression exons (TTN-AS1-276, -209, -203, and -223) using the same human cardiac tissue without heart disease (n = 7) and custom qPCR assays targeting transcript-specific exons or exon–exon junctions (Figure 1C). The highest expression in cardiac tissue was observed for TTN-AS1-276 (ENST00000659121), which is also the ENSEMBL canonical transcript, defined as the single most representative and best-supported transcript from any given gene. TTN-AS1-276 is a 6297 bp transcript with 13 exons spanning >250 kb of the TTN gene (see Supplementary material online, Figure S2A). The transcriptional start site of TTN-AS1-276 lies ∼2500 bp downstream of TTN and a high-quality map of functional genomic regions based on Roadmap Epigenomics data²⁶ showed that this genomic region is an active promoter in human atrial and ventricular tissue (see Supplementary material online, Figure S2B). RNA in situ hybridization (ISH) revealed widespread expression of TTN-AS1-276 in human left ventricular tissue (Figure 1D and Supplementary material online, Figure S3), localized predominantly in cardiomyocytes (as defined by large cell/nuclear size and presence of TTN ISH foci). Cardiomyocyte-specific expression of TTN-AS1 was corroborated by GTEx cardiac single nucleus RNA-sequencing data,²⁷ where median log-normalized and scaled expression counts for TTN-AS1 were >10 000-fold higher in cardiomyocytes than in any other cardiac cell type.

Expression and localization of *TTN* antisense transcripts in the human heart. (A) *TTN-AS1* exon usage was calculated using cardiac RNA-sequencing data from seven organ donor hearts without heart disease. The dashed line indicates the 75th percentile. (B) The *TTN-AS1* transcripts with the highest exon usage. (C) Relative expression of *TTN-AS1* transcripts and *TTN* with qRT–PCR in the seven organ donor hearts without heart disease. (D) RNA *in situ* hybridization (ISH) for *TTN-AS1-276* and *TTN* in a tissue section from one representative human cardiac biopsy. Cell membranes were stained with wheat germ agglutinin (WGA). (E) RNA ISH for *TTN-AS1-276* in human iPS-derived cardiomyocytes (iPS-CM). Nuclei were stained with DAPI. Below, quantification of the proportion of nuclear and cytoplasmic *TTN-AS1-276* ISH foci. (F) Relative expression of *TTN-AS1-276* in iPS-CM nuclear compartments with qRT–PCR. Data are derived from two different RNA preparations per group. SOL, soluble fraction; CHE, chromatin-enriched fraction.

Clues regarding the functional role of a natural antisense transcript can be drawn from its intracellular localization.²⁸ We therefore performed RNA ISH on human iPS-CMs and observed that ∼85% of TTN-AS1-276 foci were localized in the nucleus (Figure 1E). Moreover, fractionation of iPS-CM nuclear RNA revealed ∼300-fold higher levels of TTN-AS1-276 in the chromatin compartment over the soluble compartment (Figure 1F). We conclude that TTN-AS1-276 is the predominant TTN antisense transcript in the human heart and that it is localized to cardiomyocyte nuclear chromatin.

3.2. Design and validation of siRNAs targeting TTN-AS1-276

In order to study the functional role of TTN-AS1-276, we designed two independent siRNAs directed towards exons 1 and 12 of TTN-AS1-276 (si276-Ex1 and si276-Ex12). Both si276-Ex1 and si276-Ex12 caused a statistically significant ∼40% decrease in TTN-AS1-276 expression (see Supplementary material online, Figure S4A) but the expression of the other main TTN-AS1 isoforms (-203, -209, and -223) was unaffected (see Supplementary material online, Figure S4A), confirming the efficacy and specificity of these siRNAs.

3.3. TTN-AS1-276 does not regulate TTN expression

Natural antisense transcripts often play a role as cis-acting transcriptional regulators of the sense protein-coding gene.²⁹ To assess whether TTN-AS1-276 affects the expression of TTN, we transfected human iPS-CM with si276-Ex1 and si276-Ex12 but detected no effect on expression of TTN or any of the other protein-coding genes immediately within 100 kb up- and downstream of TTN (see Supplementary material online, Figure S4B). We conclude that TTN-AS1 does not play a role as a cis-acting transcriptional regulator.

3.4. TTN-AS1-276 regulates splicing of TTN I-band exons

We next hypothesized that TTN-AS1 could influence alternative splicing of TTN, as the presence of antisense transcription across the human genome has been shown to associate strongly with alternative splicing of the corresponding sense gene.¹⁶ We performed RNA-seq on human iPS-CM transfected with si276-Ex1 or, as a positive control, siRNA towards RBM20 (siRBM20), a splicing factor that mediates exon skipping in TTN,¹³ and calculated proportion spliced-in (PSI) for each TTN exon. Successful knockdown of TTN-AS1-276 and RBM20 was confirmed with qPCR (see Supplementary material online, Figure S5A). TTN PSI in control iPS-CM correlated well (Pearson r = 0.837, P < 0.0001) with previously reported data from human left ventricular tissue,³⁰ with extensive splicing out of exons across the I-band region throughout exons 50–242 (numbered according to the complete inferred TTN meta-transcript, Supplementary material online, Figure S5B, Figure 2A). Knockdown of TTN-AS1-276 caused significantly increased PSI for 26 exons (indicated by red bars in Figure 2B). Affected exons were all situated in the I-band and had baseline PSI values in the range of 70–85%. The most pronounced effect of TTN-AS1-276 knockdown was observed in exons immediately downstream of exon 49 (exons 50–89, indicated by a dashed square in Figure 2B). These exons are normally spliced out of TTN through RBM20-mediated exon skipping³¹ and alternative splicing/exon skipping from exon 49 produces a range of TTN splice products,³² notable examples of which are shown in Figure 2C. Knockdown of RBM20 resulted in a similar increase in PSI in exons downstream of exon 49 (see Supplementary material online, Figure S5C), confirming its role in mediating I-band exon skipping. To study whether the increased PSI in exons downstream of 49 following TTN-AS1 knockdown was caused by decreased exon skipping, we quantified the primary splice products across exon 49 splice junctions in iPS-CM transfected with si276-Ex1, si276-Ex12, or siRBM20 using qPCR. We observed a significant decrease in expression of TTN splice products dependent on exon skipping and a reciprocal increase in the isoform where exon 49 is spliced directly with exon 50 (Figure 2D). Considering the heterogeneity of iPS-CM, we validated the results using another iPS-CM cell line (see Supplementary material online, Figure S6A). We conclude that TTN-AS1-276 regulates exon usage in I-band TTN through facilitating exon skipping downstream of exon 49.

*TTN-AS1-276* regulates inclusion of I-band exons in *TTN*. (A) Per cent spliced in (PSI) for all exons across the *TTN* gene calculated based on RNA-sequencing reads from human iPS-derived cardiomyocytes (IPS-CM). The line represents the mean of three replicates (individual RNA preparations) per experimental group. (B) The difference in PSI (ΔPSI) comparing cells transfected with siRNA to *TTN-AS1-276* (si276-Ex1) with cells transfected with scrambled negative control siRNA (siScr). Exons with statistically significant ΔPSI are marked with red bars (adjusted P < 0.05). (C) Depiction of exon skipping from *TTN* exon 49. (D) Quantification of *TTN* splice products in iPS-CM transfected with si276-Ex1, si276-Ex12, siRBM20 or siScr, using custom qRT–PCR assays spanning the indicated exon–exon junctions. Expression data are normalized to that of total *TTN* and expressed relative to the mean of the negative control cells (siScr). Data are derived from two separate experiments with 3–6 replicates (individual RNA preparations) per experimental group. Differences between each individual experimental group and the control group were assessed with Student’s t-tests, *P < 0.05, **P < 0.01, ***P < 0.001.

3.5. TTN-AS1-276 regulates TTN isoform composition

Cardiac TTN is composed of two primary isoforms, N2BA and N2B, differing in the extent to which I-band domains are included. N2B is produced by splicing of exon 49 with exon 219, excluding many extensible I-band domains and resulting in a short and less elastic TTN protein isoform (Figure 3A). As our data showed that TTN-AS1-276 promotes exon skipping downstream of exon 49, we hypothesized that knockdown of TTN-AS1 would result in decreased expression of the N2B isoform. To address this, we quantified expression of N2B and N2BA in iPS-CM transfected with si276-Ex1, si276-Ex12, or siRBM20 using custom qPCR assays spanning the exon 49–219 junction (N2B) and the exon 108–109 junction (constitutively included in N2BA). We observed significantly lower expression of the N2B isoform in iPS-CM where TTN-AS1-276 had been knocked down, whereas expression of N2BA was unaffected (Figure 3B). This resulted in a two-fold increase in the N2BA:N2B ratio (Figure 3C). We confirmed these results in a separate iPS-CM line (see Supplementary material online, Figure S6B). Knockdown of RBM20 caused a dramatic decrease in N2B expression and a pronounced increase in the N2BA:N2B ratio, in line with previous reports.^11,13 Next, we analysed the consequence of TTN-AS1-276 knockdown on N2B and N2BA protein isoforms using gel electrophoresis. As previously reported,³⁰ iPS-CM expressed a longer, foetal-like N2BA isoform and N2B expression was considerably lower than in adult cardiac tissue (Figure 3D). Nevertheless, in line with our observation on the mRNA level, we found N2B protein expression to be significantly lower in iPS-CM where TTN-AS1-276 had been knocked down but saw no effect on N2BA expression. This was also reflected in a significantly increased N2BA:N2B ratio. Again, the effect of RBM20 knockdown had a similar effect on N2B expression and the N2BA:N2B ratio (Figure 3F). The Cronos TTN isoform, which is transcribed from an internal promoter in the distal I-band region (between exons 239 and 240 of the inferred meta-transcript), was neither affected by knockdown of TTN-AS1-276 nor RBM20 (Figure 3G). We conclude that TTN-AS1-276 regulates TTN isoform composition and can be targeted to promote translation of longer and more extensible TTN.

*TTN-AS1-276* knockdown causes a shift in TTN isoform composition. (A) Depiction of alternative splicing pathways from *TTN* exon 49 producing the two major cardiac isoforms, N2BA and N2B. (B) Quantification of N2BA and N2B isoforms in iPS-CM transfected with si276-Ex1, si276-Ex12, siRBM20, or siScr, using custom qRT–PCR assays spanning the indicated exon–exon junctions. Expression data are normalized to that of total *TTN* and expressed relative to the mean of the negative control cells (siScr). Data are derived from two separate experiments with two replicates (individual RNA preparations) per experimental group, ***P < 0.001. (C) The N2BA to N2B ratio in iPS-CM following transfection with si276-Ex1, si276-Ex12, siRBM20, or siScr, derived from data in (B). (D) Gel electrophoresis of protein from iPS-CM transfected with siScr, siRBM20, or siTTN-AS1. A sample of mouse cardiac protein (mCM) was run in one lane as a molecular ruler. Molecular weights of the major TTN isoforms (N2BA, N2B, and Cronos) are indicated. T2, TTN degradation product. The band corresponding to MYH7 is shown in the bottom. (E and G) Quantification of band intensity for N2BA, N2B, and Cronos relative to MYH7. (F) Ratio of N2BA to N2B. Data are derived from three separate experiments. Differences between each individual experimental group and the control group were assessed with Student’s t-tests *P < 0.05, **P < 0.01.

3.6. TTN-AS1-276 regulates sarcomere function and cardiomyocyte contraction dynamics

TTN isoform composition is a key determinant of cardiomyocyte mechanical properties. As knockdown of TTN-AS1-276 resulted in increased inclusion of I-band exons and a shift towards longer and more elastic TTN, we wanted to investigate the consequences of TTN-AS1-276 knockdown on sarcomere structure and dynamics. To this end, we applied sarcomere tracking analysis on live iPS-CM after transfection with si276-Ex1 or siRBM20. Cells were transfected with a plasmid encoding GFP-tagged alpha-actinin-2 (pACTN2-GFP) to visualize sarcomere Z-discs (Figure 4A) and recorded during multiple contractions with live cell imaging (see Supplementary material online, Video S1). SarcGraph v.0.2.1 was used to detect, track, and analyse functional parameters of 14 636 sarcomeres from >200 cells during contraction (Figure 4B–H). Results showed that sarcomere length was increased by ∼10% (P < 0.001) in cells where TTN-AS1-276 had been knocked down (Figure 4E), whereas the sarcomere alignment, as measured by the orientational order parameter, was unaffected (see Supplementary material online, Figure S7). Expectedly, knockdown of RBM20 caused a similar increase in sarcomere length (P < 0.001). Fractional shortening (FS) was increased from 7.5% in control cells to 9% in cells transfected with si276-Ex1 (P < 0.001), indicating increased contractility (Figure 4F) and both mean contraction (Figure 4G) and mean relaxation (Figure 4H) time increased by ∼30–40% (P < 0.001) following TTN-AS1-276 knockdown. Knockdown of RBM20 mirrored these effects as well. We conclude that knockdown of TTN-AS1-276 results in longer and more compliant sarcomeres with improved contractile properties.

*TTN-AS1* knockdown results in increased sarcomere length and altered contraction dynamics. (A) Representative fluorescent image of a human iPS-derived cardiomyocyte transfected with a ACTN2-GFP plasmid for visualization and tracking of sarcomeres in SarcGraph software. Visualization of (B) Z-disc and (C) sarcomere segmentation in SarcGraph software. (D) Time series data for average deformation and normalized sarcomere length during three contractions. (E) Sarcomere length (SL), (F) fractional shortening (FS), (G) contraction time, and (H) relaxation time from a total of 14 636 segmented iPS-CM sarcomeres following transfection with si276-Ex1, siRBM20, or siScr. Data are derived from two separate experiments. Differences between each individual experimental group and the control group were assessed with Student’s t-tests, ****P < 0.0001.

3.7. No impact of overlapping antisense transcript exons on TTN exon usage and Pol II elongation rate

Next, we wanted to explore the mechanism by which TTN-AS1 affects alternative splicing of TTN. Morrissy et al.¹⁶ hypothesized that a reason for the increased alternative splicing observed in exons with overlapping antisense exons is that polymerase elongation rates are decreased, as slower RNA polymerase II (Pol II) rates have been shown to increase the rate of alternative splicing.³³ When considering exons from all annotated TTN-AS1 and TTN transcript isoforms in GENCODE, ∼16% of TTN exons overlap with one or more TTN-AS1 exons (red bars in Supplementary material online, Figure S8A). Apart from the Z-disc, overlapping exons are present throughout the whole TTN gene. If the presence of an antisense exon was to promote exon skipping in TTN, we expected usage of such TTN exons to be lower compared to those without an antisense exon. To analyse this, we leveraged DEXSeq exon usage data from human left ventricular tissue without heart disease (n = 7), but the results revealed instead a significantly increased usage of TTN exons with an overlapping antisense exon (see Supplementary material online, Figure S8A and B), contradicting this hypothesis. Moreover, we observed no enrichment of exons with an overlapping antisense exon among the 26 TTN exons with increased PSI following TTN-AS1 knockdown (12% vs. 16% in TTN overall).

We next investigated Pol II occupancy across a selection of TTN exons with and without TTN-AS1 exon overlap in iPS-CM using ChIP-PCR. We designed primers for six pairs of adjacent TTN exons where one exon had an overlapping antisense exon and the other did not (see Supplementary material online, Figure S8C), across different TTN domains and reflecting varying PSIs (see Supplementary material online, Figure S8D). With the exception of exon 48, for which no Pol II signal could be detected, there was no robust or meaningful difference in Pol II occupancy at exons with overlapping antisense exons compared to adjacent exons without overlapping antisense exons (see Supplementary material online, Figure S8D). Moreover, inhibition of Pol II elongation by addition of DRB did not affect the degree of TTN-AS1 chromatin enrichment in iPS-CM, indicating that Pol II pausing does not influence TTN-AS1 tethering to chromatin (see Supplementary material online, Figure S8E). Taken together, these data show that TTN-AS1 influences TTN exon usage/alternative splicing in a manner independent of specific sense–antisense exon overlap and Pol II elongation rate.

3.8. TTN-AS1 facilitates interaction between RBM20 and TTN mRNA

We then wanted to explore alternative mechanisms whereby TTN-AS1 could regulate alternative splicing of TTN. Antisense transcripts have previously been reported to influence splicing either by forming an RNA-duplex with complementary sequences in the coding transcript, thereby masking splice sites in pre-mRNA of the sense gene¹⁷ or by recruiting or guiding specific components of the spliceosome to the sense gene pre-mRNA.^18,19 We reasoned that TTN-AS1-276 could exert its function through either of these mechanisms, but given the similar patterns in TTN ΔPSI, TTN isoform composition and functional consequences on sarcomere dynamics in iPS-CM following knockdown of TTN-AS1-276 and RBM20, we believed that a mechanism whereby TTN-AS1-276 recruits RBM20 to TTN mRNA would be more plausible. To explore these hypotheses, we analysed the quantity and nuclear localization of TTN-AS1, TTN mRNA, and RBM20 protein in iPS-CM using combined RNA ISH and immunofluorescence. First, we found that nuclear TTN:TTN-AS1 co-localization, indicative of duplex formation, was rare, occurring once in every fifth cell on average, and interestingly, was then almost exclusively (>90% of instances) observed as part of a cluster involving RBM20 (Figure 5A and Supplementary material online, Figure S9A), with a fluorescence pattern indicative of RBM20-mediated TTN splicing.^31,34 This observation contradicts a mechanism involving direct TTN:TTN-AS1 duplex formation and strengthens the hypothesis that TTN-AS1 regulates TTN splicing via RBM20. To test this hypothesis further, we quantified the proportion of TTN RNA ISH foci co-localized with RBM20 in iPS-CM transfected with si276-Ex1 using high content imaging analysis. In control cells, we estimated that ∼7.5% of all nuclear TTN RNA ISH foci co-localized with RBM20 (Figure 5B). Interestingly, in iPS-CM where TTN-AS1-276 had been knocked down, the proportion of TTN foci co-localized with RBM20 was more than halved (P < 0.0001, Figure 5C). This suggests that TTN-AS1-276 facilitates interaction between RBM20 and TTN pre-mRNA. We then sought to corroborate these findings using RIP on iPS-CM. iPS-CM was transfected with a plasmid encoding a RBM20-GFP fusion protein (pRBM20-GFP), and a GFP antibody was then used to pull down the RBM20 RNA interactome (Figure 5D). Transfection with pRBM20-GFP resulted in transgenic expression of RBM20 fusion protein (see Supplementary material online, Figure S10A). We observed significant enrichment of both TTN and TTN-AS1-276 in RBM20-RIP RNA compared to unrelated GAPDH RNA (Figure 5E) and to negative control IgG IP (see Supplementary material online, Figure S10B). Expectedly, the TTN-AS1-276 RIP signal was significantly reduced in cells where TTN-AS1-276 had been knocked down. Interestingly, there was also significantly less TTN interacting with RBM20 after TTN-AS1-276 knockdown, giving additional support for a mechanism whereby TTN-AS1 facilitates interaction between RBM20 and TTN mRNA (Figure 5E).

*TTN-AS1-276* facilitates interaction between RBM20 and *TTN* mRNA. (A–C) Human iPS-derived cardiomyocytes (iPS-CM) were subjected to combined RNA *in situ* hybridization (ISH) for *TTN-AS1-276* (magenta) and *TTN* (orange) and immunofluorescence for RBM20 (green) following transfection with siRNA towards *TTN-AS1-276* (si276-Ex1), *RBM20* (siRBM20), or scrambled negative control siRNA (siScr). Nuclei were counterstained with DAPI. Co-localization of *TTN* and RBM20 foci was analysed with high content imaging. (D) Depiction of experimental design for the RBM20 RNA immunoprecipitation (RIP) experiment. iPS-CM was transfected with a plasmid expressing a RBM20-GFP fusion protein. RIP was performed on iPS-CM protein using a GFP antibody and qRT–PCR was used to analyse immunoprecipitated RNA. (E) Enrichment of *TTN* and *TTN-AS1-276* in GFP-RBM20 RIP RNA from iPS-CM transfected with si276-Ex1 or siScr, analysed with qRT–PCR. Analysis of unrelated *GAPDH* RNA was included as a negative control. Data are derived from two separate experiments with two technical replicates (individual immunoprecipitates) in each group *P < 0.05, **P < 0.01. (F) The number of TCTT motifs/60 bp across intron 49 of *TTN*. The positions of RIP-qPCR assays used in (G) are indicated with dashed lines. (G) qPCR of GFP-RBM20 RIP RNA from (E) using assays targeting regions in intron 49 without TCTT motifs (‘In49 5p’) and enriched with TCTT motifs (‘In49 TCTT’). Data are derived from two separate experiments with two technical replicates (individual immunoprecipitates) in each group. Differences in the RIP signal between cells transfected within and between groups were assessed using ANOVA with Dunnett’s multiple comparisons test, *P < 0.05, **P < 0.01.

RBM20 mediates exon skipping by interacting with intronic TCTT motifs in pre-mRNA.¹³ We hypothesized that TTN-AS1-276 assists in the binding of RBM20 to such intronic motifs upstream of skipped exons. To test this hypothesis, we chose to focus on intron 49, with which RBM20 would interact for skipping of exon 50 to occur. We scanned the sequence of intron 49 and found an enrichment of TCTT motifs in the middle of the intron and towards the 3′ end, whereas the 5′ end of the intron was completely devoid of TCTT motifs (Figure 5F). To assess binding of RBM20 to different regions of intron 49, we designed qPCR assays targeting a 5′ region free from TCTT motifs (‘In49 5p’) and a region enriched in TCTT motifs in the middle of the intron (‘In49 TCTT’). We then performed RBM20-RIP qPCR using these assays and expectedly, found that RBM20 was significantly enriched in the TCTT-rich region (Figure 5G). Interestingly, RBM20 binding to the TCTT-rich region was significantly reduced upon knockdown of TTN-AS1-276 (Figure 5G). Taken together, these results point towards a mechanism whereby TTN-AS1-276 mediates exon skipping in TTN through interacting with and recruiting RBM20 to intronic TCTT motifs. We then sought to explore the mode-of-action for the protein–RNA interaction and hypothesized that RBM20 could bind directly to TTN-AS1-276 through its RRM. To test this, we performed an in-frame deletion of the RRM-coding exons in the pRBM20-GFP construct using SDM (see Supplementary material online, Figure S11A). We then transfected iPS-CM with the resulting plasmid (pRBM20ΔRRM-GFP) and performed RIP. We observed a near-complete loss of the TTN-AS1-276 RIP signal in cells transfected with pRBM20ΔRRM-GFP (see Supplementary material online, Figure S11B), suggesting that RBM20 binds TTN-AS1-276 directly through the RRM domain.

3.9. TTN-AS1 mediates splicing of other RBM20 targets

In a previous study, Bertero et al.³⁴ postulated that RBM20 forms a trans-interacting chromatin domain (TID), driving spatial proximity of multiple genomic loci representing a network of different RBM20 targets. The assembly of this splicing complex is initiated at the TTN genomic locus, where transcription of TTN nucleates RBM20 foci, which drives formation of the TID. Given the overlap and spatial proximity of TTN-AS1-276 and TTN loci, the evidence of physical interaction of TTN-AS1-276 and TTN RNA with RBM20 protein and the observation that the number of RBM20 protein foci decreased in iPS-CM after TTN-AS1-276 knockdown (see Supplementary material online, Figure S9B), we hypothesized that TTN-AS1-276 might facilitate the assembly of the RBM20 splicing factory. If this was the case, knockdown of TTN-AS1-276 would affect splicing of not just TTN, but also of other RBM20 targets. To explore this hypothesis, we first analysed PSI data from iPS-CM where either TTN-AS1-276 or RBM20 had been knocked down, focusing on three genes shown to be included in the RBM20 TID: CACNA1C, CAMK2D, and LMO7 (see Supplementary material online, Figure S12A). In line with previous reports, there was significantly different ΔPSI for exons 8 and 30 in CACNA1C (Figure 6A), exons 9–11 in LMO7 (Figure 6B), and exon 14 in CAMK2D (Figure 6C) following RBM20 knockdown. Interestingly, this effect was mirrored almost exactly in cells where TTN-AS1-276 had been knocked down. Guided by these results, we quantified splicing products using qPCR assays spanning isoform-specific exon–exon junctions (CACNA1C and LMO7) or with RT–PCR (CAMK2D). For CACNA1C, we saw a significant down-regulation of the alternative exon 8a relative to exon 8 in both cells transfected with siRBM20 and si276-Ex1 (Figure 6D). We also observed significant down-regulation of LMO7 isoforms subject to exons 9 and 10 skipping following knockdown of both siRBM20 and si276-Ex1 (Figure 6E). Finally, we detected a significant shift from CAMK2D-C, which requires skipping of exons 14–16, to the CAMK2D-B isoform, which requires skipping of exon 15, and unspliced CAMK2D, in both experimental groups (Figure 6F and Supplementary material online, Figure S12B). Results were overall strikingly similar between siRBM20 and si276-Ex1, with the only exception of the CAMK2D-9 isoform, where knockdown of RBM20 caused a significant decrease, and knockdown of TTN-AS1-276 instead caused a significant increase. Next, we confirmed the interaction between RBM20 protein and CACNA1C, LMO7, and CAMK2D RNA by RIP-qPCR (Figure 6G). All three genes showed significant enrichment compared to negative control RNA (GAPDH) in RBM20-RIP RNA and compared to the signal from non-specific IgG RIP (see Supplementary material online, Figure S12C). As expected, the RIP signals for all three genes were almost completely abolished upon TTN-AS1-276 knockdown (Figure 6G). In all, these results support a mechanism where TTN-AS1-276 is a component of the RBM20 splicing machinery and abets alternative splicing of RBM20 target genes in cardiomyocytes.

*TTN-AS1-276* facilitates splicing of additional RBM20 targets. (A–C) Difference in PSI (ΔPSI) comparing cells transfected with siRNA to *TTN-AS1-276* (si276-Ex1, blue) or *RBM20* (siRBM20, red) with cells transfected with scrambled negative control siRNA (siScr) across all exons of the (A) *CACNA1C*, (B) *LMO7*, and (C) *CAMK2D* genes in human iPS-derived cardiomyocytes (iPS-CM). The lines represent the mean of three technical replicates (individual RNA preparations). (D–F) Relative expression of alternative splice products of the (D) *CACNA1C*, (E) *LMO7*, and (F) *CAMK2D* genes in iPS-CM transfected with siRBM20, si276-Ex1 or siScr and analysed with qRT–PCR (*CACNA1C* and *LMO7*) and semi-quantitative RT–PCR (*CAMK2D*), respectively. Data are derived from three separate experiments with three technical replicates (individual RNA preparations) in each group. Differences between each individual experimental group and the control group were assessed with Student’s t-tests, *P < 0.05, **P < 0.01, ***P < 0.001. (G) Enrichment of *CACNA1C*, *LMO7*, and *CAMK2D* in GFP-RBM20 RIP RNA from iPS-CM transfected with si276-Ex1 or siScr, analysed with qRT–PCR. Analysis of unrelated *GAPDH* RNA was included as a negative control. Data are derived from two separate experiments.

3.10. TTN-AS1-276 is induced in human heart failure and correlates with TTN I-band exon usage

Cardiac TTN expression is shifted towards longer and more compliant isoforms in heart failure (HF), but the mechanistic basis for this isoform switch is not known. Given the apparent role of TTN-AS1-276 in splicing of extensible I-band exons and regulating sarcomere function in iPS-CM, we sought to analyse its expression and function in human HF. First, we confirmed that RBM20 interacts with TTN-AS1-276 in human heart tissue by performing RBM20 RIP on explanted cardiac biopsies from two HF patients (Figure 7A). Then, in order to assess the expression of TTN-AS1-276 and its association with TTN splicing in human HF, we leveraged cardiac RNA-sequencing and qPCR data from a sample of explanted cardiac biopsies from HF patients (n = 100) and unused donor hearts (n = 7). Patient characteristics are described in Table 1. We first assessed the expression of TTN-AS1-276 with qPCR and observed a marked increase in HF patients compared to controls (Figure 7B). Considering the heterogeneity of the HF patient group, we also compared the expression of TTN-AS1-276 between controls and each of the three major HF aetiologies: dilated cardiomyopathy (DCM, n = 47), ischaemic cardiomyopathy (ICM, n = 19), and hypertrophic cardiomyopathy (HCM, n = 9). Interestingly, increased TTN-AS1-276 expression was only observed in DCM and HCM. Next, we investigated the association between TTN-AS1-276 expression and clinical parameters using linear and logistic regression but found no significant correlation between TTN-AS1-276 and age (r² = 0.014; P = 0.24) or BMI (r² = 0.000032; P = 0.99), nor did we observe any significant associations with sex (β = −0.13; P = 0.7), hypertension (β = −0.67; P = 0.14), or left ventricular assist device implantation (β = −0.16; P = 0.76).

*TTN-AS1-276* expression and correlation with *TTN* splicing in human heart failure. (A) RIP was performed on explanted cardiac tissue from two heart failure patients using antibodies against RBM20 or Poly(A)-binding protein (PABP, positive control) or a non-specific IgG antibody (negative control). qRT–PCR was performed to quantify *TTN* and *TTN-AS1-276* in immunoprecipitated RNA. The RIP signal is expressed relative to the 10% input sample. Data points represent the mean of two technical replicates (individual immunoprecipitates) from each patient (n = 2). (B) Quantification of cardiac *TTN-AS1-276* in a cohort of HF patients (n = 100) and unused donor hearts (controls, n = 7) using qRT–PCR. Results from the three largest patient subgroups within the HF group, dilated cardiomyopathy (DCM), ischaemic cardiomyopathy (ICM), and hypertrophic cardiomyopathy (HCM) are shown separately to the right. Expression data are normalized to that of total *TTN* and expressed relative to the mean of the controls, *P < 0.05. (C) Correlation coefficients from linear regression analyses of *TTN-AS1-276* expression (quantified by qPCR) and PSI for each *TTN* exon in 100 HF patients and seven controls. Statistically significant correlations after Bonferroni correction are shown in red. Exons 49–90, which were among the most affected by *TTN-AS1-276* knockdown in iPS-CM, are also highlighted.

To assess whether TTN-AS1-276 influences splicing of TTN I-band exons in human hearts, we used cardiac RNA-sequencing data from the HF cohort to calculate PSI values for each TTN exon (see Supplementary material online, Figure S13) and analysed the correlation between PSI and TTN-AS1-276 expression (Figure 7C). After correction for multiple testing, we found a significant negative correlation between TTN-AS1-276 expression and the PSI of 44 exons, all situated in the I-band. We found no significant positive correlations, nor any significant correlations with exons outside the I-band. These results are in line with our findings in vitro, i.e. that TTN-AS1-276 inhibits inclusion of I-band exons, and provides support for the fact that this mechanism extends to the adult human heart.

4. Discussion

In this study, we comprehensively map TTN antisense transcription in human heart tissue and define a functional role for the most abundant transcript, TTN-AS1-276, in alternative splicing of TTN. Based on several lines of evidence from different experimental methods, we postulate that TTN-AS1-276 interacts with RBM20 to facilitate exon skipping in I-band exons of TTN. While it is well established that RBM20 represses exon inclusion in TTN and other cardiac genes,¹³ the mechanism by which RBM20 is guided to target pre-mRNA is not well studied. Recently, the splicing factor Srsf1 was shown to be recruited to Triadin (Trdn) pre-mRNA via the Triadin antisense transcript (Trdn-as) in cardiomyocytes.¹⁹ Knock out of Trdn-as resulted in dysregulated Trdn isoform composition, aberrant Ca²⁺-handling and susceptibility to arrhythmias. Our results points to a similar mechanism, where TTN-AS1-276 facilitates interaction between RBM20 and TTN pre-mRNA. While we show that TTN-AS1-276 co-localizes with both RBM20 and TTN mRNA in cardiomyocyte nuclei, that knockdown of TTN-AS1-276 results in decreased interaction between RBM20 and intronic RBM20-binding motifs in TTN and that TTN-AS1-276 seems to interact with RBM20 via its RRM domain, elucidating the structural foundation for this mechanism requires further studies. The RBM20 RRM interacts with intronic UCUU-motifs in target pre-mRNAs¹⁴ and based on the fact that TTN-AS1-276 has a comparable density of intronic UCUU-motifs (mean of 9.7/kb across all introns) to introns spanning alternatively spliced exons in the RBM20 target genes studied here, i.e. TTN (16/kb), LMO7 (7.3/kb), CAMK2D (9.9/kb), and CACNA1C (5.1/kb), we speculate that the interaction with TTN-AS1 could involve recognition of such motifs by the RBM20 RRM.

Recently, Bertero et al.³⁴ showed that RBM20 drives genomic reorganization of its target genes through a TID. Here, we show that knockdown of TTN-AS1-276 affects splicing not only of TTN, but several of the other TID-associated RBM20 target genes. The TID includes genes that are involved in cardiomyocyte excitation-contraction coupling (e.g. CACNA1C and CAMK2D) and it is therefore possible that the effects of TTN-AS1-276 knockdown on contractility could in part be due to altered Ca²⁺-flux, in addition to the direct effect on TTN isoform composition and sarcomere properties. Future studies will be warranted to study the role of TTN-AS1-176 in formation and functional consequences of the RBM20 TID beyond TTN splicing. Interestingly, TTN-AS1 has previously been predicted to be part of other gene regulatory networks in human cardiac tissue. Using bioinformatic tools, Tian et al.³⁵ showed that TTN-AS1 was predicted to regulate the Ca²⁺-channel TRPM5 in the context of right ventricular cardiomyopathy induced by tricuspid regurgitation. Moreover, Zhang et al. suggested a potential cardioprotective role for TTN-AS1 through increasing the stability of CDK6 mRNA and alleviating hypoxia-induced apoptosis in cardiomyocytes in vitro. These results must be taken into account when considering TTN-AS1 as a potential therapeutic target in heart disease.³⁶

We believe that the fact that knockdown of TTN-AS1-276 in human iPS-CM resulted in a shift towards longer and less stiff sarcomeres with improved diastolic properties merit further investigation into TTN antisense transcripts as therapeutic target for diseases characterized by increased myocardial stiffness and diastolic dysfunction, such as HFpEF. Genetic models of RBM20 deficiency have been shown to have improved diastolic function,^37,38 and we believe that TTN-AS1 could represent a direct target for regulating RBM20 activity through antisense oligonucleotide (ASO) therapies. The 5′ end of TTN-AS1-276 does not overlap TTN exons (nor those of any other gene) and thus represents a feasible target for specific ASO-mediated knockdown. Moreover, the fact that TTN-AS1-276 appears to have its own promoter (see Supplementary material online, Figure S2B) means that it is amenable to transcriptional modulation by CRISPRi.³⁹ Additional studies are required to assess whether interfering with Ttn antisense transcription can be harnessed to modulate sarcomere properties and improve cardiac function in in vivo models of diastolic dysfunction and HFpEF. One annotated antisense transcript has been identified in the mouse Ttn locus (ENSMUST00000156809). As commonly observed among antisense transcripts,⁴⁰ overall sequence conservation with human TTN antisense transcripts is low, but it is interesting to note that exon 2 of the mouse transcript has 94% similarity to exon 2 of TTN-AS1-276. An important aspect of therapeutically reducing TTN stiffness has been raised in genetic models of RBM20 loss-of-function. While homozygous deletion of the RBM20 RRM domain caused increased sarcomere length and improved diastolic function, there was a concomitant reduction in maximal systolic stress and a depression of length-dependent activation.³⁸ A rigorous evaluation of the therapeutic window and careful dosing of any ASOs targeting TTN antisense transcripts will therefore be necessary.

A limitation of the study is the lack of an adult human cardiomyocyte in vitro model. iPS-CMs are known to exhibit a foetal phenotype⁴¹ and have previously been shown to predominantly express longer N2BA isoforms and a considerably higher N2BA:N2B ratio than adult cardiomyocytes.³⁰ However, in our live cell imaging experiments, we observe sarcomere length and contraction dynamics that are comparable to adult cardiomyocytes.

In conclusion, we show that antisense transcripts play an integral role in regulation of TTN alternative splicing and sarcomere function in cardiomyocytes and could constitute targets for therapeutic modulation of cardiac stiffness.

Supplementary Material

cvaf037_Supplementary_Data

cvaf037_supplementary_data.zip^{(28.4MB, zip)}

Acknowledgements

The authors thank Lund University Bioimaging Centre (LBIC), MultiPark, Center for Translational Genomics and the National Bioinformatics Infrastructure Sweden (NBIS) at SciLifeLab for support.

Contributor Information

Selvi Celik, Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden; Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden.

Ludvig Hyrefelt, Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden.

Tomasz Czuba, Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden; Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden; The Wallenberg Laboratory/Department of Molecular and Clinical Medicine, Institute of Medicine, Gothenburg University and The Department of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden; Science for Life Laboratory, Gothenburg University, Gothenburg, Sweden.

Yuan Li, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Lund University, Lund, Sweden; Department of Immunotechnology, Lund University, Lund, Sweden.

Juliana Assis, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Lund University, Lund, Sweden.

Julia Martinez, Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden; Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden.

Markus Johansson, Systems Biology Research Center, School of Bioscience, University of Skövde, Skövde, Sweden.

Oscar André, Division of Infection Medicine, Department of Clinical Sciences, Lund University, Lund, Sweden.

Jane Synnergren, Systems Biology Research Center, School of Bioscience, University of Skövde, Skövde, Sweden; Department of Molecular and Clinical Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.

Joakim Sandstedt, Department of Laboratory Medicine, Institute of Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.

Pontus Nordenfelt, Division of Infection Medicine, Department of Clinical Sciences, Lund University, Lund, Sweden.

Kristina Vukusic, Department of Laboratory Medicine, Institute of Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.

J Gustav Smith, Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden; Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden; The Wallenberg Laboratory/Department of Molecular and Clinical Medicine, Institute of Medicine, Gothenburg University and The Department of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden; Science for Life Laboratory, Gothenburg University, Gothenburg, Sweden.

Olof Gidlöf, Division of Cardiology, Department of Clinical Sciences, Lund University, BMC D12, Solvegatan 19, Lund SE-221 84, Sweden; Wallenberg Centre for Molecular Medicine and Lund University Diabetes Centre, Lund University, Lund, Sweden.

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Authors’ contributions

S.C., L.H., J.G.S., and O.G. conceived the study. S.C., L.H., J.M., and O.G. designed and performed experiments. T.C., Y.L., and J.A. performed bioinformatic and statistical analysis. M.J., O.A., Ja.S., Jo.S., P.N., and K.V. contributed with analytical equipment, biological samples, and technical expertise. S.C., L.H., J.G.S., and O.G. analysed and interpreted results. O.G. wrote the manuscript. All co-authors contributed to the manuscript and approved of it before submission.

Funding

This work was supported by grants from the Swedish Heart and Lung Foundation (#20220344, #2023033824, and #2023033924), the Crafoord Foundation and the Royal Physiographic Society. J.G.S. was also supported by grants from the Swedish Research Council (2021-02273), the European Research Council (ERC-STG-2015-679242), Gothenburg University, Skåne University Hospital, governmental funding of clinical research within the Swedish National Health Service, a generous donation from the Knut and Alice Wallenberg foundation to the Wallenberg Center for Molecular Medicine in Lund, and funding from the Swedish Research Council (Linnaeus grant Dnr 349-2006-237, Strategic Research Area Exodiab Dnr 2009-1039) and Swedish Foundation for Strategic Research (Dnr IRC15-0067) to the Lund University Diabetes Center.

Data availability

All data are available upon reasonable request.

References

1. Loescher CM, Hobbach AJ, Linke WA. Titin (TTN): from molecule to modifications, mechanics, and medical significance. Cardiovasc Res 2022;118:2903–2918. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Furst DO, Osborn M, Nave R, Weber K. The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the Z line extends close to the M line. J Cell Biol 1988;106:1563–1572. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Granzier H, Helmes M, Cazorla O, McNabb M, Labeit D, Wu Y, Yamasaki R, Redkar A, Kellermayer M, Labeit S, Trombitas K. Mechanical properties of titin isoforms. Adv Exp Med Biol 2000;481:283–300. discussion 300–4. [DOI] [PubMed] [Google Scholar]
4. Hessel AL, Ma W, Mazara N, Rice PE, Nissen D, Gong H, Kuehn M, Irving T, Linke WA. Titin force in muscle cells alters lattice order, thick and thin filament protein formation. Proc Natl Acad Sci U S A 2022;119:e2209441119. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Wu Y, Cazorla O, Labeit D, Labeit S, Granzier H. Changes in titin and collagen underlie diastolic stiffness diversity of cardiac muscle. J Mol Cell Cardiol 2000;32:2151–2162. [DOI] [PubMed] [Google Scholar]
6. Nagueh SF, Shah G, Wu Y, Torre-Amione G, King NM, Lahmers S, Witt CC, Becker K, Labeit S, Granzier HL. Altered titin expression, myocardial stiffness, and left ventricular function in patients with dilated cardiomyopathy. Circulation 2004;110:155–162. [DOI] [PubMed] [Google Scholar]
7. Radke MH, Peng J, Wu Y, McNabb M, Nelson OL, Granzier H, Gotthardt M. Targeted deletion of titin N2B region leads to diastolic dysfunction and cardiac atrophy. Proc Natl Acad Sci U S A 2007;104:3444–3449. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Granzier HL, Radke MH, Peng J, Westermann D, Nelson OL, Rost K, King NM, Yu Q, Tschope C, McNabb M, Larson DF, Labeit S, Gotthardt M. Truncation of titin’s elastic PEVK region leads to cardiomyopathy with diastolic dysfunction. Circ Res 2009;105:557–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Herman DS, Lam L, Taylor MRG, Wang L, Teekakirikul P, Christodoulou D, Conner L, DePalma SR, McDonough B, Sparks E, Teodorescu DL, Cirino AL, Banner NR, Pennell DJ, Graw S, Merlo M, Di Lenarda A, Sinagra G, Bos JM, Ackerman MJ, Mitchell RN, Murry CE, Lakdawala NK, Ho CY, Barton PJ, Cook SA, Mestroni L, Seidman JG, Seidman CE. Truncations of titin causing dilated cardiomyopathy. N Engl J Med 2012;366:619–628. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Freiburg A, Trombitas K, Hell W, Cazorla O, Fougerousse F, Centner T, Kolmerer B, Witt C, Beckmann JS, Gregorio CC, Granzier H, Labeit S. Series of exon-skipping events in the elastic spring region of titin as the structural basis for myofibrillar elastic diversity. Circ Res 2000;86:1114–1121. [DOI] [PubMed] [Google Scholar]
11. Guo W, Schafer S, Greaser ML, Radke MH, Liss M, Govindarajan T, Maatz H, Schulz H, Li S, Parrish AM, Dauksaite V, Vakeel P, Klaassen S, Gerull B, Thierfelder L, Regitz-Zagrosek V, Hacker TA, Saupe KW, Dec GW, Ellinor PT, MacRae CA, Spallek B, Fischer R, Perrot A, Ozcelik C, Saar K, Hubner N, Gotthardt M. RBM20, a gene for hereditary cardiomyopathy, regulates titin splicing. Nat Med 2012;18:766–773. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Gregorich ZR, Zhang Y, Kamp TJ, Granzier HL, Guo W. Mechanisms of RBM20 cardiomyopathy: insights from model systems. Circ Genom Precis Med 2024;17:e004355. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Maatz H, Jens M, Liss M, Schafer S, Heinig M, Kirchner M, Adami E, Rintisch C, Dauksaite V, Radke MH, Selbach M, Barton PJR, Cook SA, Rajewsky N, Gotthardt M, Landthaler M, Hubner N. RNA-binding protein RBM20 represses splicing to orchestrate cardiac pre-mRNA processing. J Clin Invest 2014;124:3419–3430. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Dauksaite V, Gotthardt M. Molecular basis of titin exon exclusion by RBM20 and the novel titin splice regulator PTB4. Nucleic Acids Res 2018;46:5227–5238. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Katayama S, Tomaru Y, Kasukawa T, Waki K, Nakanishi M, Nakamura M, Nishida H, Yap CC, Suzuki M, Kawai J, Suzuki H, Carninci P, Hayashizaki Y, Wells C, Frith M, Ravasi T, Pang KC, Hallinan J, Mattick J, Hume DA, Lipovich L, Batalov S, Engstrom PG, Mizuno Y, Faghihi MA, Sandelin A, Chalk AM, Mottagui-Tabar S, Liang Z, Lenhard B, Wahlestedt C; RIKEN Genome Exploration Research Group; Genome Science Group (Genome Network Project Core Group); FANTOM Consortium . Antisense transcription in the mammalian transcriptome. Science 2005;309:1564–1566. [DOI] [PubMed] [Google Scholar]
16. Morrissy AS, Griffith M, Marra MA. Extensive relationship between antisense transcription and alternative splicing in the human genome. Genome Res 2011;21:1203–1212. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Beltran M, Puig I, Pena C, Garcia JM, Alvarez AB, Pena R, Bonilla F, de Herreros AG. A natural antisense transcript regulates Zeb2/Sip1 gene expression during Snail1-induced epithelial–mesenchymal transition. Genes Dev 2008;22:756–769. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Xu S, Wang P, Zhang J, Wu H, Sui S, Zhang J, Wang Q, Qiao K, Yang W, Xu H, Pang D. Ai-lncRNA EGOT enhancing autophagy sensitizes paclitaxel cytotoxicity via upregulation of ITPR1 expression by RNA–RNA and RNA–protein interactions in human cancer. Mol Cancer 2019;18:89. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Zhao Y, Riching AS, Knight WE, Chi C, Broadwell LJ, Du Y, Abdel-Hafiz M, Ambardekar AV, Irwin DC, Proenza C, Xu H, Leinwand LA, Walker LA, Woulfe KC, Bristow MR, Buttrick PM, Song K. Cardiomyocyte-specific long noncoding RNA regulates alternative splicing of the triadin gene in the heart. Circulation 2022;146:699–714. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Werner MS, Ruthenburg AJ. Nuclear fractionation reveals thousands of chromatin-tethered noncoding RNAs adjacent to active genes. Cell Rep 2015;12:1089–1098. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Hall DD, Dai S, Tseng P-Y, Malik Z, Nguyen M, Matt L, Schnizler K, Shephard A, Mohapatra DP, Tsuruta F, Dolmetsch RE, Christel CJ, Lee A, Burette A, Weinberg RJ, Hell JW. Competition between alpha-actinin and Ca²⁺-calmodulin controls surface retention of the L-type Ca²⁺ channel Ca_V1.2. Neuron 2013;78:483–497. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Greaser ML, Warren CM. Method for resolution and western blotting of very large proteins using agarose electrophoresis. Methods Mol Biol 2015;1312:285–291. [DOI] [PubMed] [Google Scholar]
23. Zhao B, Zhang K, Chen CS, Lejeune E. Sarc-Graph: automated segmentation, tracking, and analysis of sarcomeres in hiPSC-derived cardiomyocytes. PLoS Comput Biol 2021;17:e1009443. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Anders S, Reyes A, Huber W. Detecting differential usage of exons from RNA-seq data. Genome Res 2012;22:2008–2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Reyes A, Anders S, Weatheritt RJ, Gibson TJ, Steinmetz LM, Huber W. Drift and conservation of differential exon usage across tissues in primate species. Proc Natl Acad Sci U S A 2013;110:15377–15382. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Zhang Y, Hardison RC. Accurate and reproducible functional maps in 127 human cell types via 2D genome segmentation. Nucleic Acids Res 2017;45:9823–9836. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Eraslan G, Drokhlyansky E, Anand S, Fiskin E, Subramanian A, Slyper M, Wang J, Van Wittenberghe N, Rouhana JM, Waldman J, Ashenberg O, Lek M, Dionne D, Win TS, Cuoco MS, Kuksenko O, Tsankov AM, Branton PA, Marshall JL, Greka A, Getz G, Segre AV, Aguet F, Rozenblatt-Rosen O, Ardlie KG, Regev A. Single-nucleus cross-tissue molecular reference maps toward understanding disease gene function. Science 2022;376:eabl4290. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Zhao S, Zhang X, Chen S, Zhang S. Natural antisense transcripts in the biological hallmarks of cancer: powerful regulators hidden in the dark. J Exp Clin Cancer Res 2020;39:187. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Pelechano V, Steinmetz LM. Gene regulation by antisense transcription. Nat Rev Genet 2013;14:880–893. [DOI] [PubMed] [Google Scholar]
30. Hinson JT, Chopra A, Nafissi N, Polacheck WJ, Benson CC, Swist S, Gorham J, Yang L, Schafer S, Sheng CC, Haghighi A, Homsy J, Hubner N, Church G, Cook SA, Linke WA, Chen CS, Seidman JG, Seidman CE. HEART DISEASE. Titin mutations in iPS cells define sarcomere insufficiency as a cause of dilated cardiomyopathy. Science 2015;349:982–986. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Li S, Guo W, Dewey CN, Greaser ML. Rbm20 regulates titin alternative splicing as a splicing repressor. Nucleic Acids Res 2013;41:2659–2672. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Guo W, Bharmal SJ, Esbona K, Greaser ML. Titin diversity—alternative splicing gone wild. J Biomed Biotechnol 2010;2010:753675. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. de la Mata M, Alonso CR, Kadener S, Fededa JP, Blaustein M, Pelisch F, Cramer P, Bentley D, Kornblihtt AR. A slow RNA polymerase II affects alternative splicing in vivo. Mol Cell 2003;12:525–532. [DOI] [PubMed] [Google Scholar]
34. Bertero A, Fields PA, Ramani V, Bonora G, Yardimci GG, Reinecke H, Pabon L, Noble WS, Shendure J, Murry CE. Dynamics of genome reorganization during human cardiogenesis reveal an RBM20-dependent splicing factory. Nat Commun 2019;10:1538. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Tian C, Yang Y, Ke Y, Yang L, Zhong L, Wang Z, Huang H. Integrative analyses of genes associated with right ventricular cardiomyopathy induced by tricuspid regurgitation. Front Genet 2021;12:708275. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Zhang Y, Shang Z, Xu S, Zhou G, Liu A. ELF5-regulated lncRNA-TTN-AS1 alleviates myocardial cell injury via recruiting PCBP2 to increase CDK6 stability in myocardial infarction. Mol Cell Biol 2024;44:303–315. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Hinze F, Dieterich C, Radke MH, Granzier H, Gotthardt M. Reducing RBM20 activity improves diastolic dysfunction and cardiac atrophy. J Mol Med (Berl) 2016;94:1349–1358. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Methawasin M, Hutchinson KR, Lee E-J, Smith JE 3rd, Saripalli C, Hidalgo CG, Ottenheijm CAC, Granzier H. Experimentally increasing titin compliance in a novel mouse model attenuates the Frank–Starling mechanism but has a beneficial effect on diastole. Circulation 2014;129:1924–1936. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Bendixen L, Jensen TI, Bak RO. CRISPR-Cas-mediated transcriptional modulation: the therapeutic promises of CRISPRa and CRISPRi. Mol Ther 2023;31:1920–1937. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Wood EJ, Chin-Inmanu K, Jia H, Lipovich L. Sense–antisense gene pairs: sequence, transcription, and structure are not conserved between human and mouse. Front Genet 2013;4:183. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Karakikes I, Ameen M, Termglinchan V, Wu JC. Human induced pluripotent stem cell-derived cardiomyocytes: insights into molecular, cellular, and functional phenotypes. Circ Res 2015;117:80–88. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

cvaf037_Supplementary_Data

cvaf037_supplementary_data.zip^{(28.4MB, zip)}

Data Availability Statement

All data are available upon reasonable request.

[cvaf037-B1] 1. Loescher CM, Hobbach AJ, Linke WA. Titin (TTN): from molecule to modifications, mechanics, and medical significance. Cardiovasc Res 2022;118:2903–2918. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B2] 2. Furst DO, Osborn M, Nave R, Weber K. The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the Z line extends close to the M line. J Cell Biol 1988;106:1563–1572. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B3] 3. Granzier H, Helmes M, Cazorla O, McNabb M, Labeit D, Wu Y, Yamasaki R, Redkar A, Kellermayer M, Labeit S, Trombitas K. Mechanical properties of titin isoforms. Adv Exp Med Biol 2000;481:283–300. discussion 300–4. [DOI] [PubMed] [Google Scholar]

[cvaf037-B4] 4. Hessel AL, Ma W, Mazara N, Rice PE, Nissen D, Gong H, Kuehn M, Irving T, Linke WA. Titin force in muscle cells alters lattice order, thick and thin filament protein formation. Proc Natl Acad Sci U S A 2022;119:e2209441119. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B5] 5. Wu Y, Cazorla O, Labeit D, Labeit S, Granzier H. Changes in titin and collagen underlie diastolic stiffness diversity of cardiac muscle. J Mol Cell Cardiol 2000;32:2151–2162. [DOI] [PubMed] [Google Scholar]

[cvaf037-B6] 6. Nagueh SF, Shah G, Wu Y, Torre-Amione G, King NM, Lahmers S, Witt CC, Becker K, Labeit S, Granzier HL. Altered titin expression, myocardial stiffness, and left ventricular function in patients with dilated cardiomyopathy. Circulation 2004;110:155–162. [DOI] [PubMed] [Google Scholar]

[cvaf037-B7] 7. Radke MH, Peng J, Wu Y, McNabb M, Nelson OL, Granzier H, Gotthardt M. Targeted deletion of titin N2B region leads to diastolic dysfunction and cardiac atrophy. Proc Natl Acad Sci U S A 2007;104:3444–3449. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B8] 8. Granzier HL, Radke MH, Peng J, Westermann D, Nelson OL, Rost K, King NM, Yu Q, Tschope C, McNabb M, Larson DF, Labeit S, Gotthardt M. Truncation of titin’s elastic PEVK region leads to cardiomyopathy with diastolic dysfunction. Circ Res 2009;105:557–564. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B9] 9. Herman DS, Lam L, Taylor MRG, Wang L, Teekakirikul P, Christodoulou D, Conner L, DePalma SR, McDonough B, Sparks E, Teodorescu DL, Cirino AL, Banner NR, Pennell DJ, Graw S, Merlo M, Di Lenarda A, Sinagra G, Bos JM, Ackerman MJ, Mitchell RN, Murry CE, Lakdawala NK, Ho CY, Barton PJ, Cook SA, Mestroni L, Seidman JG, Seidman CE. Truncations of titin causing dilated cardiomyopathy. N Engl J Med 2012;366:619–628. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B10] 10. Freiburg A, Trombitas K, Hell W, Cazorla O, Fougerousse F, Centner T, Kolmerer B, Witt C, Beckmann JS, Gregorio CC, Granzier H, Labeit S. Series of exon-skipping events in the elastic spring region of titin as the structural basis for myofibrillar elastic diversity. Circ Res 2000;86:1114–1121. [DOI] [PubMed] [Google Scholar]

[cvaf037-B11] 11. Guo W, Schafer S, Greaser ML, Radke MH, Liss M, Govindarajan T, Maatz H, Schulz H, Li S, Parrish AM, Dauksaite V, Vakeel P, Klaassen S, Gerull B, Thierfelder L, Regitz-Zagrosek V, Hacker TA, Saupe KW, Dec GW, Ellinor PT, MacRae CA, Spallek B, Fischer R, Perrot A, Ozcelik C, Saar K, Hubner N, Gotthardt M. RBM20, a gene for hereditary cardiomyopathy, regulates titin splicing. Nat Med 2012;18:766–773. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B12] 12. Gregorich ZR, Zhang Y, Kamp TJ, Granzier HL, Guo W. Mechanisms of RBM20 cardiomyopathy: insights from model systems. Circ Genom Precis Med 2024;17:e004355. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B13] 13. Maatz H, Jens M, Liss M, Schafer S, Heinig M, Kirchner M, Adami E, Rintisch C, Dauksaite V, Radke MH, Selbach M, Barton PJR, Cook SA, Rajewsky N, Gotthardt M, Landthaler M, Hubner N. RNA-binding protein RBM20 represses splicing to orchestrate cardiac pre-mRNA processing. J Clin Invest 2014;124:3419–3430. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B14] 14. Dauksaite V, Gotthardt M. Molecular basis of titin exon exclusion by RBM20 and the novel titin splice regulator PTB4. Nucleic Acids Res 2018;46:5227–5238. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B15] 15. Katayama S, Tomaru Y, Kasukawa T, Waki K, Nakanishi M, Nakamura M, Nishida H, Yap CC, Suzuki M, Kawai J, Suzuki H, Carninci P, Hayashizaki Y, Wells C, Frith M, Ravasi T, Pang KC, Hallinan J, Mattick J, Hume DA, Lipovich L, Batalov S, Engstrom PG, Mizuno Y, Faghihi MA, Sandelin A, Chalk AM, Mottagui-Tabar S, Liang Z, Lenhard B, Wahlestedt C; RIKEN Genome Exploration Research Group; Genome Science Group (Genome Network Project Core Group); FANTOM Consortium . Antisense transcription in the mammalian transcriptome. Science 2005;309:1564–1566. [DOI] [PubMed] [Google Scholar]

[cvaf037-B16] 16. Morrissy AS, Griffith M, Marra MA. Extensive relationship between antisense transcription and alternative splicing in the human genome. Genome Res 2011;21:1203–1212. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B17] 17. Beltran M, Puig I, Pena C, Garcia JM, Alvarez AB, Pena R, Bonilla F, de Herreros AG. A natural antisense transcript regulates Zeb2/Sip1 gene expression during Snail1-induced epithelial–mesenchymal transition. Genes Dev 2008;22:756–769. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B18] 18. Xu S, Wang P, Zhang J, Wu H, Sui S, Zhang J, Wang Q, Qiao K, Yang W, Xu H, Pang D. Ai-lncRNA EGOT enhancing autophagy sensitizes paclitaxel cytotoxicity via upregulation of ITPR1 expression by RNA–RNA and RNA–protein interactions in human cancer. Mol Cancer 2019;18:89. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B19] 19. Zhao Y, Riching AS, Knight WE, Chi C, Broadwell LJ, Du Y, Abdel-Hafiz M, Ambardekar AV, Irwin DC, Proenza C, Xu H, Leinwand LA, Walker LA, Woulfe KC, Bristow MR, Buttrick PM, Song K. Cardiomyocyte-specific long noncoding RNA regulates alternative splicing of the triadin gene in the heart. Circulation 2022;146:699–714. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B20] 20. Werner MS, Ruthenburg AJ. Nuclear fractionation reveals thousands of chromatin-tethered noncoding RNAs adjacent to active genes. Cell Rep 2015;12:1089–1098. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B21] 21. Hall DD, Dai S, Tseng P-Y, Malik Z, Nguyen M, Matt L, Schnizler K, Shephard A, Mohapatra DP, Tsuruta F, Dolmetsch RE, Christel CJ, Lee A, Burette A, Weinberg RJ, Hell JW. Competition between alpha-actinin and Ca²⁺-calmodulin controls surface retention of the L-type Ca²⁺ channel Ca_V1.2. Neuron 2013;78:483–497. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B22] 22. Greaser ML, Warren CM. Method for resolution and western blotting of very large proteins using agarose electrophoresis. Methods Mol Biol 2015;1312:285–291. [DOI] [PubMed] [Google Scholar]

[cvaf037-B23] 23. Zhao B, Zhang K, Chen CS, Lejeune E. Sarc-Graph: automated segmentation, tracking, and analysis of sarcomeres in hiPSC-derived cardiomyocytes. PLoS Comput Biol 2021;17:e1009443. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B24] 24. Anders S, Reyes A, Huber W. Detecting differential usage of exons from RNA-seq data. Genome Res 2012;22:2008–2017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B25] 25. Reyes A, Anders S, Weatheritt RJ, Gibson TJ, Steinmetz LM, Huber W. Drift and conservation of differential exon usage across tissues in primate species. Proc Natl Acad Sci U S A 2013;110:15377–15382. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B26] 26. Zhang Y, Hardison RC. Accurate and reproducible functional maps in 127 human cell types via 2D genome segmentation. Nucleic Acids Res 2017;45:9823–9836. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B27] 27. Eraslan G, Drokhlyansky E, Anand S, Fiskin E, Subramanian A, Slyper M, Wang J, Van Wittenberghe N, Rouhana JM, Waldman J, Ashenberg O, Lek M, Dionne D, Win TS, Cuoco MS, Kuksenko O, Tsankov AM, Branton PA, Marshall JL, Greka A, Getz G, Segre AV, Aguet F, Rozenblatt-Rosen O, Ardlie KG, Regev A. Single-nucleus cross-tissue molecular reference maps toward understanding disease gene function. Science 2022;376:eabl4290. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B28] 28. Zhao S, Zhang X, Chen S, Zhang S. Natural antisense transcripts in the biological hallmarks of cancer: powerful regulators hidden in the dark. J Exp Clin Cancer Res 2020;39:187. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B29] 29. Pelechano V, Steinmetz LM. Gene regulation by antisense transcription. Nat Rev Genet 2013;14:880–893. [DOI] [PubMed] [Google Scholar]

[cvaf037-B30] 30. Hinson JT, Chopra A, Nafissi N, Polacheck WJ, Benson CC, Swist S, Gorham J, Yang L, Schafer S, Sheng CC, Haghighi A, Homsy J, Hubner N, Church G, Cook SA, Linke WA, Chen CS, Seidman JG, Seidman CE. HEART DISEASE. Titin mutations in iPS cells define sarcomere insufficiency as a cause of dilated cardiomyopathy. Science 2015;349:982–986. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B31] 31. Li S, Guo W, Dewey CN, Greaser ML. Rbm20 regulates titin alternative splicing as a splicing repressor. Nucleic Acids Res 2013;41:2659–2672. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B32] 32. Guo W, Bharmal SJ, Esbona K, Greaser ML. Titin diversity—alternative splicing gone wild. J Biomed Biotechnol 2010;2010:753675. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B33] 33. de la Mata M, Alonso CR, Kadener S, Fededa JP, Blaustein M, Pelisch F, Cramer P, Bentley D, Kornblihtt AR. A slow RNA polymerase II affects alternative splicing in vivo. Mol Cell 2003;12:525–532. [DOI] [PubMed] [Google Scholar]

[cvaf037-B34] 34. Bertero A, Fields PA, Ramani V, Bonora G, Yardimci GG, Reinecke H, Pabon L, Noble WS, Shendure J, Murry CE. Dynamics of genome reorganization during human cardiogenesis reveal an RBM20-dependent splicing factory. Nat Commun 2019;10:1538. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B35] 35. Tian C, Yang Y, Ke Y, Yang L, Zhong L, Wang Z, Huang H. Integrative analyses of genes associated with right ventricular cardiomyopathy induced by tricuspid regurgitation. Front Genet 2021;12:708275. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B36] 36. Zhang Y, Shang Z, Xu S, Zhou G, Liu A. ELF5-regulated lncRNA-TTN-AS1 alleviates myocardial cell injury via recruiting PCBP2 to increase CDK6 stability in myocardial infarction. Mol Cell Biol 2024;44:303–315. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B37] 37. Hinze F, Dieterich C, Radke MH, Granzier H, Gotthardt M. Reducing RBM20 activity improves diastolic dysfunction and cardiac atrophy. J Mol Med (Berl) 2016;94:1349–1358. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B38] 38. Methawasin M, Hutchinson KR, Lee E-J, Smith JE 3rd, Saripalli C, Hidalgo CG, Ottenheijm CAC, Granzier H. Experimentally increasing titin compliance in a novel mouse model attenuates the Frank–Starling mechanism but has a beneficial effect on diastole. Circulation 2014;129:1924–1936. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B39] 39. Bendixen L, Jensen TI, Bak RO. CRISPR-Cas-mediated transcriptional modulation: the therapeutic promises of CRISPRa and CRISPRi. Mol Ther 2023;31:1920–1937. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B40] 40. Wood EJ, Chin-Inmanu K, Jia H, Lipovich L. Sense–antisense gene pairs: sequence, transcription, and structure are not conserved between human and mouse. Front Genet 2013;4:183. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cvaf037-B41] 41. Karakikes I, Ameen M, Termglinchan V, Wu JC. Human induced pluripotent stem cell-derived cardiomyocytes: insights into molecular, cellular, and functional phenotypes. Circ Res 2015;117:80–88. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Antisense-mediated regulation of exon usage in the elastic spring region of Titin modulates sarcomere function

Selvi Celik

Ludvig Hyrefelt

Tomasz Czuba

Yuan Li

Juliana Assis

Julia Martinez

Markus Johansson

Oscar André

Jane Synnergren

Joakim Sandstedt

Pontus Nordenfelt

Kristina Vukusic

J Gustav Smith

Olof Gidlöf

Abstract

Aims

Methods and results

Conclusions

Graphical Abstract

Graphical Abstract.

1. Introduction

2. Methods

2.1. Human samples

Table 1.

2.2. Human heart muscle cells

2.3. RNA isolation

2.4. siRNA and plasmid DNA transfection

2.5. RNA in situ hybridization of human cardiac tissue

2.6. Combined RNA in situ hybridization and immunofluorescence

2.7. Protein gel electrophoresis

2.8. RNA immunoprecipitation

2.9. Chromatin immunoprecipitation

2.10. Sarcomere tracking

2.11. qPCR and RT–PCR

2.12. RNA-sequencing

2.13. Statistical analysis

3. Results

3.1. Cardiac antisense transcription in the TTN locus

Figure 1.

3.2. Design and validation of siRNAs targeting TTN-AS1-276

3.3. TTN-AS1-276 does not regulate TTN expression

3.4. TTN-AS1-276 regulates splicing of TTN I-band exons

Figure 2.

3.5. TTN-AS1-276 regulates TTN isoform composition

Figure 3.

3.6. TTN-AS1-276 regulates sarcomere function and cardiomyocyte contraction dynamics

Figure 4.

3.7. No impact of overlapping antisense transcript exons on TTN exon usage and Pol II elongation rate

3.8. TTN-AS1 facilitates interaction between RBM20 and TTN mRNA

Figure 5.

3.9. TTN-AS1 mediates splicing of other RBM20 targets

Figure 6.

3.10. TTN-AS1-276 is induced in human heart failure and correlates with TTN I-band exon usage

Figure 7.

4. Discussion

Supplementary Material

Acknowledgements

Contributor Information

Supplementary material

Authors’ contributions

Funding

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases