Follow the LINE: A novel case of dilated cardiomyopathy caused by a LINE-1 insertion in the TTN gene

Abstract

Introduction

Dilated cardiomyopathy (DCM) is a substantial cardiac pathology characterized by nonischemic left ventricular or biventricular dilatation and systolic dysfunction in the absence of congenital or valvular heart disease and hypertension. Dilated cardiomyopathy is the second-most common cause of heart failure, affecting approximately 1 in 250 individuals.^1,2 Prognosis of DCM, especially for patients with severe systolic dysfunction, is poor.³ Dilated cardiomyopathy is responsible for 10,000 deaths in the United States and more than 400,000 deaths worldwide annually.³

Genetic factors are important in DCM, with studies suggesting a genetic basis in 20% to 50% of cases.^4,5 Genetic forms of DCM are most frequently autosomal dominant and often affect sarcomere function.^2,5 Pathogenic variants—specifically, truncating variants—in TTN are the most common genetic cause of DCM.²

TTN encodes titin, the largest human protein. Titin contains 4 microscopically visible regions: the Z disk, I band, A band, and M band. Titin is essential for sarcomere organization and striated muscle function.⁶ A landmark study by Herman et al uncovered a strong association between A band TTN truncating variants (TTNtvs) and DCM.⁷ Further studies revealed that TTNtvs in constitutively expressed non–A band exons are also strongly associated with DCM.⁸ The onset of TTN-associated DCM is usually between 40 and 59 years of age.²

Most TTNtvs described to date are single-nucleotide variations (formerly, single-nucleotide polymorphisms) or small insertions and deletions (indels).⁹ Structural variants (eg, mobile element insertions [MEIs]), however, could also cause protein truncation. Mobile element insertions involve inserting a mobile genetic element (eg, Alu, long interspersed nuclear element [LINE]) into a new genomic location.¹⁰ Mobile element insertions have long been established as a cause of various human diseases,^11-13 but they have not yet been identified as a mechanism for TTN-associated DCM.

Mobile element insertions pose substantial analytical and interpretive challenges for clinical laboratorians. Although detectable by next-generation sequencing (NGS), raw data (NGS reads) that support an MEI may be mistaken for noise during manual review. Orthogonal confirmation and clinical interpretation of MEIs are also challenging. Collectively, these factors may cause pathogenic MEIs to be overlooked in clinical laboratories.

Here we describe a 17-year-old patient with DCM in whom a likely pathogenic LINE-1 insertion in the A band of TTN was identified. The LINE-1 insertion was initially detected by NGS and orthogonally confirmed by long-range polymerase chain reaction (PCR). This article marks the first report of an MEI as the cause of TTN-associated DCM. Our report extends the mutational spectrum of TTNtv in DCM and has important educational value for clinical laboratorians, geneticists, and cardiologists.

Methods

A 17-year-old male patient was transferred to the pediatric intensive care unit for congestive heart failure and acute kidney injury secondary to newly diagnosed DCM. Transthoracic echocardiography revealed severe biventricular dilatation and systolic dysfunction (left ventricular ejection fraction, 21%), severe biatrial enlargement, mild to moderate tricuspid and mitral valve regurgitation, and left ventricular hypertrabeculation. Follow-up echocardiograms 2 and 4 days later showed slight improvements in systolic dysfunction (left ventricular ejection fraction, 34% and 28%, respectively), but severe left ventricular dilatation persisted. The DCM and left ventricular noncompaction cardiomyopathy (LVNC) gene panel was ordered. Subsequently, the patient was transferred to another facility and received heart transplantation.

Mayo Clinic Laboratories’ DCM/LVNC gene panel analyzes 63 genes associated with hereditary DCM and LVNC, primarily using NGS and supplemented by Sanger sequencing. This panel contains the TTN gene and analyzes genetic variations within or adjacent to (within 10 base pairs [bp] of) exons. The full list of genes is available at https://www.mayocliniclabs.com/-/media/it-mmfiles/Special-Instructions/A/7/F/Targeted-Genes-Dilated-Cardiomyopathy-Left-Ventricular-Noncompaction.

Briefly, a peripheral blood specimen was received from the patient from which DNA was extracted using the chemagic 360 instrument (Revvity). Next-generation sequencing library preparation was carried out using the KAPA HyperPrep Kit (Roche), with co-capture using xGen Exome probes (Integrated DNA Technologies). Sequencing was performed on a NovaSeq 6000 Sequencing System (Illumina). The NGS reads were aligned to the hs37ds human reference genome (GRCh37/hg19 build). All quality metrics of the patient sample (eg, percentage of reads mapped, sequencing depth) were within normal limits. Thus, the NGS data from the patient proceeded to secondary and tertiary analyses. Manta was used to detect structural variants.¹⁴ The Integrative Genomics Viewer¹⁵ was used for manual NGS data review when needed.

Results

We first evaluated whether the patient had any pathogenic or likely pathogenic single-nucleotide variations, small indels, or copy number variants; none were detected in any of the 63 analyzed genes. There were 2 heterozygous variants of uncertain significance found in TTN (NM_001256850.1) (1) c.59170+4_59170+7del and (2) c.86951G>A, p.(Gly28984Glu). Nonetheless, they were not considered explanatory for the patient’s DCM.

Intriguingly, Manta detected a putative structural variant (SV) involving the TTN gene—specifically, a heterozygous insertion in exon 276 (NM_001256850.1) at genomic position 2:179,425,653 (hg19). Manta provided no additional details, such as the size or type of the insertion. Thus, we manually reviewed the NGS alignment data. As shown in FIGURE 1A, 3 distinctive patterns were observed near the putative SV site: (1) a slight but discernible increase in read depth over an 18-bp region, (2) right-clipped reads containing a poly-T sequence, and (3) left-clipped reads containing a sequence that could be mapped to multiple locations in the human genome using the BLAST-like Alignment Tool (BLAT, https://genome.ucsc.edu/cgi-bin/hgBlat) FIGURE 1B.

FIGURE 1

Next-generation sequencing alignment data supporting the LINE-1 insertion. A, IGV screenshot of the region in which Manta identified a putative insertion. Three characteristic features of MEIs were observed: (1) a slight increase in read depth over an 18-bp region that corresponds to the target site duplication sequence, (2) right-clipped reads containing a poly-T sequence (red arrows), and (3) left-clipped reads containing a nonuniquely mappable sequence (blue arrows). B, BLAT analysis revealed that the left-clipped sequence maps perfectly (ie, 100% identity) to multiple locations in the reference genome. BLAT, BLAST-like Alignment Tool; bp, base pairs; IGV, Integrative Genomics Viewer; LINE, long interspersed nuclear element; MEI, mobile element insertion.

Open in new tabDownload slide

The observed patterns were characteristic of an MEI. As illustrated in FIGURE 2A, an MEI typically consists of the sequence of the mobile element followed by a poly-A tail and is flanked by a short target site duplication sequence on both ends.¹⁰ The 18-bp region with increased read depth corresponds to the target site duplication, while the 2 types of clipped reads represent the 5ʹ end of the mobile element and the poly-A tail, respectively. Using BLAT, we identified the inserted mobile element as LINE-1.

FIGURE 2

Orthogonal confirmation and clinical interpretation of the LINE-1 insertion. A, Structure of LINE-1 insertions. B, Schematic of the LINE-1 insertion identified in this study. Red diamonds: locations of the PCR primers. Red lines: PCR products (with size indicated above or below). Regions in panels A and B are not to scale. C, Gel electrophoresis (following long-range PCR) confirms the presence of an approximately 4- to 5-kilobase pair insertion in the patient but not in the normal control sample. Both the patient and the control sample had a 431-bp PCR product that corresponds to the wild-type allele in panel B. D, The LINE-1 insertion sequence (magenta) disrupted the reading frame of TTN, leading to a premature stop codon. bp, base pair; LINE, long interspersed nuclear element; NTC, no template control; PCR, polymerase chain reaction; TSD, target site duplication.

Open in new tabDownload slide

To orthogonally confirm the LINE-1 insertion, we performed long-range PCR using a primer set that flanks the MEI FIGURE 2B. The PCR product sizes were estimated using gel electrophoresis. In the normal control, a single PCR product of 431 bp was observed. In the patient sample, 2 PCR products were observed: 1 of 431 bp and the other approximately 5 kilobase pairs (kb) FIGURE 2C. The 5-kb PCR product confirmed the presence of a heterozygous MEI of approximately 4 to 5 kb in the patient.

Based on the NGS data, we concluded that the LINE-1 element had inserted into exon 276 of TTN, which encodes part of the A band of titin. This insertion was predicted to disrupt the reading frame, resulting in premature protein truncation FIGURE 2D. Given that TTNtv is a known cause of DCM, we classified this variant as likely pathogenic and considered it likely explanatory for the patient’s DCM.

Discussion

Here we present the first documented case of an MEI causing TTN-associated DCM. Although TTNtvs are an established cause of DCM, our case extends the spectrum of genetic alterations that cause titin truncation. Overall, MEIs are rare,¹⁶ but they are increasingly recognized as contributors to human diseases.¹⁷ Our case underscores the value of reanalyzing genetic sequencing data for rarer types of pathogenic variants (eg, MEIs) and other SVs. This finding is especially pertinent when the initial result is negative but family history or clinical presentation strongly suggests a genetic cause.

Our case also offers important learning points for clinical laboratorians. The false-positive rate of SV detection is relatively high; therefore, manual review of NGS alignment data is usually required for SV identification in clinical laboratories. Because of the rarity of MEIs, laboratorians may not be familiar with their hallmark features, which may result in these variants being overlooked or misinterpreted. For example, the poly-T sequence and the left-clipped sequence that could not be uniquely mapped may be initially confused with noise or sequencing error. As detailed in this report, however, these are characteristic features of MEIs that warrant careful consideration during test interpretation.

Orthogonal confirmation, nomenclature, and clinical interpretation of MEIs also pose challenges. Polymerase chain reaction–based strategies that amplify across 1 or both breakpoints of the MEI FIGURE 2B, followed by fragment size analysis, are generally sufficient for orthogonal confirmation. LINE-1 elements are several kilobases in size and not uniquely mappable; thus, short-read NGS cannot sequence through the entire insertion. For this reason, we estimated the MEI size by gel electrophoresis and used the Human Genome Variation Society uncertainty nomenclature to describe this MEI at the complementary DNA level: TTN (NM_001256850.1): c.80282_80283insN[(4000_5000)]. Because the left-clipped sequence enabled exact identification of the new stop codon FIGURE 2D, however, the MEI was precisely described at the protein level as p.(Arg26762Glyfs*23). Given that this truncating variant is in the A band of TTN, it was interpreted as likely pathogenic for DCM.

This study has several limitations. First, we were unable to obtain parental samples to clarify the inheritance patterns of the variants, which limited our ability to definitively classify them. Second, although the LINE-1 insertion was predicted to result in frameshift, RNA analyses were not performed to confirm this effect because of the lack of TTN expression in clinically obtainable tissues (eg, peripheral blood and skin biopsies). Thus, we could not fully rule out that the LINE-1 insertion may have other effects (eg, activation of cryptic splice sites).

In conclusion, this case highlights the need to consider rarer types of pathogenic variants (eg, MEIs) in clinical practice. Detecting, confirming, and interpreting MEIs require expertise because their distinguishing features may easily be overlooked. We are hopeful that our report aids clinicians, laboratorians, and researchers in more readily recognizing MEIs and ultimately enhances genetic diagnostics.

Acknowledgements

The authors thank the staff of the Molecular Technologies Laboratory and the Clinical Genome Sequencing Laboratory at Mayo Clinic for generating the data used in this report.

Conflict of Interest Disclosure

The authors declare no conflicts of interests.

Funding

No funding was received for this work.

Ethics statements

The Mayo Clinic Institutional Review Board determined that approval was not required for case reports. The mother of the patient consented to the publication of this report.

Data Availability

All data generated during this study are included in this published article.

References

Author notes