Whole-Exome Sequencing Insights into Adult Pulmonary Fibrosis. Repeating the Telomere Theme

Pulmonary fibrosis is currently the leading indication for lung transplantation in the United States (1). Despite recent Food and Drug Administration–approved medications that slow the course of idiopathic pulmonary fibrosis (IPF), fibrosis can be relentlessly progressive and fatal. Life expectancy ranges between 3 and 5 years (2, 3). Increased understanding of the basic mechanisms underlying the initiation and progression of lung fibrosis is needed to close therapeutic gaps.

Genome-wide association studies (GWAS) have reported a number of pulmonary fibrosis susceptibility variants (4). The most replicated of these, a single-nucleotide polymorphism in the MUC5B promoter, accounts for a proportion of the estimated heritability of pulmonary fibrosis (5). But given its high frequency in the general population, it cannot substantially inform an individual patient’s risk of disease. As with other complex diseases, rare variants of large effect size are generally not captured by GWAS but are identified using next-generation sequencing technologies, such as whole-exome sequencing.

In this issue of the Journal, Petrovski and colleagues (pp. 82–93) used whole-exome sequencing to study a cohort of patients who underwent lung transplantation for end-stage pulmonary fibrosis (6). The protein-coding sequences of more than 18,000 genes were captured and sequenced. Genetic data for 262 pulmonary fibrosis cases and 4,141 ethnically matched control subjects were pruned so that there was a similar opportunity to identify genetic variants in both groups. The authors used prespecified measures of variant allele frequency and in silico–predicted variant effect to define different categories of “qualifying variants.” The number of qualifying variants found per gene were counted and compared between cases and control subjects. Those genes that surpassed a study-wide significance threshold—accounting for the number of analyzed genes and six models of different variant classes—were reported to be enriched in this cohort of patients with predominantly sporadic IPF.

The authors demonstrate multiple ultrarare variants in genes that converge on the telomerase pathway (Figure 1). Although most of the patients (87%) did not report a family history of pulmonary fibrosis, both the genes and the types of variants discovered recapitulate those found in family studies (7–11). Qualifying variants in the TERT, RTEL1, and PARN genes were found to be significantly enriched in cases compared with control subjects. Altogether, 11.5% of pulmonary fibrosis cases (31 of the 262 cases) were found to carry a qualifying variant in one of these genes. Subgroup analysis of IPF cases replicated these findings. All three genes are linked by their shared role in maintaining telomere length. RTEL1 is a helicase that unwinds the specialized structures at the ends of telomeres, allowing the enzyme telomerase (TERT) access to elongate the distal telomere ends. PARN is necessary for the maturation of the telomerase RNA (hTR, encoded by the TERC gene) (12), which serves as the template for the TTAGGG hexanucleotide telomere repeat. Although the study did not identify any new genes, it did highlight the role of variants in these three genes as significantly contributing to the overall genetic architecture of sporadic pulmonary fibrosis and, specifically, IPF.

Figure 1.

Ultrarare variants identified by whole-exome sequencing of patients with sporadic pulmonary fibrosis converge on three genes (TERT, RTEL1, and PARN) in the telomerase pathway. Human telomeres, capping the ends of chromosomes, consist of several thousand bases of tandemly repeated 5′-TTAGGG-3′ sequences. Telomeres terminate with a T-loop structure and can fold into G-quadruplex structures. The regulator of telomere length 1 (RTEL1) is a DNA helicase that disassembles a variety of DNA secondary structures, including G-quadruplex structures and T-loops, to maintain telomere integrity. Telomerase, a ribonucleoprotein that contains both an intrinsic RNA (hTR, TERC) and a protein catalytic subunit (TERT), adds single-stranded TTAGGG repeats to the end of the chromosome and leads to telomere elongation. Polyadenylation-specific RNase (PARN) removes oligo(A) tails from the 3′ end of TERC, which serves as the template for the telomere repeat. Ultrarare qualifying variants reported in this issue of the Journal (6) are predicted to be damaging and are likely associated with short telomeres.

This study extends our understanding of the genetic risk for pulmonary fibrosis in a number of important ways. It demonstrates the significant enrichment of two types of genetic variants: (1) ultrarare variants with an allele frequency less than 0.1% that lead to premature truncation or extensive change in protein sequence, and (2) variants that are so rare that they are not found in large reference databases and whose predicted substitution of a single amino acid likely leads to a damaging effect. These two categories of variants had calculated mean odds ratios of 30 to 250, which is much greater than the odds ratios of common variants discovered by GWAS. All discovered variants are heterozygous. Because they are predicted to have a deleterious effect on protein function, these variants are likely associated with short telomeres. Although most other studies have identified variants that have been private to the individual and/or family in which they were found, this study demonstrated the presence of recurrent ultrarare variants in multiple unrelated individuals, echoing findings of recurrent TERT variants in a nationwide French cohort (13). Finally, the authors find evidence for an additive contribution of the common variant MUC5B promoter single-nucleotide polymorphism (rs35705950) in patients with rare variants in TERT, RTEL1, or PARN, thus increasing our understanding of the relationship between common and rare genetic variants in patients with pulmonary fibrosis.

The study is limited by a lack of functional analysis of the variants. There was no analysis of TERT, PARN, or RTEL1 protein function. In addition, there was no measurement of telomere lengths. Despite computational predictions that support a deleterious effect, genetic counseling for many of the reported variants would be difficult, and likely ambiguous, without additional data. In fact, some of the discovered rare missense variants that exceeded prespecified allele frequencies and were excluded as qualifying variants may be determined to be risk alleles after functional analyses are completed. This study was limited to individuals of European ancestry and could not be translated to other ethnic backgrounds without analysis of appropriate matched control subjects. Gaps in the sequencing, accounting for 8% of the coding portion of the genome and including the RNA component of telomerase (hTR, TERC), may have masked the identification of other risk genes.

Overall, the work provides insight into the genetic architecture of sporadic IPF and is consistent with several prior studies. Telomere shortening has been previously found to be a risk factor for the development of IPF (14, 15) and for predicting survival of patients with IPF (16, 17). Studies of some murine models of telomere shortening demonstrate the particular importance of the type II alveolar epithelial cell in the development of lung fibrosis (18, 19). This noteworthy study repeats the theme of telomerase dysfunction in IPF and puts the telomere end at the center of disease pathogenesis.