The pathophysiology of bone marrow failure (BMF) can be immune, as in acquired aplastic anemia (AA), or constitutional, due to germline mutations in genes critical for DNA repair and telomere maintenance. Variability in penetrance and phenotype can complicate diagnosis, as patients with underlying genetic defects may present in adulthood and without characteristic physical anomalies. RTEL1 encodes a helicase crucial for telomere maintenance and DNA repair. The gene has two main transcripts in human cells: the 1300 amino acid isoform 3 and the 1219 amino acid isoform 1. RTEL1 isoform 3 contains a conserved C4C4-RING domain responsible for resolving the t-loop required for telomere replication. Using next-generation sequencing (NGS), RTEL1 germline variants with unknown clinical significance have been found in AA patients. Functional tests may elucidate RTEL1 variants' pathogenic role in telomere biology. Here, we describe RTEL1 heterozygous germline mutations in patients with BMF and investigate their impact in telomere maintenance.

We screened 63 patients with a suggestive familial phenotype for germline mutations in peripheral blood cells using a targeted, 49 gene NGS panel. To investigate variants' impact in telomere functions, telomere length (TL) was measured by Southern blot (SB), t-circles were quantified by telomere circle assay, and single-stranded overhang was measured by non-denaturing SB.

Eight patients carried novel heterozygous non-synonymous RTEL1 variants: four nucleotide changes were located in the RAD3 domain, six in the harmonin-like domain, and one in the RING domain. Clinical features and TL were heterogeneous (Table 1). The only RTEL1 variant predicted as pathogenic in silico was F1262L (c.3786 C>G) in patient 2; this mutation affects a highly conserved amino acid residue located in the RING domain, which is responsible for RTEL1 interaction with TRF2 at telomeres and t-loop unwinding. Patient 2 had very short telomeres, abnormal accumulation of t-circles, and eroded single-stranded telomeric overhangs in leukocytes, indicating a disrupted RTEL1 RING domain. To confirm observations made in clinical samples, 293T cells transfected with a plasmid carrying wild-type RTEL1-FLAG isoform 3 or its F1262L mutated version were assessed for TRF2 and FLAG co-localization in the nucleus. By confocal microscopy, wild-type RTEL1, but not mutant RTEL1 co-localized with TRF2. These findings strongly implicate RTEL1-F1262L as pathogenic, and thus the first autosomal dominant mutation in the RING domain in an AA patient. In patient 1, D743N variant in silico prediction was indeterminate, but telomeres were very short and there was a family history of typical telomeropathy (AA, liver cirrhosis, and pulmonary fibrosis) without any other suspicious germline mutations. The D743N variant is located close to the V745M variant that has been reported in a patient with dyskeratosis congenita. Increased amounts of t-circles and telomeric overhang attrition were observed in three other patients (#4, 5, and 7). While not specific for RTEL1 function, these results suggest telomere dysfunction, despite TLs in the normal range for patient 4 and 5. The RTEL1 P82L variant also appeared related to clonal evolution and leukemic progression observed in patient 5. For patients 3, 4, 6, 7, and 8, several mutations were observed in other genes concomitant to RTEL1, and a more complex genomic architecture may be the cause of patients' phenotype. A previously reported TERC variant, and a TERT variant of undetermined in silico prediction, could be pathogenic in patients 7 and 6, respectively. In these cases, RTEL1 variants may modulate disease, or represent only coincidental abnormalities.

To our knowledge, this is the first report of heterozygous RTEL1 mutations in AA. We also describe a TL-independent association between RTEL1 haploinsufficiency and telomere dysfunction in humans. Haploinsufficiency of RTEL1 may disrupt DNA repair, destabilize the genome, and promote leukemogenesis by a mechanism different than typical accelerated telomere attrition associated with very short telomeres. T-circle quantification and overhang measurement may be better measures of telomere dysfunction in patients with RTEL1 variants than simple TL assessment. The combination of different functional tests was useful to the assessment of novel variants impact in telomere maintenance and DNA repair.

Disclosures

Fernandez Ibanez:GSK/Novartis: Research Funding. Desierto:GSK/Novartis: Research Funding. Townsley:GSK/Novartis: Research Funding. Young:GSK/Novartis: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution