The Genetic Landscape of Diamond-Blackfan Anemia

Abstract

Diamond-Blackfan anemia (DBA) is a rare bone marrow failure disorder that affects 7 out of 1,000,000 live births and has been associated with mutations in components of the ribosome. In order to characterize the genetic landscape of this heterogeneous disorder, we recruited a cohort of 472 individuals with a clinical diagnosis of DBA and performed whole-exome sequencing (WES). We identified relevant rare and predicted damaging mutations for 78% of individuals. The majority of mutations were singletons, absent from population databases, predicted to cause loss of function, and located in 1 of 19 previously reported ribosomal protein (RP)-encoding genes. Using exon coverage estimates, we identified and validated 31 deletions in RP genes. We also observed an enrichment for extended splice site mutations and validated their diverse effects using RNA sequencing in cell lines obtained from individuals with DBA. Leveraging the size of our cohort, we observed robust genotype-phenotype associations with congenital abnormalities and treatment outcomes. We further identified rare mutations in seven previously unreported RP genes that may cause DBA, as well as several distinct disorders that appear to phenocopy DBA, including nine individuals with biallelic CECR1 mutations that result in deficiency of ADA2. However, no new genes were identified at exome-wide significance, suggesting that there are no unidentified genes containing mutations readily identified by WES that explain >5% of DBA-affected case subjects. Overall, this report should inform not only clinical practice for DBA-affected individuals, but also the design and analysis of rare variant studies for heterogeneous Mendelian disorders.

Keywords: Diamond-Blackfan anemia; RNA sequencing; congenital hypoplastic anemia; hematopoiesis; human genetics; rare disease; whole-exome sequencing.

Figure 1

Mutational Spectrum of Likely Pathogenic Variants in DBA (A) PCA of genetic ancestry based upon 1000 Genomes for the DBA cohort. Filled circles represent individual DBA-affected families and open circles represent 1000 Genomes individuals. (B) Percentage of putative causal mutations in each gene. A total of 78% of case subjects have a putative causal mutation. (C) Types of putative causal mutations. Across all affected individuals, 176 had a LoF, 73 had a missense, 68 had a splice related mutation, 31 had a deletion, 13 had multiple putative causal variants, and 7 had variants that fell into other categories. (D) Relative frequency of putative causal mutations in the DBA cohort. Among unrelated individuals, 169 variants were singletons, 23 variants were doubletons, and 19 variants were seen 3 or more times. (E) Structure of the assembled ribosome highlighting the 19 known and 7 new RP genes that are mutated in the DBA cohort. The color coding reflects the frequency of mutations: blue (none found in this study, but reported in other studies), cyan (1 mutation), green (2–3 mutations), yellow (7–15 mutations), orange (31–54 mutations), and red (more than 100 mutations). RPS7, RPL27, RPL34, and RPL19 are mutated in DBA but are obscured by other proteins from this viewpoint.

Figure 2

Non-canonical Splice Variants in Known DBA-Associated Genes (A and B) Sashimi plots of non-canonical splice mutants and a representative control are shown. The number of reads spanning each junction is indicated by the size of the sashimi plot curve, and the number of reads spanning mutant junctions is indicated in red text. Coverage plots for total mapped reads are shown (total exon coverages is shown on the y axis). Mutant junctions due to the mutation indicated in (C) are highlighted in red. For the last panel in (B), the mutation disrupts a poly(A) binding site resulting in an extended 3′ UTR. The best guess mutant poly(A) site used by this extended transcript is indicated. (C) Location and consequence of each mutation is shown, in addition to coverage plots for the exon extension mutants.

Figure 3

Copy Number Variants Identified by WES Copy number variants identified by WES sequence are shown for (A) RPS24, (B) RPS19, (C) RPL35A, and (D) RPS17. A total of 558 controls consisting of other DBA samples or samples sequenced at the same time are shown in gray. (A) In this case, unaffected parents underwent WES and the inheritance of the deletion was determined to be de novo. (B) Several different deletions affecting only 5′ exons, 3′ exons, or the entire gene were detected across RPS19. (D) RPS17 deletions were determined to almost exclusively be due to microdeletions of this region. A total of 11 individuals with RPS17 deletions were detected.

Figure 4

Penetrance and Prevalence of RP Genes (A) Near complete penetrance for LoF mutations in the top three most frequently mutated genes (57% of case subjects) was observed. Slightly lower estimates were obtained for LoF mutations in other known DBA-mutated RP genes. Penetrance was much lower for rare RPS19 missense mutations (58% of all missense) but substantially increased when considering only predicted damaging mutations. (B) The majority of missense mutations identified in the DBA cohort are predicted to be damaging, whereas mutations of similar frequency in gnomAD are predicted to be benign. (C) RPS19 missense mutations cluster into 3–4 groups along the mRNA transcript, although without clear separation from gnomAD mutations. (D) RPS19 missense mutations appear to predominantly disrupt the stability of RPS19 by altering the hydrophobic core or by disrupting interactions with rRNA in the assembled ribosome (Table S8). The core α helices (1, 2, 4, and 5) and the β-hairpin are labeled. (E) Altering hydrophobic (Hydro.) amino acids to another type of amino acid was common in DBA but not in gnomAD. Special amino acids include glycine, proline, and cysteine.

Figure 5

Phenotypic Associations Differences in (A) presence of ≥1 congenital abnormality, (B) heart malformations, (C) treatment requirements, and (D) eADA levels were observed between different RP genes. Differences remained significant after removing missense mutations. A Fisher’s exact test was used to test the hypothesis that there were differences in proportion of the outcome between RP genes. (D) The association between RP gene and treatment requirements appeared to primarily be due to differences in remission, and this association was no longer significant after removing individuals who went into remission.

Figure 6

Gene Burden Results (A) The burden of rare synonymous mutations in DBA indicate limited deviance from expected. (B) We observed an exome-wide significant association between rare LoF dominant mutations in RPS19, RPL5, RPS26, RPL11, and RPS10, five of the most commonly mutated DBA-associated genes. (C) Similar results were observed after including rare damaging missense. (D) We observed an exome-wide association between rare LoF and damaging missense mutations with recessive inheritance for CECR1 (ADA2). A Fisher’s exact test was used to test for differences in each class of mutation between the DBA cohort and the gnomAD population control dataset, after filtering for high confidence variants and well covered regions consistent in both variant call sets.