Novel and known ribosomal causes of Diamond-Blackfan anaemia identified through comprehensive genomic characterisation

Abstract

Background: Diamond-Blackfan anaemia (DBA) is an inherited bone marrow failure syndrome (IBMFS) characterised by erythroid hypoplasia. It is associated with congenital anomalies and a high risk of developing specific cancers. DBA is caused predominantly by autosomal dominant pathogenic variants in at least 15 genes affecting ribosomal biogenesis and function. Two X-linked recessive genes have been identified.

Objectives: We aim to identify the genetic aetiology of DBA.

Methods: Of 87 families with DBA enrolled in an institutional review board-approved cohort study (ClinicalTrials.gov Identifier:NCT00027274), 61 had genetic testing information available. Thirty-five families did not have a known genetic cause and thus underwent comprehensive genomic evaluation with whole exome sequencing, deletion and CNV analyses to identify their disease-associated pathogenic variant. Controls for functional studies were healthy mutation-negative individuals enrolled in the same study.

Results: Our analyses uncovered heterozygous pathogenic variants in two previously undescribed genes in two families. One family had a non-synonymous variant (p.K77N) in RPL35; the second family had a non-synonymous variant (p. L51S) in RPL18. Both of these variants result in pre-rRNA processing defects. We identified heterozygous pathogenic variants in previously known DBA genes in 16 of 35 families. Seventeen families who underwent genetic analyses are yet to have a genetic cause of disease identified.

Conclusions: Overall, heterozygous pathogenic variants in ribosomal genes were identified in 44 of the 61 families (72%). De novo pathogenic variants were observed in 57% of patients with DBA. Ongoing studies of DBA genomics will be important to understand this complex disorder.

Keywords: Diamond-Blackfan anemia; RPL18; RPL35; genetics; ribosome; whole exome sequencing.

Figure 1.

Flowchart showing genetic breakdown of all families enrolled in the NCI’s DBA cohort. WES: Whole exome sequencing; aCGH: Array comparative genomic hybridization; CNV: Copy number variation

Figure 2.

Family structure pedigree and corresponding mutation information of families with novel ribosomal genes causative of DBA. Figure 2A. Family NCI-138 with mutation in RPL35 Figure 2B. Family NCI-172 with mutation in RPL18 Figure 2C. Summary mutation information and in silico analyses of novel ribosomal genes in families NCI-138 and NCI-173

Figure 3.

Northern blot analysis of pre-rRNA processing was performed with hybridization probe sequences within ITS1 and ITS2 (shown as ** and * above the schematic) using mononuclear cells recovered from the peripheral blood from the affected child of the proband in each of the families NCI-138 and 172, and an unaffected control individual wildtype (Wt) expression of the novel ribosomal genes. Figure 3A. Pre-rRNA processing analysis showing a characteristic increase in 32S and 41S intermediates in family NCI-138 with mutation with RPL35 Figure 3B. Pre-rRNA processing analysis showing an increase in the 36S pre-rRNA intermediate in family NCI-172 with mutation in RPL18

Figure 4.

Copy number variation (CNV) analysis showing SNP array data depicting deletions and copy-neutral loss of heterozygosity surrounding deleted region. Each dot represents a SNP. Black dots represent the logR ratio (LRR, scale along Y axis). Red dots represent the B-allele frequency (scale to the right of the graph). The blue line denotes the region of the chromosome deleted. Figure 4A. CNV analysis of SNP array in Family NCI-51 showing deletion in chromosome 15q25 involving RPS17 and RPS17L Figure 4B. CNV analysis of SNP array in Family NCI-71 showing deletion in chromosome 3q29 involving RPL35A

Figure 5.

Frequencies of ribosomal protein mutations in the NCI Inherited Bone Marrow Failure Syndromes Diamond Blackfan anemia cohort. Figure 5A. Percentage of known genetic cause of disease in the cohort attributed to each ribosomal gene identified. Further breakdown of each column indicated the specific type of genetic change identified. Figure 5B. Number of families in the cohort with known inherited or proven de novo changes in each gene. ‘Unknown’ indicates that there was incomplete or no data on testing of the proband’s parents. Figure 5C. Breakdown by number of families and number of individuals identified with each causative ribosomal gene. More individuals than families indicate multi-case families with changes in that particular gene.