DNAJC21: the new kid on the SDS block

In this issue of Blood, Dhanraj et al identify biallelic mutations in DNAJC21 in patients with phenotypical Shwachman-Diamond syndrome (SDS) who tested negative for mutations in the SBDS gene, thus extending the genes that can be mutated in SDS.¹ 

SDS was described originally in 1964²  and classically presents with short stature, exocrine pancreatic insufficiency, cognitive and behavioral impairment, and bone marrow failure, with predisposition to myeloid malignancies (Online Mendelian Inheritance of Man [OMIM] 260400).^3-5  SDS is an autosomal recessive disorder. The majority of patients who present with phenotypic SDS are found to have biallelic mutations in the SBDS gene, which encodes a protein with multiple functions, including the maturation of the 80S ribosome by the release of eIF6.⁴ 

Two recent articles now describe expansion of the molecular pathology in SDS to include biallelic mutations in DNAJC21, collectively observed in 8 individuals from 7 families.⁶  These patients showed classic features of bone marrow failure with peripheral pancytopenia. Tummala et al described additional features in their patients, which included intrauterine growth restriction, short stature, and a myeloid malignancy in a 12-year-old child.⁶  In the paper published in this issue of Blood by Dhanraj et al, the patients underwent detailed phenotypic characterization including analysis of pancreatic function to confirm the SDS phenotype.

The mutations identified in these patients encode mutations in the N-terminal half of the DNAJC21 protein (see figure). Missense mutations are found in the extreme N-terminus: P32A and K34E, which are likely to disrupt the function of the highly conserved H-P-D motif that is critical for function of the J domain. The 5 other mutations, if expressed, would be expected to truncate the protein before the 2 predicted C-terminal zinc finger domains: R173X, Q174X, E265X, and 2 splicing defects that result in frameshift mutations, V148Kfs*30 and G299Afs*2. The 3 nonsense mutations may be subject as well to nonsense-mediated decay of the corresponding messenger RNAs, although Dhanraj et al provide evidence for expression for at least 1 of the truncating frameshift mutations, V148Kfs*30.

The DNAJC21 protein associates with ribosomal RNA and contributes to 60S ribosomal subunit maturation. Tummala et al showed that mutant proteins have decreased interaction with cofactors HSPA8 and ZNF622 as well as cytoplasmic accumulation of the export factor PA2G4, resulting in late cytoplasmic maturation of 60S subunit, aberrant ribosome profiles, and increased cell death.⁶ 

These 2 papers are emblematic of a new trend in molecular diagnostics: the adaptation of next-generation sequencing techniques to determine the molecular basis of disease based on rare patients who phenocopy a well-described syndrome. With 10% to 20% of patients diagnosed with a clinical syndrome of SDS testing negative for SBDS, it is unclear what fraction of those cases has DNAJC21 mutations, leaving open the exciting possibility that there are other causative gene mutations for SDS. Moreover, in the past, when clinical criteria were used to define a syndrome, it is likely that the most apparent phenotypes that were described corresponded to the strongest disrupting alleles. Now that next-generation sequencing and molecular testing of individuals are more accessible and affordable, we are likely to expand our clinical definitions of syndromes, including the ribosomopathies.⁷