AUTHORS
Martina Sarchi, MD, PhD, Francisco Saavedra Cantillana2, Carmen Iannini3, Connor Mahoney4, Laura Daniela Baquero Galvis5, Alessandro Enrico6, Sara Gonske7, Yu-Yen Hsiao8, Moises Di Sante9, Anna Gallì, Bsc10, Martina Gandossini11, Francesco Pasqualini12, Luca Malcovati, MD13, Thelma M Escobar14, Sergei Doulatov, PhD15
AFFILIATIONS
1 IRCCS San Matteo Hospital Foundation, Pavia, Italy, University of Pavia, Pavia, Italy
2 University of Washington, Seattle, Washington, United States
3 University of Pavia, Pavia, Italy
4 University of Washington, Seattle, Washington, United States
5 Columbia University Medical Center, New York, New York, United States, University of Washington, Seattle, Washington, United States
6 University of Pavia, Pavia, Italy
7 University of Washington, Seattle, Washington, United States
8 University of Washington, Seattle, Washington, United States
9 University of Pavia, Pavia, Italy
10 IRCCS San Matteo Hospital Foundation, Pavia, Italy
11 University of Pavia, Pavia, Italy
12 University of Pavia, Pavia, Italy
13 IRCCS San Matteo Hospital Foundation, Pavia, Italy, University of Pavia, Pavia, Italy
14 University of Washington, Seattle, Washington, United States
15 Columbia University Medical Center, New York, New York, United States
ABSTRACT
DEAD-box (DDX) RNA helicases regulate RNA metabolism, translation, and genome stability. Mutations in DDX helicases, notably DDX41, cause inherited and somatic disorders, but the broader role of the DDX family in normal and neoplastic hematopoiesis remains poorly characterized. Here, we identify DDX54 as the second most frequently mutated DDX gene in myelodysplastic neoplasms (MDS), with frequencies between 1 and 10% across independent cohorts. DDX54 mutations are exclusively missense and include both germline and somatic variants. DDX54 has a known role in ribosomal maturation; however, its function in the hematopoietic system is unknown.
To define DDX54 role in hematopoiesis, we first performed CRISPR/Cas9-mediated DDX54 knock-out (KO) in primary human CD34+ hematopoietic stem and progenitor cells (HSPCs). DDX54-KO impaired HSPC expansion and induced erythroid skewing. Transcriptomic profiling of DDX54-KO HSPCs revealed down-regulation of translation and ribosome biogenesis programs. Consistent with its role in ribosomal maturation, DDX54-KO impaired protein synthesis, which was restored upon DDX54 re-expression. Mass spectrometry (MS) of the DDX54 interactome identified proteins involved in ribosome biogenesis, rRNA processing and splicing. Notably, NPM1, a nucleolar protein required for ribosome maturation and recurrently mutated in myeloid neoplasms, emerged as a top interactor. We confirmed direct binding using purified recombinant proteins, and established that DDX54 C-terminal alpha helix, the most common target of MDS mutations, is required for NPM1 interaction. These findings establish DDX54 as a critical regulator of hematopoiesis and identify NPM1 as its direct binding partner.
To investigate MDS-associated DDX54 mutations, we first overexpressed mutant isoforms in DDX54-KO cells. Mutant DDX54 rescued cell growth and protein synthesis, indicating that these mutations are likely gain or change of function. Since DDX54 C-terminus is recurrently mutated in MDS and critical for NPM1 binding, we engineered a precise C-terminal R835H mutation, or a silent R835R control, in CD34+ HSPCs by CRISPR/Cas9 homology directed repair. By contrast with DDX54-KO, R835H mutation promoted HSPC expansion and impaired erythropoiesis, phenotypic hallmarks of MDS. HSC differentiation and lineage commitment genes were downregulated, while ribosome biogenesis and oxidative phosphorylation programs were upregulated. Consistently, protein synthesis was modestly increased in HSPCs, but decreased in erythroid precursors. These data demonstrate that DDX54 R835H mutation promotes HSPC expansion and impairs erythroid differentiation through cell type-specific regulation of protein synthesis.
To define the underlying mechanisms, we compared DDX54 R835H protein interactome to wild-type, showing reduced binding to ribonucleoprotein biogenesis factors, critically NPM1. Co-IP confirmed decreased NPM1 binding. Strikingly, DDX54:NPM1 crosslinking MS revealed that DDX54 mutations in MDS map to NPM1 interacting sites, supporting the idea that the DDX54:NPM1 interaction is essential for mutant DDX54 pathogenicity. To dissect this interaction in a cellular context, we found that DDX54 primarily localizes in the inner core of the nucleolus, while NPM1 is a key component of the outer granular zone. We hypothesized that DDX54 governs NPM1 distribution in the granular zone. DDX54-KO increased NPM1 condensate area, while overexpression had the opposite effect, with more modest impact on the fibrillar core. This was associated with NPM1 redistribution in the nucleolus and increased ribosomal proteins and 28S rRNA in the cytosol of R835H cells. These data show that DDX54 regulates nucleolar structure and its mutations promote NPM1 redistribution, altering rRNA maturation and export.
In conclusion, our study identifies DDX54 as a novel genetic driver in MDS, uncovering its physiological role in hematopoiesis and demonstrating that DDX54 mutations induce MDS-like phenotypes in human HSPCs. Mechanistically, DDX54 regulates protein synthesis and nucleolar structure via direct interaction with NPM1. DDX54 mutations in MDS occur at NPM1 binding sites, frequently at the C-terminus. C-terminal R835H mutation disrupts DDX54:NPM1 interaction, remodeling the nucleolar granular zone and increasing ribosomal components in the cytosol. These findings expand the pathogenic landscape of DDX helicases, highlighting the broader DEAD-box family as a critical axis in MDS.
DISCLOSURES
Martina Sarchi, MD, PhD: No relevant conflicts of interest to declare.