Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: Pedigrees, DDX41 variant segregation, protein domain organization, variant distribution, and conservation across species. (a) Pedigrees of the nine analyzed families, showing DDX41 variants (M1-M8) described at the cDNA and protein levels, and their segregation. “+” indicates the normal allele; asterisks (*) mark variants previously reported in Diness et al. 2018 (Family 9). Symbols filled in black denote individuals affected with EOSRD, with or without extraocular involvement ( Supplementary Table 2 ); those filled in grey indicate visual symptoms without complete ophthalmological evaluation (Family 9). (b) Protein domain organization and positions of DDX41 variants. Variants are predominantly located in the DEAD-box domain. aa, amino acids. (c) Conservation of the amino acid sequence at the variant sites across species. The arrow indicates the amino acid affected by the variant. BPDCN: blastic plasmacytoid dendritic cell neoplasm.

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: Variant Assay, Sequencing

Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: Structural modeling and biochemical analysis of DDX41 variants. (a) Predicted 3D structure of human DDX41 bound to poly-uridine RNA, modeled with AlphaFold3 and adjusted based on Dbp5-RNA structure ; N-terminal region truncated for clarity. (b-d) Locations of DDX41 variants on the DEAD domain, highlighting residues predicted to affect RNA binding (S266, P321) and the role of I396 in stabilizing the hydrophobic pocket and Q-motif. Models were visualized in ChimeraX . (e-g) Expression and purification of DDX41 variants from HEK293F cells. Proteins were visualized by (e) Coomassie staining, (f) after Heparin/StrepTactin affinity purification, by silver staining and (g) immunoblot of purified proteins for EMSA using an anti-DDX41 antibody shows co-purification of Hsp70, with increased association observed for the p.R339C variant, supporting partial misfolding of this mutant. Hsp70 represents a contaminant co-purifying with DDX41 rather than a specific target of the antibody. (h-j) EMSA analysis of DDX41 binding to RNA and structural/kinetic model. (h) Representative EMSA blot showing single (S) and double (D) and higher order (M) DDX41-RNA complexes that form sequentially. Schematic on the left depicts the molecular species producing the observed bands: free Cy5-labeled RNA (RNA shown as a black ribbon, Cy5 as a yellow box), RNA bound by a single (S) DDX41 (purple) and by two (D) DDX41 molecules. (i) Quantification of bound fractions at 100 nM protein; error bars represent standard deviation from at least two independent purifications (n = 4). The R339C variant shows a marked decrease in single-bound complexes and a concomitant increase in double-bound species, consistent with disrupted RNA binding and partial misfolding. (j) Kinetic model for DDX41-RNA binding, with a schematic depicting single (S), double (D), and higher-order (M) assemblies. Kon/Koff and K2on/K2off indicate association/dissociation rates for S and D complexes, respectively, while KMon/KMoff correspond to higher-order assemblies. Higher-order complexes may represent either aggregates of multiple DDX41 monomers or multimeric assemblies that bind RNA less specifically.

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: RNA Binding Assay, Expressing, Purification, Staining, Affinity Purification, Silver Staining, Western Blot, Copurification, Variant Assay, Mutagenesis, Binding Assay, Labeling, Standard Deviation

Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: DDX41 expression, stability, and subcellular localization in control and patient fibroblasts. (a, b) DDX41 protein levels in total cell lysates from control (CTL) and patient (P1) fibroblasts. (a) Representative Western blot; β-actin was used as a loading control. Molecular weight markers are shown on the right. (b) Quantification of DDX41 levels normalized to β-actin. Data are mean ± SEM from three independent experiments. **, p = 0.0013 (unpaired two-tailed Student t-test). (c, d) DDX41 expression in control (CTL) and patients (Ps, P1-P2) fibroblasts after treatment with 10 μM MG-132 for 18 h. (c) Representative Western blot; β-actin is the loading control. (d) Quantification of DDX41 levels in treated and untreated cells, normalized to β-actin. Data represent mean ± SEM from three independent experiments. **, p = 0.0039, ***, p = 0.0008, ****, p < 0.0001 (two-way ANOVA with Tukey post hoc test). (e, f) Immunofluorescence analysis of DDX41 (orange) and the nuclear speckles marker SC35 (green) (n=3). (e) Representative images of control and patient fibroblasts; merged images show predominantly nuclear DDX41 with partial colocalization with SC35. Scale bars, 10 µm. (f) Quantification of DDX41 fluorescence intensity in cytoplasmic and nuclear compartments. Data represent mean ± SEM from three independent experiments; ****, p < 0.0001 (two-way ANOVA with Tukey post hoc test).

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: Expressing, Control, Western Blot, Molecular Weight, Two Tailed Test, Immunofluorescence, Marker, Fluorescence

Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: Whole-transcriptome analysis in control and patient fibroblasts. (a) DDX41 transcript levels were comparable between the affected individual (P1) and healthy controls (CTLs), as determined by RNA-seq. Bars represent mean ± SEM; n.s., not significant (unpaired two-tailed Student t-test). (b) Volcano plot of differentially expressed genes (DEGs) between patient and control fibroblasts. The x-axis shows log₂ fold change (FC) in gene expression; the y-axis represents statistical significance as -log₁₀ adjusted false discovery rate (FDR). Genes with |log₂FC| ≥ 1.5, TPM ≥ 5, and FDR < 0.05 were considered significant. Significantly upregulated genes are shown in orange, downregulated genes in blue. (c) Venn diagrams depicting the overlap between DEGs and differentially spliced genes (DSGs) identified in patient versus control fibroblasts. (d) Distribution of splicing event types based on rMATS analysis. Skipped exons (SE) were the most frequent event, followed by alternative 5′ splice sites (A5SS), alternative 3′ splice sites (A3SS), retained introns (RI), and mutually exclusive exons (MXE). (e) RNA-seq quantification of spliceosomal small nuclear RNAs (snRNAs) U1, U2, U4, U5, and U6 in fibroblasts from the affected individual (P1) and controls (CTLs), expressed as reads per million (RPM). Bars represent mean ± SEM. (f) RT-qPCR validation of U1, U2, U4, U5, and U6 snRNA levels in fibroblasts from four controls and the affected individual. Bars represent mean ± SEM from three independent experiments; n.s., not significant (unpaired Welch t-test was used to assess statistical differences).

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: Control, RNA Sequencing, Two Tailed Test, Gene Expression, Quantitative RT-PCR, Biomarker Discovery

Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: Functional and structural characterization of the Ddx41 I396T/I396T knock-in mouse model. (a) Schematic of the CRISPR-Cas9 genome editing strategy used to generate the Ddx41 I396T/I396T mouse. Coding exons are shown as solid black bars, non-coding exons as open bars. The sgRNA sequence (black underline) targets exon 11, spanning the isoleucine codon at position 396 (I396). The protospacer adjacent motif (PAM) is shown in grey, and the introduced nucleotide substitution is indicated in bold. (b) Electroretinography (ERG) analysis of cone- and rod-mediated responses in wild-type (WT) and Ddx41 I396T/I396T mice at postnatal months 1, 3, and 6. Dot plots show a-wave and B-wave amplitudes. Statistical significance between age-matched genotypes was assessed by two-way ANOVA followed by Sidak post hoc test. Bars represent mean ± SEM from n = 10 mice per genotype per time point. Significance: *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001; ****: p ≤ 0.0001. (c) Representative retinal histology images of WT and Ddx41 I396T/I396T mice at postnatal months 1, 3, and 6 (40× magnification). Retinal layers appear intact in both genotypes across all time points. ONL: Outer Nuclear Layer; INL: Inner Nuclear Layer; GCL: Ganglion Cell Layer. (d) Retinal histology at postnatal month 12 showing thinning of the ONL and reduced photoreceptor nuclei in Ddx41 I396T/I396T mice compared with WT. (e) Quantification of ONL nuclei counts per 250 µm retinal length at postnatal month 12. Each dot represents one biological replicate (n = 4-6 retinas per genotype). Statistical analysis was performed using two-way ANOVA with Sidak correction for multiple comparisons. Bars represent mean ± SEM. Significance: *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001; ****: p ≤ 0.0001.

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: Functional Assay, Knock-In, CRISPR, Sequencing

Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: Transcriptomic profiling of retinas from 1 month-old wild-type and Ddx41 I396T/I396T mice. (a) Volcano plot of differentially expressed genes (DEGs) between wildtype (WT) and Ddx41 I396T/I396T mutant retinas (n = 4 per genotype) at postnatal month 1. The x-axis shows log₂ fold change (FC) in gene expression, and the y-axis shows statistical significance as -log₁₀ FDR-adjusted p-values. Genes with |log₂FC| ≥ 1.2, TPM ≥ 5, and P < 0.01 were considered significant. Upregulated genes are shown in orange, downregulated genes in blue. (b) Heatmaps generated using Metascape showing enrichment of Gene Ontology (GO) biological pathways among DEGs. Each row represents a gene set, and color intensity reflects relative expression levels (blue: low expression; orange: high expression).

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: Mutagenesis, Gene Expression, Generated, Expressing

Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: Proteomic analysis of from 1 month-old wild-type and Ddx41 I396T/I396T mice. (a) Volcano plot of proteins significantly upregulated (orange) or downregulated (blue) Ddx41 I396T/I396T mutant versus wild-type (WT) retinas (n = 3 per genotype) at postnatal month 1. Differentially expressed proteins were defined as |log₂FC| ≥ 1.5 and FDR ≤ 0.05. (b) Dot plots showing Gene Ontology (GO) biological pathway enrichment among upregulated proteins. Pathways related to ribonucleoprotein complex biogenesis and RNA splicing are prominently represented. (c) Dot plots showing GO biological pathway enrichment among downregulated proteins. Pathways related to cilium assembly, organization, and photoreceptor cell differentiation are prominently represented. (d) Table displaying GO enrichment analysis of the 50 most significantly altered proteins (FDR < 0.05). BP: Biological Process; CC: Cellular Component.

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: Mutagenesis, Cell Differentiation

Journal: medRxiv

Article Title: Biallelic germline variants in the hematologic malignancy predisposition gene DDX41 cause retinal dystrophy through dysregulation of retinal homeostasis

doi: 10.64898/2026.01.28.26344834

Figure Lengend Snippet: Immunolabeling of DDX41 and Müller glia markers in the retina of wild-type and Ddx41 I396T/I396T mice. (a, b) Western blot analysis and quantification of DDX41 expression in retinal protein extracts from wildtype (WT) and Ddx41 I396T/I396T mice (n = 4 per genotype) at 1 and 9 months of age. (a) Representative Western blot; β-catenin was used as a loading control, and molecular weight markers are indicated on the right. (b) Quantification of DDX41 levels normalized to β-catenin, showing reduced expression in Ddx41 I396T/I396T mice compared with WT at both time points. Data represent mean ± SEM, **, p = 0.0042, ***, p = 0.0002 (two-way ANOVA with Tukey post hoc test). (c-f) Immunolabeling and quantitative analysis of DDX41 and Müller glia organization in retinas from 8-month-old WT and Ddx41 I396T/I396T mice. (c-e) Immunofluorescence staining showing (c) DDX41 (orange), (d) glutamine synthetase (GS; cytoplasmic Müller glia marker), and (e) SOX9 (nuclear Müller glia-specific transcription factor). Scale bars, 50 µm. (f) Quantification of Müller glia nuclear spatial dispersion within the inner nuclear layer (INL) based on SOX9 staining between WT and Ddx41 I396T/I396T . The ratio between the convex hull area encompassing SOX9-positive nuclei and the total INL area was calculated using QuPath (version 0.5.1) . This dimensionless index reflects glial spatial organization, with higher values indicating broader nuclear dispersion. Bars represent mean ± SEM (n = 3 per genotype). *, p = 0.0307 (unpaired two-tailed t-tests). (g) GFAP immunostaining of retinal sections from 12-month-old WT and Ddx41 I396T/I396T mice, indicating gliosis. Scale bar, 50 μm.

Article Snippet: DDX41 was immunostained overnight at 4 °C using the following primary antibodies: polyclonal rabbit anti DDX41 (1:200; HPA017911, Merck), monoclonal mouse anti SC 35 (1:2000; S4045, Sigma Aldrich), and monoclonal mouse anti DNA/RNA hybrid (clone S9.6, 1:200; MABE1095, Merck).

Techniques: Immunolabeling, Western Blot, Expressing, Control, Molecular Weight, Immunofluorescence, Staining, Marker, Dispersion, Two Tailed Test, Immunostaining