A novel dominant-negative variant of IRF8 in a mother and son: Clinical, phenotypic and biological characteristics

Abstract

Background: The few reported patients with pathogenic IRF8 variants have manifested 2 distinct phenotypes: (1) an autosomal recessive severe immunodeficiency with significant neutrophilia and absence of or significant decrease in monocytes and dendritic cells and (2) a dominant-negative form with only a decrease in conventional type 2 dendritic cells (cDC2s) and susceptibility to mycobacterial disease.

Objectives: Genetic testing of a child with persistent EBV viremia identified a novel IRF8 variant: c.1279dupT (p.∗427Leuext∗42). The variant was also found in his mother, who was subsequently diagnosed with a human papillomavirus-positive tumor. We sought to examine the pathogenicity of the identified IRF8 variant and its phenotypic and functional characteristics.

Methods: Immunophenotypic and functional flow cytometry, natural killer cell cytotoxicity, matrix-assisted laser desorption/ionization-time of flight mass spectrometry, T-cell receptor Vβ spectratyping, Sanger sequencing, RNA-sequencing, Olink proteomics, immunoblotting, molecular cloning, dual-luciferase reporter assay, immunofluorescence microscopy, and image analysis.

Results: The 42 amino acid C-terminal extension of the mutant IRF8 (∼4 kDa heavier than wild type) impaired IRF8 nuclear localization in a dominant-negative manner and inhibited IRF1/IRF8-mediated transcriptional activities. Both patients had a decrease in plasmacytoid dendritic cells (pDCs) and in cDC1s, a mild neutrophilia and a mild monocytosis. Their existing pDCs had impaired IFN-α production. On TLR engagement, the production of IL-1β, IL-6, IL-10, and IL-12 by their monocytes and of IL-12 by their myeloid DCs were within normal limits. Natural killer cell development and cytolytic activity were essentially normal. RNA-sequencing and proteomic approaches bolstered the phenotypic and functional findings.

Conclusions: This study defines the pathogenic nature of the c.1279dupT (p.∗427Leuext∗42) IRF8 variant, determines its dominant-negative mechanism of action, and broadens the existing phenotype of human IRF8 immunodeficiency.

Keywords: EBV; HPV; IRF8; dendritic cell; interferon; loss-of-stop variant; monocyte.

Figure 1.. Validation of the IRF8 Mutation

(A) IRF8 cDNA sequence and corresponding amino acids from healthy unrelated controls (Controls 1 and 2) and the patients. Arrow: location of the NM_002163.2:c.1279dupT, (p.*427Leuext*42) patient mutation within the stop codon. (B) IGV screenshot showing the read coverage for the mutant loci in the mother and son. (C) Immunoblot of PBMC lysates as indicated. The heavier bands in the patient samples represent the mutant IRF8 protein.

Figure 2.

(A) Hemoglobin levels, platelet, neutrophil and monocyte counts of the patients during the course of observation. The spikes observed in the mother’s samples happened during the peri-operative period for removing the HPV⁺ tumor and were temporary. (B) Flow cytometric analysis of the patients’ monocytes (B, left panel), pDCs and mDCs (B, middle panel) and cDC1s (B, right panel).

Figure 3.. Cytokine Production by pDC and mDC in Response to TLR Engagement

(A) Assessment of IFN-α production by pDCs in response to R848 and to ODN2216 in the patients and the maternal grandparents. (B) Assessment of IL-12 production by mDCs in response to R848 in the patients and the maternal grandparents.

Figure 4.

(A) Enumeration of the patients’ CD3⁺ T cells, CD4⁺ T cells, CD8⁺ T cells, CD19⁺ B cells and NK cells during the course of observation. (B) Flow cytometric analysis of the patients’ memory B cells (B, left panel), the patients’ immunoglobulin levels (B, three middle panels), and mass spectrometric analysis of their serum immunoglobulins (B, right panel). (C) Monitoring the phenotypic evolution of CD8⁺ T cells in the patients over time. (D) Assessment of the patients’ TCRVβ repertoire. (E) Evaluation of NK cell cytotoxicity in the patients. The numbers at or above the shaded area fall within the normal range for this assay. (F) Flow cytometric analysis of CD107a/b expression and of IFN-γ production in CD8⁺ T cells after stimulation with PMA and ionomycin. (G) Phosphorylation of STAT3 in response to IL-6, IL-10 and IL-21 in CD4⁺ T cells (G, left panel), and phosphorylation of STAT1 in CD14⁺ monocytes after stimulation with IFN-α and IFN-γ (G, right panel). The PerFix EXPOSE reagent used for this assay cuts CD4 epitopes. Therefore, the definition of CD4⁺ T cells in this instance is based on the cells being CD45⁺CD3⁺CD8⁻.

Figure 5.. RNA-seq and Proteomic Analysis in Patients with the IRF8 Mutation

(A) Immune cell deconvolution in the control and patient samples. (B) Genes common between the two patients with an abs(log₂ fold-change)>2. (C) Genome-wide RNA-seq correlation between the two patients. Log₂ fold-change reflects the comparison of each patient versus control sample. Pearson correlation rho=0.6996, p-value <2.2e⁻¹⁶. Highlighted are genes with abs(log₂ fold-change)>5 in both conditions. (D) Heatmap showing genes that are common between the two patients with an abs(log₂ fold-change)>2. (E) Pathway enrichment for genes with a log₂ fold-change>2 across both samples. (F) Heatmap showing IRF8 target genes that are common and show the same directionality in the two patient samples with a threshold of abs(log₂ fold-change)>0.5. (G) Heatmap showing relative abundance of proteins measured using the Olink Explore panel across patients. (H) Differential abundance of proteins across the two patient and two control plasma samples. The horizontal dashed line represents a threshold of p-value 0.05. Selected significantly changing proteins are labeled. (I) The relative changes in protein abundance in each patient with reference to controls. The log₂-transformed fold-change for each protein for the son’s and mother’s samples are plotted on the x-axis and the y-axis respectively. (J) The relative changes in RNA expression levels and protein abundance in patients. The log₂-transformed fold-change for common RNA transcripts and proteins for patient samples are plotted on the x-axis and the y-axis respectively.

Figure 6.. The C-terminal extension of the Mutant IRF8 Impairs Nuclear Localization of IRF8.

(A-C) 293T cells were transfected with the indicated plasmids. (A) Immunoblot of transfected 293T cell lysates. (B) Immunofluorescence microscopy of transfected cells. Yellow lines show cell boundaries. (C) MFI of YFP within the nucleus to that within cytoplasm of cells from (B). (D-F) 293T cells were transfected with the indicated plasmids. The 42 amino acid extension present at C-terminal end of the patient mutant IRF8 (‘tail’) was attached to the C-terminal end of the IRF4. (D) Immunoblot of transfected 293T cell lysates. (E) Immunofluorescence microscopy of transfected cells. (F) Quantification as in (C) for transfected cells in (E). Mut: Patient mutant (p.*427Leuext*42) (A-C), F: FLAG (A, D). Scale bar = 10 μm (B, E). (C and F) Each dot represents a single cell; dots are color-coded based on the 3 independent experiments. The results presented are representative or means ± SD from 3 experiments. Statistical analyses were performed using one-way ANOVA with Dunnett’s multiple comparisons. ****P < 0.0001.

Figure 7.. The IRF8 Mutation is Dominant-Negative and Impairs IRF1-mediated Transcription.

(A and B) 293T cells were transfected with WT or mutant YFP-IRF8 and mCherry-IRF8 plasmids. (A) Immunofluorescence microscopy of transfected cells as indicated. Yellow lines show cell boundaries. (B) MFI ratio of YFP/mCherry within the nucleus to that within cytoplasm of cells from (A). (C) IRF8 distribution in IRF8⁺ PBMCs from healthy donor controls and patients. (D) MFI ratio of IRF8 in the nucleus relative to the cytoplasm in (C). (A and C) Scale bar = 10 μm (A) or 3 μm (C). (B, D) Each dot represents a single cell; dots are color-coded based on the independent experiments. (E-G) 293T cells were transfected with HA-IRF1 and YFP-IRF8 plasmids as indicated for dual-luciferase reporter assays. (E) Immunoblot of transfected 293T cell lysates. (F and G) Dual-luciferase reporter assays for TAPBP (F) and IL12B (G) promoter regions. Each color-coded dot represents an independent experiment. Mut: Patient mutant (p.*427Leuext*42) (A, B, E-G). The results presented are representative or means ± SD from 3 (A, B, E-G) and 1 (C and D) experiments. Statistical analyses were performed using one-way ANOVA with Dunnett’s multiple comparisons. *P < 0.05, **P < 0.01, ****P < 0.0001. ns = not significant.