Vexas Syndrome Is a New Cause of p53-Mediated Erythroblastopenia

Introduction:

VEXAS (vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic) is a recently identified autoinflammatory disorder caused by somatic mutations in the UBA1 gene, resulting in a deep but cytoplasmic-restricted ubiquitylation defect¹. VEXAS is characterized by a macrocytic anaemia despite inflammation, which is not fully understood. We hypothesized that UBA1-mutated erythroblasts could generate pathological red cells responsible for the anaemia and contributing to systemic inflammation through impaired erythrophagocytosis. Thus, we characterized red cells and erythroblasts in VEXAS.

Methods:

Blood and marrow samples from VEXAS patients were collected under ethics approval number DC2016-2618 in Paris Hospitals. Mass spectrometry-based proteomics was performed on patients and controls erythrocytes using label-free protein quantification. Absolute copy number quantification for total erythrocyte proteomes was done using calculated MCH values. Erythropoiesis from peripheral blood (PB) and cord blood (CB) CD34+ cells and Hudep-2 erythroid cell line was analysed as previously described² and edited using the ABEmax enzyme two days after isolation. TUBE ubiquitin capture was used to analyse protein ubiquitinylation.

Results:

Red cells isolated from VEXAS patients did not present significant differences in mean projected surface area, lactadherin staining, mitochondrial retention, deformability and polyubiquitin content compared to age and sex-matched controls (p > 0.35 for each, n=7). Patients and controls erythrocyte proteomes were highly correlated in absolute quantification (r² > 0.95 for each comparison, n =5). By using primary bone marrow samples of VEXAS patients and cell line (Hudep-2) or CB cells edited with an UBA1 mutation and cultivated to induce erythroid differentiation, we were able to show that i) UBA1 pathogenic variants were present in patients non-erythroid bone marrow cells (GPA-) (VAF 31% +/- 12.1, n= 4), but not in polychromatophilic or orthochromatophilic erythroblasts (GPA+Band3^highC49d+ and GPA+Band3^highC49d- (1% +/-2 and 1% +/- 1, respectively, n=4). ii) Knocking-in the p.Met41Thr variant resulted in massive lethality in the proerythroblastic Hudep-2 cell line and in CB CD36+ erythroid cultures (day 5 mortality 62% +/- 8.3). iii) No massive lethality was observed in cord blood CD36- monocytic/macrophagic cultures (21% +/- 9.2). In vitro erythropoiesis from patients PB CD34+ cells showed clearance of UBA1 pathogenic variants (day 0 VAF 68.25% +/- 15.0 versus day 15 VAF 1%, SD = 1.73, n=4) and generated mature erythroblasts were UBA1 negative. We assessed P53 in our base edited CB model and showed that i) ubiquitylation of P53 was defective in base edited cells in TUBE capture and immunoprecipitation assays, and as a consequence ii) P53 was markedly overexpressed six days after base editing at an early stage of erythroid differentiation and two days before cell death onset (signal quantification 7.89 +/- 0.42 vs 1.25 +/- 0.18 in the mock condition).

Conclusion:

VEXAS-causing UBA1 variants result in massive cell death at the early stages of the erythroid differentiation, explaining the absence of such variants in patients erythroid marrow precursors and the absence of pathological features in circulating red cells. Our findings enlight a previously unknown role of cytoplasmic ubiquitylation in the regulation of P53 expression, and suggest that P53 is a key player of VEXAS physiopathology as in Diamond Blackfan and Fanconi anaemias. Collectively, these findings highlight a previously unknown role of P53 as a key player of VEXAS phyisopathology with a tightly regulated P53 expression during erythroid differentiation. Further treatment choices may include that erythropoiesis in VEXAS is carried out by the UBA1-WT marrow compartment.

1 Beck, D. B. et al. Somatic Mutations in UBA1 and Severe Adult-Onset Autoinflammatory Disease. N. Engl. J. Med. 383, 2628–2638 (2020)

2 Lamarque, M. et al. Role of Caspase-10-P13tBID axis in erythropoiesis regulation. Cell Death Differ. 30, 208–220 (2023).