A Novel Cause of Type 1 VWD: Impaired Exocytosis of Weibel-Palade Bodies Due to Biallelic Mutations in Madd

Background: Quantitative and qualitative defects of the hemostatic plasma protein von Willebrand factor (VWF) give rise to the inherited bleeding disorder von Willebrand disease (VWD). Approximately 30% of the patients with type 1 VWD with VWF <50 IU/dL do not have pathogenic VWF variants. While this suggests that other determinants outside the VWF locus can be responsible for VWF deficiencies, so far, no other genes have been causally implicated in VWD. VWF in plasma is synthesized by endothelial cells, where it resides in secretory organelles called Weibel-Palade bodies (WPBs) before exocytosis into the vascular lumen. We previously identified the Rab3/Rab27 guanine nucleotide exchange factor MADD as an important regulator of VWF secretion in cultured endothelial cells. Biallelic mutations in MADD have recently been implicated in a rare multisystem disorder that also presents with hematological abnormalities and bleeding diathesis. Aim: In this study we aimed to determine the in vivo relevance of MADD in regulation of plasma VWF by studying patients with biallelic MADD mutations. Results: We isolated Endothelial Colony Forming Cells (ECFCs) from 3 unrelated pediatric patients with biallelic mutations in MADD (P1: 4 yr (f), p.R327*, p.L1467Cfs*20; P2: 14 yr (f), p.G305V, p.?; P3: 12 yr (m), p.S1213*, p.R1532*), heterozygous family members and matching controls. Plasma VWF levels were within normal range for all heterozygous parents (100-126 IU/dL). In all 3 biallelic MADD patients VWF:Ag was reduced (P1: 22, P2: 30 and P3: 24 IU/dL) and below the cutoff for VWD type 1. Notably, P3 experienced recurrent nosebleeds, requiring frequent emergency hospital visits and blood transfusions. Genomic analysis using our in-house developed Illumina Next-Generation Sequencing-based gene panel confirmed the MADD genotype in all study participants, but did not identify pathogenic VWF variants in any of the 3 patients. To determine the effect of MADD mutations on protein composition and abundance in endothelial cells, as well as on the broader protein expression profile, we performed whole proteome analysis using mass spectrometry-based label-free quantification. MADD specific peptides were robustly detected in control (mean coverage 17%) and family member ECFCs (mean coverage 11-14%), but were entirely undetectable in patient ECFCs, indicating that our patient-derived biallelic MADD mutant ECFCs represent a model of MADD deficiency. Strongly correlating with the absence of MADD were reductions in protein expression of VWF and two components of the WPB secretory machinery: the MADD substrate Rab3D and Rab-effector Slp4-a. To study the functional consequences of loss of MADD we determined activation state and recruitment of Rab3D and Rab27A in patient ECFCs. Pull-down active GTPase assays for endogenous Rab3D and Rab27A in control and patient ECFCs revealed that all 3 patient ECFCs had strongly reduced levels of active, GTP-bound Rab3D (75-97% reduction, p< 0.001) and Rab27A (77% reduction, p< 0.001) compared to controls. Morphological analysis showed this was accompanied by complete absence of Rab27A, Rab3D and Slp4-a on WPBs, indicating that loss of MADD severely impairs acquisition of exocytotic machinery to WPBs. To investigate the impact on secretion of WPB cargo we evaluated histamine-induced secretion of VWF from patient ECFCs. Interestingly, we found severely reduced stimulus induced VWF secretion in ECFCs from P1 (80% reduced, p<0.01) and P3 (77% reduced, p<0.01) compared to controls. Finally, we determined the effect on exocytotic behavior of WPBs using a live cell optical imaging assay in which WPBs were fluorescently labeled with a VWF propeptide-StayGold probe and fura-2 was used to monitor intracellular calcium levels. Patient ECFCs showed strongly reduced histamine-induced degranulation (P2: 17%; P3: 5%) compared to controls (36-39%) and significantly increased delay of first fusion events following rise of intracellular calcium (P2: 12±2 s, P3: 26±9 s; control: 3±1 s). Conclusion: Taken together, our data are consistent with a model in which MADD determines secretion competence of WPBs and magnitude of VWF secretion by regulating the acquisition of exocytotic machinery to WPBs. This study underscores the in vivo relevance of the contribution of WPB exocytosis to plasma VWF levels and identifies MADD as the first causal gene for type 1 VWD in patients without pathogenic VWF variants.