Expanding the genetic spectrum of TUBB1-related thrombocytopenia

Abstract

β1-Tubulin plays a major role in proplatelet formation and platelet shape maintenance, and pathogenic variants in TUBB1 lead to thrombocytopenia and platelet anisocytosis (TUBB1-RT). To date, the reported number of pedigrees with TUBB1-RT and of rare TUBB1 variants with experimental demonstration of pathogenicity is limited. Here, we report 9 unrelated families presenting with thrombocytopenia carrying 6 β1-tubulin variants, p.Cys12LeufsTer12, p.Thr107Pro, p.Gln423*, p.Arg359Trp, p.Gly109Glu, and p.Gly269Asp, the last of which novel. Segregation studies showed incomplete penetrance of these variants for platelet traits. Indeed, most carriers showed macrothrombocytopenia, some only increased platelet size, and a minority had no abnormalities. Moreover, only homozygous carriers of the p.Gly109Glu variant displayed macrothrombocytopenia, highlighting the importance of allele burden in the phenotypic expression of TUBB1-RT. The p.Arg359Trp, p.Gly269Asp, and p.Gly109Glu variants deranged β1-tubulin incorporation into the microtubular marginal ring in platelets but had a negligible effect on platelet activation, secretion, or spreading, suggesting that β1-tubulin is dispensable for these processes. Transfection of TUBB1 missense variants in CHO cells altered β1-tubulin incorporation into the microtubular network. In addition, TUBB1 variants markedly impaired proplatelet formation from peripheral blood CD34+ cell-derived megakaryocytes. Our study, using in vitro modeling, molecular characterization, and clinical investigations provides a deeper insight into the pathogenicity of rare TUBB1 variants. These novel data expand the genetic spectrum of TUBB1-RT and highlight a remarkable heterogeneity in its clinical presentation, indicating that allelic burden or combination with other genetic or environmental factors modulate the phenotypic impact of rare TUBB1 variants.

Graphical abstract

Figure 1.

Family pedigrees and location of the β1-tubulin variants. (A) Pedigrees of the affected families with inherited thrombocytopenia. The index cases are indicated with black arrows. The upper right quarter red shading in symbols indicates individuals with thrombocytopenia, lower right quarter green shading denotes increased MPV, and upper and lower left quarter blue shading indicates heterozygous or homozygous for TUBB1 variant. NA, not available for the study; unlabeled symbols correspond to nonblood relatives not included in the study. (B) Representative peripheral blood smears from pedigree I after May-Grünwald Giemsa staining (×100). Variable platelet size was observed with large (arrows) and giant (cross) platelets. (C) Schematic representation of the β1-tubulin protein with all the reported variants. Variants in pedigrees reported in the present study are highlighted in bold; underlined variant is novel and the one within blue rectangle was found in homozygosity. (D) Structural analysis of the p.Gly109Glu missense variant using a β1-tubulin 3D model (software https://www3.cmbi.umcn.nl/hope/). The protein is colored in gray; the side chains of both the wild-type and the mutant residues are shown in green and red, respectively.

Figure 2.

Effect of the novel missense variants in a CHO cellular model. Distribution of (A) transfected wild-type β1-tubulin, mutants p.260Ser, p.107Pro, p.359Trp, and p.269Asp and (B) mutant p.109Glu (in a different set of experiments), in transfected CHO cells by immunostaining for β1- (green) and α-tubulin (red). (C) Microtubule length in CHO cells transfected with wild-type or mutant β1-tubulin was assessed using ImageJ software. All β1-tubulin mutants resulted in significantly shorter microtubules. All images were acquired with a Carl Zeiss Axio Observer A1 fluorescence microscope with a 63× objective lens. Scale bars are 20 μm. Data are means ± SD of values obtained from at least 10 different microscopy fields. *P < .05; **P < .005.

Figure 3.

Effect of the p.Arg359Trp β1-tubulin variant in platelets and megakaryocytes. The functional and structural consequences of p.Arg359Trp β1-tubulin variant were assessed in members of Pedigree F. Immunofluorescence analysis of β1-tubulin was performed in platelets on poly-l-lysine–coated coverslips under (A) resting (platelet-rich plasma) and (B) spreading (washed platelets) conditions from p.Arg359Trp carriers with (II.1 and III.1) or without thrombocytopenia (III.2), a noncarrier (III.3), and an unrelated healthy control. Platelets were labeled with a β1-tubulin antibody (red) and with fluorescein isothiocyanate-phalloidin (green). Scale bars are 5 μm, and the objective lens is 63×. (C) Bar plot showing the percentage of spread platelets relative to total adhered platelets. Data are means ± SD of values obtained from at least 10 different microscopy fields in patient and control samples. (D) Representative Western blot picture of β1-tubulin levels in platelet lysates from p.Arg359Trp carriers with or without thrombocytopenia, a noncarrier, and 2 healthy controls, using β-actin as internal control. (E) Illustrative images of CD34⁺ peripheral blood cell–derived megakaryocytes in patients from Pedigree F (p.Arg359Trp carriers with or without thrombocytopenia and a noncarrier) and healthy controls. MKs were labeled with β1-tubulin (green) and rhodamine-phalloidin (red). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; blue). Images were acquired in a Carl Zeiss Axio Observer A1 fluorescence microscope with a 63× objective lens. Scale bars are 5 μm. (F) Classification of MKs in family members and controls. Polynucleated cells extending protrusions with terminal tips were considered as proplatelet-forming MKs, whereas those displaying a flattened shape with actin organized in focal adhesion points and fibers as spread MKs. Some MKs were small and isolated, whereas MKs that were present in aggregates of 2 or more MKs were defined clusters. At least 100 MKs from 5 different replicates were analyzed; ^#P < .05 vs TCP carriers. (G) Microtubule length in CD34⁺ peripheral blood cell–derived MKs from controls and patients with the p.Arg359Trp β1-tubulin variant was assessed using the ImageJ software. Data are mean ± SD of values obtained from at least 6 different microscopy fields. TCP, thrombocytopenic; non-TCP, nonthrombocytopenic. **P < .005.

Figure 4.

Effect of the p.Gly269Asp β1-tubulin variant in platelets and megakaryocytes. The functional and structural consequences of the novel p.Gly269Asp β1-tubulin variant were assessed in members of Pedigree H. (A) Immunofluorescence analysis of β1-tubulin was performed on poly-l-lysine–coated coverslips in platelets (platelet-rich plasma) from p.Gly269Asp carriers with (II.2 and III.3) or without thrombocytopenia (II.3) and a noncarrier (II.4). Platelets were labeled with a β1-tubulin antibody (red). Scale bars are 5 μm, and the objective lens is 63× (magnification 3×). (B) Western blot pictures of β1-tubulin levels in platelet lysates; β-actin was used as an internal control. (C) Representative images of CD34⁺ peripheral blood cell–derived MKs in a thrombocytopenic p.Gly269Asp carrier (II.2), noncarrier (II.4), and healthy control. (D) Representative images of spread MKs. MKs were labeled with β1-tubulin (green) and rhodamine-phalloidin (red). Nuclei were stained with DAPI (blue). The white arrow indicates a MKs with disorganized β1-tubulin. Images were acquired in a Leica SP8 confocal microscope with a 63× objective lens (magnification 1.5×). Scale bars are 5 μm. (E) Classification of MKs in family members and controls (as described in Figure 3; ^#P < .05 vs TCP carriers). (F) Microtubule length of CD34⁺ peripheral blood cell–derived MKs from controls and patients with p.Gly269Asp β1-tubulin variant was assessed using ImageJ software. Values are mean ± SD of values obtained from at least 10 different microscopy fields. TCP, thrombocytopenic; non-TCP, nonthrombocytopenic. **P < .005.

Figure 5.

The variant Gly109Glu in β1-tubulin alters platelet ultrastructure. Electron microscopy of platelets from 2 homozygous and a heterozygous carrier of the β1-tubulin p.Gly109Glu and from 2 healthy controls. In contrast to control platelets, most platelets from homozygous p.Gly109Glu carriers were round and large and showed a slightly hypertrophic open canalicular system and an irregular membrane demarcation system (white arrows and asterisks, respectively). Magnification for each image is shown (8000-15 000×). Images were acquired in a Philips/FEITecnai12 transmission electron microscope.

Figure 6.

The p.Gly109Glu variant disrupts β1-tubulin incorporation into microtubules with a minimal effect on platelet spreading. Representative immunofluorescence analysis of β1-tubulin in patient (Pedigree I) and control washed platelets in (A) resting and (B-C) spreading conditions. Platelets were labeled with a β1-tubulin antibody (red) and with FITC-phalloidin (green). Images were acquired in a Carl Zeiss Axio Observer.A1 fluorescence microscope with a 100× objective lens. Scale bars are 5 μm. Platelet spreading on poly-l-lysine was quantified, and the percentage of totally spread platelets relative to the total adhered platelets is shown in the bar plot. Values are mean ± SD. Images were acquired in a Leica SP8 confocal microscope with a 63× objective lens (magnification, 2×). Scale bars are 20 μm. (D) Western blot pictures of β1-tubulin levels in platelet lysates from patients and controls, using 2 different β1-tubulin antibodies; β-actin was used as internal control.

Figure 7.

Effect of the TUBB1 p.Gly109Glu variant in CD34⁺ peripheral blood cell–derived MK cultures. (A) Representative images of CD34⁺ peripheral blood cell–derived MKs in 2 controls and in homozygous carriers (pedigree I) of the novel variant p.Gly109Glu. (B) Detailed images of spread MKs. MKs were labeled with β1-tubulin (green) and rhodamine-phalloidin (red). Nuclei were stained with DAPI (blue). Images were acquired in a Carl Zeiss Axio Observer.A1 fluorescence microscope with a 63× objective lens. Scale bars are 5μm. (C) Classification of MKs (as described in Figure 3; ^#P < .05) in homozygous carriers and controls. (D) Microtubule length in MKs from controls and homozygous carriers of the p.Gly269Asp β1-tubulin variant was assessed using ImageJ software. Values are mean ± SD of values obtained from at least 6 different microscopy fields. *P < .05.