Platelet proteomic profiling in sitosterolemia suggests thrombocytopenia is driven by lipid disorder and not platelet aberrations

Abstract

Sitosterolemia is a rare autosomal recessive genetic disorder in which patients develop hypercholesterolemia and may exhibit abnormal hematologic and/or liver test results. In this disease, dysfunction of either ABCG5 or ABCG8 results in the intestinal hyperabsorption of all sterols, including cholesterol and, more specifically, plant sterols or xenosterols, as well as in the impaired ability to excrete xenosterols into the bile. It remains unknown how and why some patients develop hematologic abnormalities. Only a few unrelated patients with hematologic abnormalities at the time of diagnosis have been reported. Here, we report on 2 unrelated pedigrees who were believed to have chronic immune thrombocytopenia as their most prominent feature. Both consanguineous families showed recessive gene variants in ABCG5, which were associated with the disease by in silico protein structure analysis and clinical segregation. Hepatosplenomegaly was absent. Thrombopoietin levels and megakaryocyte numbers in the bone marrow were normal. Metabolic analysis confirmed the presence of strongly elevated plasma levels of xenosterols. Potential platelet proteomic aberrations were longitudinally assessed following dietary restrictions combined with administration of the sterol absorption inhibitor ezetimibe. No significant effects on platelet protein content before and after the onset of treatment were demonstrated. Although we cannot exclude that lipotoxicity has a direct and platelet-specific impact in patients with sitosterolemia, our data suggest that thrombocytopenia is neither caused by a lack of megakaryocytes nor driven by proteomic aberrations in the platelets themselves.

Graphical abstract

Figure 1.

Patients’ genetics and hematologic characteristics before treatment. (A-B) Family pedigrees: double line shows a consanguineous union; filled symbol, affected individuals; half-filled symbols, heterozygous unaffected carrier; open symbols, not affected. (A) Pedigree A with propositus A-II-1, A-II-2, and A-II-3. (B) Pedigree B with propositus B-II-1. (C) Molecular annotation of the ABCG5 variants reported here. (D-F) Sterol levels in plasma. (D) Sitosterol. (E) Campesterol. (F) Cholestanol. (G) Platelet count. (H) Mean platelet volume. (I) Thrombopoietin (TPO). Dotted lines show the normal plasma parameters.

Figure 2.

Longitudinal hematologic parameters before and after treatment. Patients were followed up for ∼1 year, with samples collected throughout this time. (A) Sitosterol plasma levels. (B) Platelet count. (C) MPV.

Figure 3.

Proteomic depth and stability. (A) Dynamic range of quantified proteins; label-free quantification (LFQ) intensities are shown. (B) GO term enrichment network of all quantified proteins. The node size is proportional to the number of proteins associated with GO annotation. Node color intensity denotes P values, whereas edge width denotes term similarities based on Jaccard’s similarity. (C) Scatterplot of SD and mean protein abundances. (D) LFQ intensities of proteins covering the range of measurement. Lines denote the mean intensities.

Figure 4.

Proteomic landscape of platelets from controls and patients before and after treatment. (A) Volcano plot comparing the platelet proteomes of patients with sitosterolemia before treatment and healthy controls. Upregulated proteins in patients are shown in orange, whereas downregulated proteins are shown in blue. (B) Volcano plot comparing proteomic profiles of patients before and after treatment. Upregulated proteins after the treatment are shown in red. (C) Volcano plot comparing the platelet proteomes of patients with sitosterolemia after treatment and healthy controls. Upregulated proteins in patients are shown in red, whereas downregulated proteins are shown in dark blue. (A-C) Circles are drawn around granule-derived proteins: purple = alpha granules; green = dense granules. The difference in the expression is shown on the x-axis, and the logarithmic P value (–log(P value)) is shown on the y-axis. Dotted lines indicate the threshold of significance with the horizontal line marking a –log10 (adjusted P value) of 1, and the vertical dotted lines marking a log2(fold change) of 1. For a complete list of the significantly regulated proteins, refer to the supplemental Tables. (D) Heat map of Pearson correlation coefficient of the pairwise comparison of all quantified proteins in this study. The row and column splits are based on WGCNA-defined functional modules, which are numbered. Color gradients denote coefficients (purple: −1, white: 0, orange:1). A heat map of the median module intensity per condition is annotated below. Colors reflect the LFQ intensity. (E) GO term enrichment of functional modules. Color indicates module, node size indicates amount of proteins with a GO term annotation, and edge width denotes GO term similarities based on Jaccard’s similarity. (F-G) Box plots showing protein intensities (LFQ) of a selection of proteins in module 8 associated with an inflammatory response (F) or blood microparticles (G). The black dots represent the individual measurements.

Figure 5.

Proteomic landscape of platelets from controls and patients before and after treatment. Volcano plot comparing proteomic profiles before and after treatment of individual patients. (A) Index case A-II-1. (B) Index case A-II-2. (C) Index case A-II-3. (D) Index case B-II-1. Upregulated proteins in patients are shown in orange, whereas downregulated proteins are shown in dark blue. Difference in expression is shown on the x-axis, and the logarithmic P value (–log(P value)) is shown on the y-axis. Dotted lines indicate the threshold of significance with the horizontal line marking a –log10 (adjusted P value) of 1, and the vertical dotted lines marking a log2(fold change) of 1. (E) GO term enrichment of functional modules. Color indicates module, node size indicates amount of proteins with a GO term annotation, and edge width denotes GO term similarities based on Jaccard’s similarity. (F) Box plots showing the protein intensities (LFQ) of a selection of mitochondrial proteins. Black dots represent individual measurements.