A novel mechanism of thrombocytopenia by PS exposure through TMEM16F in sphingomyelin synthase 1 deficiency

Abstract

Sphingomyelin synthase 1 (SMS1) contributes to the generation of membrane sphingomyelin (SM) and affects SM-mediated physiological functions. Here, we describe the hematologic phenotypes, such as reduced circulating platelets and dysfunctional hemostasis, in SMS1-deficient (SMS1-KO) mice. SMS1-KO mice display pathologic manifestations related to idiopathic thrombocytopenia (ITP), including relatively high amounts of peripheral blood reticulated platelets, enhanced megakaryopoiesis in the bone marrow and spleen, and splenomegaly. Deficiency of SMS1, but not SMS2, prevented SM production and enhanced phosphatidylserine (PS) externalization on the plasma membranes of platelets and megakaryocytes. Consequently, SMS1-KO platelets were excessively cleared by macrophages in the spleen. Multimer formation in the plasma membrane of TMEM16F, a known calcium (Ca2+)-activated nonselective ion channel and Ca2+-dependent PS scramblase, was enhanced; the result was PS externalization to outer leaflets through increased Ca2+ influx in immortalized mouse embryonic fibroblasts established from SMS1-KO mice (SMS1-KO tMEFs), as seen with SMS1-KO platelets. Thus, SMS1 deficiency changed the TMEM16F distribution on the membrane microdomain, regulating Ca2+ influx-dependent PS exposure. SMS1-KO tMEFs in which TMEM16F was knocked out by using the CRISPR/Cas9 system lacked both the Ca2+ influx and excess PS exposure seen in SMS1-KO tMEFs. Therefore, SM depletion on platelet membrane microdomains due to SMS1 deficiency enhanced PS externalization via a Ca2+ influx through TMEM16F activation, leading to elevated platelet clearance and causing hemostasis dysfunction through thrombocytopenia. Our current findings show that the SM-rich microdomain generated by SMS1 is a potent regulator of thrombocytopenia through TMEM16F, suggesting that its dysfunction may be a novel additional mechanism of ITP.

Graphical abstract

Figure 1.

Platelet numbers and bleeding time in WT, SMS1-KO, and SMS2-KO mice. (A-C) Complete blood count analyses were performed on whole blood extracted from WT, SMS1-KO, and SMS2-KO mice (n = 3 per group). The numbers of platelets (A), red blood cells (B), and white blood cells (C) in mice of each genotype are shown. (D) The tail vein bleeding times of WT (n = 18), SMS1-KO (n = 12), and SMS2-KO (n = 4) mice are shown. Values show the mean ± standard deviation. *P < .05; **P < .005.

Figure 2.

Assessment of extramedullary hematopoiesis and splenomegaly. (A) The reticulated platelets of WT (n = 10), SMS1-KO (n = 5), and SMS2-KO (n = 6) mice were stained with thiazole orange and analyzed by using flow cytometry. Paraffin sections of spleen (B) and bone marrow (C) were stained with hematoxylin and eosin (HE) or with anti-vWF antibody. The vWF-positive MKs in several randomly chosen high-power fields (HPF) were quantified. Scale bars, 50 µm. (D) Representative images of spleens from WT, SMS1-KO, and SMS2-KO mice. The ratios of the spleen to whole body weight were calculated in nine WT, ten SMS1-KO, and seven SMS2-KO mice. Values show the mean ± standard deviation. *P < .05; **P < .005.

Figure 3.

Generation of membrane SM in platelets and MKs. (A) Platelets were detected by using phycoerythrin-conjugated anti-CD41 antibody (anti–CD41-phosphatidylethanolamine [PE]) and flow cytometry. (B) Cell surface SM (stained with Venus-lysenin) in CD41⁺platelets from WT (n = 3), SMS1-KO (n = 4), or SMS2-KO (n = 4) mice was analyzed by flow cytometry and quantified as the mean fluorescent intensity (MFI). NS, no staining. (C) SM amounts in platelets from WT (n = 4), SMS1-KO (n = 6), or SMS2-KO (n = 5) mice were measured by using liquid chromatography–tandem mass spectrometry. (D) MKs were extracted from bone marrow cells, then stained with Venus-lysenin and anti–CD41-PE. Nuclei were counterstained with 4,6-diamidino-2-phenylindole (DAPI), and images were obtained with confocal microscopy. The fluorescence intensity of Venus-lysenin was quantified by using ImageJ software (National Institutes of Health, Bethesda, MD) and is indicated as arbitrary units (AU). Bars, 20 µm. DIC, Differential interference contrast. Values show the mean ± standard deviation. *P < .05; **P < .005.

Figure 4.

PS exposure and Ca²⁺ influx in platelets and MKs. (A) Membrane PS on platelets from SMS1-KO or WT mice (n = 3 per group) was stained with FITC-conjugated annexin V (FITC–annexin V) and detected by using flow cytometry. WT mouse platelets were pretreated with calcium ionophore A23187 for 10 minutes before PS staining. The percentages of FITC-annexin V–positive platelets (CD41⁺) are presented. (B) MKs were stained with FITC–annexin V and anti–CD41-phosphatidylethanolamine (PE). The nuclei were then counterstained with 4,6-diamidino-2-phenylindole (DAPI), and images were obtained with confocal microscopy. The fluorescent intensity of FITC–annexin V was quantified with ImageJ software and is presented in arbitrary units (AU). Scale bar, 20 µm. Values show the mean ± standard deviation. (C) Sections of mouse spleen were stained with Alexa Fluor 488–conjugated anti-CD68 antibody and anti–CD41-PE, and then observed with fluorescent microscopy. Scale bar, 50 µm. (D) Splenectomy was performed in WT and SMS1-KO mice (n = 3-5 each). Blood was then collected from the tail vein, and platelet numbers were counted on the indicated day after surgery. Values show the mean ± standard deviation. (E) WT (n = 5) and SMS1-KO (n = 3) mice were intravenously injected with N-hydroxysuccinimide ester (NHS)-conjugated biotin. Platelets were then isolated from whole blood collected at the indicated time points and stained with FITC-conjugated anti-CD41 antibody and allophycocyanin-conjugated streptavidin. The biotinylated platelets were analyzed with flow cytometry. t_1/2 represents the half-life of platelets in hours. Values show the mean ± standard deviation. (F) Platelets from WT (n = 4) and SMS1-KO (n = 4) mice were stained with Furo-4 AM and anti–CD41-PE and then analyzed with flow cytometry. The intracellular Ca²⁺ levels in platelets were detected by measuring the Furo-4 AM fluorescence intensity and are presented as mean fluorescent intensity (MFI). Values show the mean ± standard deviation. *P < .05; **P < .005.

Figure 5.

PS externalization and cytosolic Ca²⁺ level in tMEFs. (A) WT, SMS1-KO, and SMS2-KO tMEFs were stained with FITC–annexin V, and then analyzed with flow cytometry. The fluorescence of FITC–annexin V was quantified as mean fluorescent intensity (MFI). (B) Intracellular Ca²⁺ levels were measured with Fura2-AM and calcium imaging. The fluorescence was measured with Argus (aHamamtsu Photonics, Hamamatsu, Japan) and is presented as the 340 nm/380 nm fluorescence ratio (R_340nm/380nm). Scale bars, 50 µm. (C) The DRM fraction was extracted with lysis buffer containing 1% Brij 58 (Sigma-Aldrich, St. Louis, MO) and OptiPrep (Axis-Shield Alere Technologies, Oslo, Norway) discontinuous gradients (5% and 30%). Fractions were collected from the top to the bottom of the gradient, and the DRM fractions (fractions 8-10) were analyzed by western blot analysis with antibodies against TMEM16F and the lipid raft marker flotillin-1. (D) Western blotting of TMEM16F protein in WT and SMS1-KO platelets was performed. (E) TMEM16F-KO tMEFs were established from SMS1-KO tMEFs by using CRISPER/Cas9 plasmids constructed based on lentiCRISPR-V2. The sequence of TMEM16F is shown at the top with the target sequence and the protospacer adjacent motif (PAM) sequence. Insertions are shown in red, and deletions are shown as black dashes. The change in length caused by each indel mutation is listed to the left of each sequence (+, insertion; −, deletion). TMEM16F protein in the SMS1-KO/TMEM16F-KO (1KO/16FKO) and SMS1-KO/vector (1KO/vec) tMEFs was detected by western blot analysis. Arrows indicate TMEM16F protein. (F) Cell surface PS was stained with FITC–annexin V and then analyzed with flow cytometry. The fluorescence was quantified as MFI. (G) Cytosolic Ca²⁺ levels were measured with Fura2-AM and are presented as the R_340nm/380nm. Scale bars, 50 µm. Values show the mean ± standard deviation. **P < .005.