Phycoerythrin (PE)-conjugated hamster anti-CD61 (Itgb3) monoclonal antibody (cat. no. 553347) and PE/cyanine7 (CY7)-conjugated mouse anti-CD41 (Itga2b) antibody (cat. no. 133916) were purchased from BD Pharmingen (BD Biosciences) and Biolegend, Inc., respectively. Rabbit anti-CD3 monoclonal antibody (cat. no. MAB4841) was purchased from R&D Systems China Co., Ltd. and rat anti-transferrin R (CD71) monoclonal antibody (8D3; cat. no. NB 100-64979-0.05mg) was purchased from Novus Biologicals, LLC. Rabbit anti-CD3ε monoclonal antibody (cat. no. 99940S) was purchased from Cell Signaling Technology, Inc. Rat anti-CD19 antibody (1D3; cat. no. ARG55048) was purchased from Arigo Biolaboratories. Goat anti-rat IgG (H+L) HRP (cat. no. FMS-Rt01) for immunohistochemistry was purchased from Fcmacs Biotech Co., Ltd. Goat anti-rabbit IgG (H+L) HRP and DAB buffer kit (cat. no. RQ7100) for immunohistochemistry was purchased from Quanhui Imp and Exp Int'l Co., Ltd. Alexa-Fluor 647-conjugated human fibrinogen (cat. no. F35200) was purchased from Molecular Probes (Thermo Fisher Scientific, Inc.). Tetramethyl rhodamine isothiocyanate (TRITC)-conjugated phalloidin (cat. no. P1951) was purchased from Sigma-Aldrich (Merck KGaA). Rabbit anti-Itgb3 polyclonal antibody (cat. no. 4702) for western blotting was purchased from Cell Signaling Technology, Inc. Rabbit anti-β-actin polyclonal antibody (cat. no. 4968) for western blot was purchased from Cell Signaling Technology, Inc. HRP-conjugated goat anti-rabbit IgG secondary antibodies (cat. no. 5196-2504) was purchased from Bio-Rad, Laboratories, Inc. RIPA buffer (cat. no. 89900) and protease inhibitor mixture (cat. no. 87786) were purchased from Thermo Fisher Scientific, Inc. The 2X SDS Laemmli sample buffer (cat. no. S3401) was purchased from Sigma-Aldrich. ECL (cat. no. 1705070) was purchased from Bio-Rad, Laboratories, Inc. Collagen type I (cat. no. 385) and ADP (cat. no. 384) were purchased from Chrono-Log Corporation. Purified human fibrinogen (cat. no. FIB1) was purchased from Enzyme Research Labs Inc. Multimer purified human von Willebrand Factor (vWF; cat. no. HCVWF-0190) was purchased from Hematologic Technologies, Inc. Calcein AM (cat. no. C1430) was purchased from Thermo Fisher Scientific, Inc. D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone dihydrochloride (PPACK; cat. no. BML-P1117-0025) was purchased from Enzo Life Sciences, Inc. Arg-Gly-Asp-Ser (RGDS) peptide and the protease activated receptors 4 (PAR4)-thrombin receptor activating peptide, AF [Ala-Tyr-Pro-Gly-Lys-Phe (AYPGKF)], were synthesized at GL Biochem (Shanghai) Ltd. Mouse plasma ferritin (cat. no. JL11908) and mouse fecal occult blood test ELISA test kits (cat. no. JL50073) were purchased from Shanghai Jianglai Biological Technology Co., Ltd. All other biochemical reagents were obtained from Sigma-Aldrich (Merck KGaA).

The Itgb3^−/− C57BL/6 mice were received as a generous gift from Professor J. Liu (Shanghai Jiao Tong University School of Medicine, Shanghai, China). Wild-type (Itgb3^+/+) C57BL/6 mice were obtained from the Animal Experiment Center of Jiangsu University (Jiangsu, China). A total of 2 Itgb3^−/− male mice were mated with 2 Itgb3^+/+ female mice to obtain Itgb3 heterozygous (Itgb3^+/−) mice. Subsequently, Itgb3^−/− male mice were mated with Itgb3^+/− female mice to obtain continuous offspring Itgb3^−/− mice. A total of 50 mice per group were used. The mice were housed (5 mice per cage) under a 12-h light/dark cycle at 23°C in a specific-pathogen-free environment with ad libitum access to autoclaved food and water. The cages were changed each week. Routine sanitation and environmental controls, including temperature, humidity (40–70%), ventilation, illumination and light schedule and noise abatement, were performed by the animal care staff. Mice (age, 6–8 weeks; 1:1 male/female ratio) were used for each procedure with 3–10 mice/group. The animal study protocol was approved by the Jiangsu University Institutional Animal Care & Use Committee (approval no. UJS-IACUC-AP-20190307021).

Itgb3^−/− mice (offspring) were identified by PCR according to previously published methods (11,14). Whole blood containing the anticoagulant sodium citrate was collected by cardiac puncture from Itgb3^−/− mice following anesthetization with an intraperitoneal injection of 60 mg/kg pentobarbital sodium. Subsequently, 20 µl of the blood was put aside for measurement of whole-blood count. The anesthetized mice were sacrificed by cervical dislocation after whole blood collection. Platelets were separated from whole blood by sequential centrifugation as previously described (14). The platelet suspensions were rested at room temperature for 1 h before being used for spreading assay, fibrinogen binding assay, adhesion assay and other experiments. Itgb3^+/+ and/or Itgb3^+/− mice served as controls.

Briefly, 10 µl whole blood containing the anticoagulant sodium citrate was collected from the Itgb3^+/+, Itgb3^+/− and Itgb3^−/− mice by tail cutting. The whole blood diluted in PBS was incubated with PE-conjugated hamster anti-CD61 (Itgb3) monoclonal antibody (1:100) and PE/CY7-conjugated mouse anti-CD41 (Itga2b) antibody (1:50) at room temperature for 30 min. The expressions of Itga2b and Itgb3 in platelets were tested using flow cytometry (n=3 mice/group).

Itgb3^−/− platelets (1×10⁸) were lysed for 30 min at 4°C in ice-cold using RIPA buffer containing the protease inhibitor mixture to obtain proteins. Following centrifugation of the platelet lysate at 13,800 × g for 15 min at 4°C, the supernatant was added to an equal volume of 2X SDS Laemmli sample buffer. The proteins from 1×10⁸ platelets per mouse were dissolved in 200 µl buffer, 10 µl of which was loaded per lane. The proteins, after being boiled at 100°C for 5 min, were separated by SDS-PAGE using 12% gels and transferred to PVDF membranes. Membranes were blocked with 3% BSA for 1 h at room temperature, then incubated with rabbit β3 polyclonal antibody (1:1,000) or rabbit anti-β-actin polyclonal antibody (1:1,000) overnight at 4°C followed by HRP-conjugated goat anti-rabbit IgG secondary antibodies (1:10,000). Immunoreactive bands were detected by ECL. The Itgb3^+/− and Itgb3^+/+ mice served as controls (n=3 mice/group).

Soluble fibrinogen binding assay was performed as previously described (14). Briefly, (1×10⁶/ml) washed platelets from Itgb3^−/− mice resuspended in HEPES-Tyrode's buffer (137.0 mM NaCl, 2.0 mM KCl, 12.0 mM NaHCO₃, 0.3 mM NaH₂PO₄, 1.0 mM CaCl₂, 1.0 mM MgCl₂, 5.5 mM glucose, 5.0 mM HEPES, 0.1% BSA, pH 7.4) were divided into several groups and treated at 37°C for 30 min with agonists alone or co-treated with antagonists as follow: Mn²⁺ (presented by formation of MnCl₂), Mn²⁺ + RGDS, ADP, ADP + RGDS, AF and AF + RGDS. The final concentration was 0.5 mM for Mn²⁺, 50 µM for ADP, 0.5 mM for AF peptide and 2 mM for RGDS. Following stimulation, the platelets were incubated with 100 µg/ml of Alexa-Fluor 647-conjugated fibrinogen for 30 min at 37°C in the dark. The reaction was then stopped by fixation with 4% formaldehyde for 15 min at room temperature. Then the platelets were washed with PBS and centrifuged at 600 × g for 5 min at room temperature. The Itgb3^+/+ and Itgb3^+/− platelets were used as controls. Fibrinogen binding of platelets was tested using a flow cytometer (Beckman Coulter, Inc.) and the data were analyzed with FlowJo 7.6 software (FlowJo LLC). Specific fibrinogen binding was calculated by total binding minus nonspecific binding in the absence of any agonists (n=3 mice/group).

The assay was performed according to the instruction manual of the Chrono-log whole blood lumi-aggregometer (Chrono-log Corporation). Briefly, blood containing 3.8% sodium citrate from Itgb3^−/− mice by cardiac puncture was collected and diluted 1:1 in physiological saline. Platelet aggregation was measured at 37°C in whole blood lumi-aggregometer (Chrono-log Corporation) using electrical impedance following the addition of 2 U/ml thrombin, 6 µg/ml collagen or 12 µg/ml collagen. An AC voltage in the millivolt range is applied to the probe circuit. During a brief period of equilibration, a monolayer of platelets forms on the exposed portions of the wires, resulting in a stable baseline of impedance which is assigned a value of zero ohms of resistance. Then, 20.0±0.2 ohms were added to the impedance baseline and the gain was adjusted, so that 20 ohms equals 50% of scale. Aggregation was recorded for 14 min. Itgb3^+/− and Itgb3^+/+ mice served as controls (n=3 mice/group).

Thrombus formation of Itgb3^−/− platelets under flow was performed as previously described (14). Briefly, 100 µg/ml collagen was perfused at a shear rate of 125 s⁻¹ (5 dynes/cm² for shear stress) for 2 min through the Bioflux 200 microfluidic channels (Fluxion Biosciences, Inc.) to coat the channels at 4°C overnight, followed by blocking with 2% BSA at room temperature for 1 h. The PPACK- anticoagulant whole blood (without sodium citrate) collected from Itgb3^−/− mice through cardiac puncture was labeled with the fluorescence dye, Calcein AM, and perfused at a shear rate of 1,500 s⁻¹ for 10 min through the microfluidic channels. The process of thrombus formation was recorded by a video using NIS D image software version 3.1 (Nikon Corporation) and Bioflux 200 software. Itgb3^+/− and Itgb3^+/+ mice served as controls (n=3 mice/group).

For the adhesion assays, a 96-well plate was coated with 20 µg/ml fibrinogen, 4 µg/ml collagen, or 4 µg/ml vWF at 4°C overnight and then blocked with 2% BSA at room temperature for 1 h. Washed platelets (1×10⁶/ml) from Itgb3^−/− mice were allowed to adhere for 1 h at 37°C. Next, the wells were rinsed with PBS to remove non-adherent or unstable adherent platelets. 4-Nitrophenylphosphate (PNPP) was used to quantify the adherent platelets. In detail, the buffer (1% Triton X-100, 3 mg/ml PNPP, 100 mM sodium acetate, pH 5.0) was added and incubated at 37°C for 1 h. Then, sodium hydroxide (0.5 M) was added to stop the reaction. The optical density (OD) value was read at a wavelength of 405 nm using a microplate reader. Washed Itgb3^+/+ platelets were used as a control.

Platelet spreading assays were performed as previously described (14). Briefly, the chamber slides were coated with 20 µg/ml fibrinogen overnight at 4°C and then blocked with 2% BSA. Washed platelets (2×10⁵/ml) resuspended in HEPES-Tyrode's buffer were allowed to adhere and spread on fibrinogen-coated slides at 37°C for 2 h. After washing three times with PBS, the attached platelets were fixed with 4% paraformaldehyde for 15 min at room temperature, permeabilized with 0.2% Triton X-100 and stained with TRITC-conjugated phalloidin as previously described (15). Itgb3^+/+ platelets were used as a control. The coverage area of spreading was measured using ImageJ software version 1.4.3.67 (National Institutes of Health). A total of five randomly selected fields from different tests were used for statistical analysis (n=3 mice analyzed from each group).

Itgb3^−/− mice were anesthetized with intraperitoneal injection of 60 mg/kg pentobarbital sodium. A 3-mm section was cut from the tail tip in 6-week-old mice and then the bleeding time was counted. Every 30 sec the wounded tails were touched to a filter paper to confirm whether the bleeding had stopped or not. The maximum time for counting was 30 min. Itgb3^+/− and Itgb3^+/+ mice served as controls. Simultaneously, blood smear slides were made from the cut tail followed by Wright's staining at room temperature for 15 min (n=10 mice/group).

A total of 3 8-week-old male Itgb3^+/+ mice were anesthetized with intraperitoneal injection of 60 mg/kg pentobarbital sodium in accordance with animal welfare standards. Mice were restrained and the neck gently pressed to cause head vein congestion. A sterilized capillary pipette was gently pressed into the posterior orbital venous plexus. The blood was naturally sucked into the tube. After the required blood had been obtained, the pressure on the neck was removed and the blood collection tube was pulled out at the same time to prevent postoperative bleeding from the puncture hole and sterile gauze or cotton ball was used to compress the eyeball to stop bleeding. During the operation, a veterinarian was nearby to observe the breathing and palpate heart rate. Following the operation, 0.4 ml of 0.9% sodium chloride was given through gavage twice to add fluid to prevent shock in accordance with animal welfare standards. The animal was covered with insulation blanket to keep it warm. The animal was placed in a cage in warm environment after awakening. Blood (0.5 ml) was collected every time and twice a week. After 1 month, CHM mice were established successfully, and three CHM mouse models were established. A total of 3 8-week-old male Itgb3^+/+ mice without treatment served as controls. The animals were kept individually, as aforementioned, and monitored daily during the CHM induction for animal welfare.

Itgb3^−/− mice were sacrificed after anesthetization with pentobarbital sodium injection by cervical dislocation and their organs, including the spleen, heart, kidneys and liver were collected and weighed. Mouse femurs were dissected with a scissor. Muscles and connective tissues were removed as thoroughly as possible. The cartilages on both end of the femur were removed. The marrow of femurs was leaked out and prepared for bone marrow smear. Wright staining and iron staining (Prussian Blue) at room temperature for 20 min were performed on bone marrow smears. Itgb3^+/+, Itgb3^+/− and CHM mice served as controls (n=10 mice/group).

Half of the spleens collected from Itgb3^+/+ and Itgb3^−/− mice were fixed in 10% formalin overnight at room temperature and subsequently embedded in paraffin. The other half of the spleens were embedded in optimal cutting temperature compound and sectioned (1 mm) at low temperatures and fixed with precooled acetone to make the frozen section. The spleen tissue sections were stained sequentially with hematoxylin and eosin at room temperature for 2–3 min. The immunohistochemical procedure was performed on paraffin sections or frozen sections according to a previous study (16). The slides were incubated with monoclonal anti-CD3 (1:50), anti-CD19 (1:200) or anti-CD71 (1:100) antibodies at 4°C overnight followed by goat anti-rat IgG (H+L) HRP (1:1,000) or goat anti-rabbit IgG (H+L) HRP (1:1,000) at 37°C for 1 h and diaminobenzidine substrate with hematoxylin counterstaining (n=3 mice/group). The slides were observed under a light microscope (Olympus BX53F; Olympus Corporation) with Smart V digital camera and analyzed using JD-801 Image analysis system (JEDA Science-Technology Development Co., Ltd.).

Whole blood containing the anticoagulant sodium citrate was collected by cardiac puncture from Itgb3^−/− mice after anesthetization as aforementioned. Then plasma was obtained by centrifugation at 1,000 × g for 30 min at room temperature. Plasma ferritin was tested using a ferritin ELISA test kit following the manufacturer's protocol. Itgb3^+/+ mice served as controls (n=8 mice/group).

Feces from Itgb3^−/− mice were collected then were tested using fecal occult blood ELISA test kit (Shanghai Jianglai Biological Technology Co., Ltd.) following the manufacturer's protocol. Itgb3^+/+ mice served as controls (n=8 mice/group).

SPSS version 18 (SPSS, Inc.) was used for statistical analysis. Quantitative data were expressed as the mean ± standard deviation. Differences between two groups were assessed using Student's t-test. For datasets containing three or more groups, one-way ANOVA followed by Tukey's test was used. P<0.05 was considered to indicate a statistically significant difference.

DNA was extracted from the tail tissue of Itgb3^+/+, Itgb3^+/− and Itgb3^−/− mice and then genotyped by PCR. A band of 538 bp was observed for Itgb3^−/− mice, 446 bp for Itgb3^+/+ mice and double bands for Itgb3^+/− mice (Fig. S1A). Western blotting results showed no Itgb3 protein expression in washed Itgb3^−/− platelets (Fig. S1B). Whole-blood from Itgb3^+/+, Itgb3^+/− and Itgb3^−/− mice were incubated with PE-conjugated anti-CD61 (Itgb3) antibody and PE/CY7-conjugated anti-CD41 (Itga2b) antibody and then subjected to flow cytometry. Almost no Itga2b and Itgb3 was expressed on the Itgb3^−/− platelet surface, compared with that on the Itgb3^+/− and Itgb3^−/− platelet surface (Fig. S1C-E).

The washed Itgb3^+/+ platelets can bind to soluble fibrinogen under the stimulation of Mn²⁺, ADP or AF peptide, which can be partially inhibited by co-treatment with the antagonist RGDS peptide (Fig. 1A). The fibrinogen binding of Itgb3^+/− platelets was similar to that of Itgb3^+/+ platelets. However, the Itgb3^−/− platelets did not bind to soluble fibrinogen under the stimulation of any agonists.

Electronic resistance mediated by platelet aggregation from Itgb3^+/+ mice was 8 ohms induced by 2 U/ml thrombin, 10 ohms induced by 6 µg/ml collagen and 11 ohms induced by 12 µg/ml collagen, whereas that from Itgb3^−/− mice was 2, 0 and 2 ohms, respectively (Fig. 1B).

On a collagen-coated surface at shear rates of 1,500 s⁻¹, the thrombus formed from Itgb3^+/+ mice was large and thick, showing strip-shaped distribution along the direction of blood flow (Fig. 1C and D). By contrast, the thrombus formed from Itgb3^−/− mice was scattered, thin and easily washed off.

The adhesion of washed Itgb3^−/− platelets was notably lower on the fibrinogen or collagen-coated surface compared with that of washed Itgb3^+/+ platelets, no difference was observed in the adhesion on the vWF-coated surface (Fig. 2A and B). The Itgb3^+/+ platelets spread well on the fibrinogen-coated surface with the regular actin arrangement. On the surface without any matrix (control), Itgb3^+/+ platelets also spread, but the actin arrangement was disordered and diffuse. However, the spreading of Itgb3^−/− platelets on fibrinogen-coated surface and control was notably impaired (Fig. 2C and D).

Itgb3^−/− mice exhibited prolonged bleeding time (Fig. 3A and B) after tail cutting and they were more prone to subcutaneous hemorrhage (Fig. 3C) at birth compared with Itgb3^+/+ mice. Itgb3^−/− mice had a mean red blood cell count of 6.38±2.02×10¹² cells/l, hemoglobin of 89.38±25.73 g/l and a platelet count of 536.18±195.58×10⁹ cells/l, which were lower compared with Itgb3^+/+ mice (Table I). Wright staining showed increased microcytic hypochromic erythrocytes in peripheral blood smear from Itgb3^−/− mice compared with that from Itgb3^+/+ mice (Fig. 3D).

The bone marrow smears of Itgb3^−/− mice exhibited significantly more early erythroblasts compared with Itgb3^+/+ mice (Fig. 4A and B), and the Itgb3^−/− mice exhibited a relatively lower amounts of iron granules compared with Itgb3^+/+ mice, indicating iron deficiency in Itgb3^−/− mice (Fig. 4C). Enlarged spleens were observed in Itgb3^−/− mice (Fig. 4D). The ratio of spleen to whole body weight was higher in Itgb3^−/− mice than in Itgb3^+/+ mice suggesting a splenomegaly in Itgb3^−/− mice (Fig. 5A). CHM mice were established to confirm whether CHM also exhibits splenomegaly and to determine whether the splenomegaly in Itgb3^−/− mice is a result of chronic bleeding (Fig. 5B). Immunohistochemical staining of spleen biopsies from Itgb3^−/− mice showed no change in the expression of CD3 and CD19, which are the markers of T lymphocytes and B lymphocytes, respectively, whereas the expression of CD71, which is a marker of erythrocytes, increased compared with Itgb3^+/+ mice (Fig. 5C and D).

Itgb3^−/− mice had significantly lower level of plasma ferritin compared with Itgb3^−/− mice (Fig. 5E). There was no difference of fecal occult blood between Itgb3^+/+ mice and Itgb3^−/− mice (Fig. 5F).