Delineating Mechanisms Of Desialylated Platelet Clearance In The Liver

1.Determine whether Kupffer cells are the primary cells to clear desialylated platelets in the liverObjectiveMost platelet membrane receptors are glycoproteins(GP),which have sialic acids as the capping glycan structure.Recent studies have shown that platelet desialylation is critical for platelet clearance in many pathological states and the liver is the major site for the clearance of desialylated platelets.However,which cell type in the liver phagocytizing desialylated platelets in vivo has remained elusive.Since 2017,studies have suggested that Kupffer cells(KCs)phagocytize desialylated platelets,but there is no conclusive evidence in vivo.The liver also contains an abundance of a professional type of F4/80 positive phagocytes known as Kupffer cells,which comprises the largest population of resident tissue macrophages in the body.We aim to determine whether the Kupffer cell is the main cell type for the clearance of desialylated platelets in vivo.Methods1.Generating megakaryocyte/platelet specific reporter(ROSAmTmG;Pf4Cre)mice.ROSAmTmG mice(#007576,the Jackson Laboratory)were crossed with a transgenic mouse line expressing Cre recombinase under the control of the mouse Pf4 promoter(Pf4Cre mice,#008525,the Jackson Laboratory)to generate ROSAmTmG;Pf4Cre mice.This ROSAmTmG;Pf4Cre reporter mouse line specifically expresses enhanced green fluorescent protein(EGFP)in platelets,and tdTomato fluorescence in all other cell types.2.Tracking desialylated platelets in ROSAmTmG;Pf4Cre mice to determine the association between desialylated platelets and KCs.ROSAmTmG;Pf4Cre mice were intravenously injected with Arthrobacter ureafaciens-derived neuraminidase to remove sialic acids from platelet surface glycoproteins.In addition,the KCs from neuraminidase-treated ROSAmTmG;Pf4Cre mice were isolated and purified.Confocal microscopy imaging analysis was performed to determine the association of desialylated platelets with KCs and other cell types in the liver.3.KCs in mice were removed by clodronate liposomes and then the clearance of desialylated platelets in the liver was observed.Results1.ROSAmTmG;Pf4Cre mice were successfully generated,which is a novel mouse line critical for tracking the fate of platelets in vivo.2.Z-stack images of confocal microscope showed EGFP positive desialylated platelets were primarily associated with F4/80 positive KCs in the liver of neuraminidase-treated ROSAmTmG;Pf4Cre mice,but not in PBS-treated(untreated)control mice.Desialylated platelets were primarily present inside of the KCs,which was further confirmed by imaging of primary KCs freshly isolated from mice following neuraminidase treatment.In contrast,our in vivo tracking,and confocal microscopy imaging experiments did not detect any association of desialylated platelets with hepatocytes,which were considered a major cell type in the clearance of desialylated platelets.3.After removal of KCs by clodronate liposomes,the clearance of desialylated platelets in the liver was significantly reduced.ConclusionROSAmTmG;Pf4Cre mice were successfully generated.KCs are the primary cells in the liver to clear desialylated platelets.2.Generation and characterization of Clec4f-/-miceObjectivesDespite of many years of research,the major molecular receptor for mediating desialylated platelet clearance in the liver is still unclear.Our recent in vitro results suggest that KC receptor CLEC4F(C-type lectin domain family 4 member F or KC receptor,KCR)plays an important role in the clearance of desialylated platelets;however,convincing in vivo evidence has been lacking.To address this critical question,we aimed to generate and characterize mice lacking CLEC4F(Clec4f-/-mice).Methods1.Clec4f-/-mice were generated using CRISPR/Cas9-mediated(Clustered Regularly Interspaced Short Palindromic Repeats)gene targeting.Briefly,the gRNA targeting the Clec4f exon 2 and saCas9(staphylococcus aureus Cas9)mRNA were injected into C57BL/6J zygotes,and injected zygotes were transferred into the oviduct of CD-1 foster females.The female F0 founder was born with bi-allelic frameshifting indel-mutations of Clec4f.Direct sequencing of PCR products using tail-derived genomic DNA was used to validate the indel-mutations of F0 Clec4f founder.The F1 Clec4f-/-mice were obtained by crossing the female F0 mice with C57BL/6J mice.2.The genotypes of F1 generation were identified by PCR(polymerase chain reaction).The expression of Clec4fmRNA and protein in the liver of wild-type(WT)and Clec4f-/-mice were analyzed by qRT-PCR(quantitative real-time PCR)and western blot.Immunofluorescent staining of formalin-fixed paraffin-embedded tissue sections were used to analyze the distribution of CLEC4F in the liver and spleen of WT and Clec4f-/-mice.The body weight and peripheral blood count of WT and Clec4f-/-mice were compared.Results1.Direct sequencing of PCR products using tail-derived genomic DNA identified bi-allelic frameshifting indel-mutations of Clec4f from a F0 founder.Two independent F1 lines with two different heritable indel-mutations,termed Clec4f-4 and Clec4f-6+1,were then generated by crossing the founder with a WT C57BL/6J mouse.Further DNA sequencing results confirmed that Clec4f-4 mice exhibited a 4-bp(GCGG)frameshifting deletion,and Clec4f-6+1 mice had a 6-bp deletion with an insertion of a t(+1)in Clec4f exon 2.2.qRT-PCR and western blot analyses indicated that the homozygote deletions resulted in a loss of Clec4f mRNA transcripts as well as CLEC4F protein relative to that of WT littermate controls.Immunofluorescent staining indicated that CLEC4F was specifically colocalized with F4/80 positive KCs in WT but not Clec4f-/-mouse liver sections.Moreover,CLEC4F was not detectable in hepatocytes or the spleen of either WT or Clec4f-/-mice.The body weights and peripheral blood counts of both lines of Clec4f-/mice were comparable to that of WT littermates.ConclusionsWe have generated two independent Clec4f-/-lines.Clec4f-/-mice lacking Clec4f mRNA transcripts as well as CLEC4F protein relative to that of WT littermates.CLEC4F is specifically expressed on KCs in the liver but not in other tissues.The body weights and number of peripheral platelets in Clec4f-/-mice were comparable to that of WT littermates,indicating that CLEC4F does not alter platelet homeostasis under normal physiological conditions.There is no significant differences in phenotypes between Clec4f-4 and Clec4f-6+1 knockout mouse lines.Therefore,the Clec4f-4 line was primarily used in this study,while Clec4f-6+1 was used in some key experiments for validation.3.CLEC4F is the primary receptor mediating phagocytosis of desialylated platelets by Kupffer cells in the liverObjectivesThere is a considerable confusion regarding receptor(s)that are important for the clearance of desialylated platelets in the mouse liver.For example,the asialoglycoprotein receptor(ASGPR or Ashwell Morell receptor,AMR)has been reported to mediate the phagocytosis of desialylated platelets by hepatocytes;the macrophage galactose lectin(MGL)on KCs is also considered important for the clearance of desialylated platelets.Our own previous study supports an important role for CLEC4F in this regard.In this study,we aimed to determine whether KC CLEC4F is a major receptor mediating the phagocytosis of desialylated platelets in the liver.Methods1.To determine the survival rate of desialylated platelets and the distribution of desialylated platelets in the liver of WT and Clec4f-/-mice,washed platelets were firstly treated with neuraminidase.Flow cytometry analysis was used to detect efficiency of desialylation based on the binding of platelets to RCA I-FITC and biotinylated MAL Ⅱ.CellTracker Deep Red-labeled desialylated platelets were then transfused into either WT or Clec4f-/-recipients through intravenous injections.The percentage of labeled platelets was analyzed by flow cytometry.Immunofluorescent staining was used to analyze the distribution of labeled platelets in the liver and spleen of WT and Clec4f-/-mice.2.For in vivo desialylation,mice were intravenously injected with different types of bacteria-derived neuraminidases.The percentage of desialylated platelets was analyzed by flow cytometry.Confocal microscopy imaging of immunofluorescently stained tissue sections were used to analyze the distribution of desialylated platelets in the liver and spleen of these mice.3.To provide in vitro evidence that the CLEC4F-mediated interaction of desialylated platelets with KCs leads to phagocytosis,EGFP-positive platelets were firstly isolated from ROSAmTmG;Pf4Cre mice and then transfused into either WT or Clec4f-/-recipients.Recipient mice were injected with neuraminidase.Immunofluorescent staining was used to analyze the association of desialylated platelets with KCs in the liver.4.Transmission electron microscopy(TEM)was used to confirm that Kupffer cells are the primary cells phagocytizing desialylated platelets.5.WT and mice either genetically lacking or with antibody-mediated functional blocking of MGL,AMR,and/or CLEC4F were directly compared for their roles in mediating the clearance of desialylated platelets.Results1.Flow cytometry analysis indicated that the binding of MALII(specific for α-2,3-sialic acids)was decreased on platelets from neuraminidase-treated mice compared with the non-neuraminidase-treated group.Complementarily,RCA 1(specific for terminal β-gal without sialic acids)binding was increased on platelets treated with neuraminidase compared with non-neuraminidase-treated platelets.The survival rate analysis indicated that fewer transfused platelets detected in the WT peripheral blood compared with Clec4f-/mice at time points analyzed.Confocal microscopy analysis of immunostained cryosections from the liver showed that the transfused desialylated platelets were primarily associated with KCs in the liver of WT recipients,but not in Clec4f-/-recipients.In contrast,normal platelets were rarely found to be associated with KCs of either WT or Clec4f-/-recipients.2.In vivo desialylation assay showed that a marked increase in the percentage of platelet RCA1 binding(unmasked galactose)of both WT and Clec4f-/-mice within the initial 30 min after injection of neuraminidase,which was continuously increased at 24 h post-neuraminidase administration.WT mice exhibited a faster clearance rate of endogenous platelets than Clec4f-/-mice,with a major reduction of platelets within the initial 30 min following the administration of the neuraminidase.Confocal imaging analysis of liver sections from WT and Clec4f-/-mice showed a significant increase in numbers of platelets associated with KCs in the WT than those in the Clec4f-/-mouse liver.No obvious association of platelets with hepatocytes was detected in the WT and Clec4f-/mouse liver.Further Z-stack high-resolution confocal imaging and three-dimensional reconstruction analysis showed that the adherent platelets were mainly phagocytosed by KCs rather than hepatocytes.3.Single segment high resolution confocal imaging and rotating 3D imaging showed EGFP-positive desialylated platelets were primarily present inside of the primary KCs freshly isolated from WT mice following neuraminidase treatment,but not in Clec4f-/mice.4.TEM micrographs indicated that WT,but not Clec4f-/-,KCs phagocytosed the desialylated platelets.5.Comparison of WT and mice either genetically lacking or with antibody-mediated functional blocking of MGL,AMR,and/or CLEC4F demonstrated that CLEC4F was the key receptor in mediating the clearance of desialylated platelets by using different types of bacteria-derived neuraminidases in vivo and in vitro.ConclusionOur results indicate that the ability of KCs from Clec4f-/-mice to phagocytize desialylated platelets was significantly lower than that of WT mice.KC CLEC4F is a major receptor for mediating the phagocytosis of desialylated platelets in the liver of mice.