De novo glycosylation of MHC class I in platelets Free

Whether platelets can intracellularly de novo glycosylate proteins and how this affects platelet functions remains largely unknown. Historically, platelets have been considered deficient in protein synthesis and glycosylation due to their lack of a rough endoplasmic reticulum (ER) and a mature Golgi apparatus. However, we found that under disease conditions such as sepsis, stressed platelets (in both humans and mice) alter their functions and significantly increase the de novo synthesis of approximately 3,000 proteins—nearly half of which are glycoproteins. Notably, one of the most upregulated glycoproteins is major histocompatibility complex class I (MHC I).

We previously showed that platelets suppress CD8+ T cells in an MHC I dependent manner during sepsis. The synthesis, membranous expression, and antigen (Ag) presentation of MHC I is significantly increased. The MHC I heavy chain (MHC Ia) possesses a highly conserved Asn86 site for N-glycosylation, which is critical for several functions, including intracellular binding to chaperone proteins, trafficking, expression on the plasma membrane, and extracellular interactions with T cell receptors (TCRs). However, whether MHC I could be glycosylated in platelets remains unknown.

In our current study, we first investigated whether MHC I expression and function requires de novo glycosylation in platelets. Platelets were isolated from healthy donors and treated with tunicamycin (5 µg/mL), a specific inhibitor of N-glycosylation, for 16 hours at 37°C. Both total MHC I expression and plasma membrane expression were significantly decreased in the treated platelets, as measured by western immunoblotting and flow cytometry, respectively. That this was a defect in intracellular trafficking was suggested by confocal images that showed a punctate distribution of MHC I in the cytoplasm. This is likely due to the intracellular accumulation of nascently synthesized MHC I without proper glycosylation. Platelet viability following tunicamycin treatment was confirmed by negative annexin V staining. Tunicamycin also inhibited platelet-CD8+ T cell interactions in co-culture in an MHC I-dependent manner. To investigate the de novo glycosylation of MHC I in platelets, we carried out a Click-iT assay, in which an azide-labeled GlcNac analog (GlcNaz) is incorporated into newly synthesized glycans on proteins, which can then be detected by reaction with an alkyne. Platelets from healthy donors were treated with GlcNAz (100 μM) for 6 hours at 37°C. De novo N-glycosylated proteins incorporating GlcNAz were then labeled with alkyne-biotin and pulled down using streptavidin beads. Newly glycosylated MHC-I was successfully detected by immunoblotting. Mechanistically, we examined glycosyltransferases involved in platelets using our platelet RNA-Seq data and proteomic datasets publicly available. Platelets express all glycosyltransferases required for the synthesis of Glc₁Man₉GlcNAc₂, the dominant N-glycan structure on MHC Ia, which is critical for proper MHC I assembly. However, platelet-synthesized proteins may not undergo the full spectrum of N-glycan maturation steps typically observed in nucleated cells, as we found different molecular weight of MHC I in platelets as compared to leukocytes from the same donors by western blot using w6/32 antibody. This incomplete processing of MHC I glycans may result in a platelet-specific MHC I:Ag peptide complex that accounts for the subsequent CD8⁺ T cell suppression during sepsis. We next investigated whether the glycan structure of MHC I changes in platelets during sepsis. Platelets were then treated with lipopolysaccharide (LPS, 50 μg/mL) or polyinosinic:polycytidylic acid (poly I:C, 10 μg/mL) to mimic bacterial or viral-mediated sepsis conditions, respectively. Expression of MHC I was increased, with slightly shifted molecular weight by western blot. The molecular weight shift was inhibited in the presence of tunicamycin, suggesting it was due to altered N-glycosylation. We are currently conducting N-glycoproteomic analyses to characterize the glycan structures of platelet-derived MHC I.

In summary, we found that platelets can actively glycosylate newly synthesized proteins, including MHC I, although their capacity to generate fully mature glycans is limited, leading to unique structures that may affect protein function including antigen presentation.