Alterations in sialic-acid O-acetylation glycoforms during murine erythrocyte development

Vinay S. Mahajan; Faisal Alsufyani; Hamid Mattoo; Ian Rosenberg; Shiv Pillai

Given that the Casd1-deficient mice exhibited a complete loss of TER-119 staining in the erythroid lineage, we considered the possibility that the TER-119 monoclonal antibody recognized a 9-O-acetyl sialic acid dependent epitope. However, the Ery-A and Ery-B erythroid precursors that normally express the TER-119 epitope in wild-type mice do not express any 9-O-acetyl sialic acid that is detectable by CHE-FcD staining. Therefore, we decided to examine the relationship between 9-O-acetyl sialic acid and the TER-119 epitope in greater detail. The hemagglutinin-esterase from bovine coronavirus Mebus strain (BHE) also recognizes 9-O-acetyl sialic acid, but in a manner that is structurally distinct from CHE (Zeng et al., 2008) . Furthermore, unlike CHE, the BHE lectin recognizes 7,9-di-O-acetyl sialic acid in addition to 9-O-acetyl sialic acid (Langereis et al., 2015; Bakkers et al., 2016) . Therefore, we expressed BHE as an Fcfusion protein (BHE-Fc) analogous to the approach used to generate CHE-Fc. In addition, we also generated a catalytically inactive form of BHE-Fc bearing the S40A substitution (BHE-Fc-S40A) whose ability to bind 9-O-acetyl sialic acids is unhindered as previously described (Zeng et al., 2008) . We found that the TER-119 + CD71 + Ery-A and Ery-B precursors from wild-type mice exhibited strong staining with BHE-Fc-S40A but not with CHE-FcD (Fig 3B and 4A) . This suggested to us that the TER-119 epitope may be recognized by the BHE virolectin but not by CHE. We next used murine RBCs to examine the interactions of the TER-119 epitope with CHE and BHE. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/469254 doi: bioRxiv preprint 4B). Yet, the TER-119 epitope on murine erythrocytes was surprisingly resistant to treatment with CHE ( Fig 4B) . As a control, and as previously shown by Varki et al., we also showed that CHE-Fc pre-treatment completely eliminated CHE-FcD binding on CD4 + T lymphocytes (Fig 2D) . Interestingly, the TER-119 epitope on murine RBCs was sensitive to BHE-Fc but not CHE-Fc esterase treatment (Fig 4B) . In contrast, pretreatment with an enzymatically inactive version of BHE-Fc bearing the S40A catalytic site mutation (BHE-S40A-Fc) had a negligible effect on TER-119 staining (data not shown).