Leukemia is a complex, heterogeneous disease characterized by various genetic abnormalities. Acute leukemias include acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL). Recent advances in understanding the molecular landscape of these diseases have led to the development of targeted therapies, particularly those targeting CD38, widely expressed in hematologic malignancies. CD38, a 45 kDa transmembrane glycoprotein, is significantly overexpressed in AML and T-cell acute lymphoblastic leukemia (T-ALL), playing a crucial role in immune modulation. Targeted therapies against CD38, including monoclonal and bispecific antibodies such as XmAb18968 (Xencor), have demonstrated efficacy with manageable toxicity. However, resistance can occur despite persistent CD38 expression, possibly due to posttranslational modification blocking epitope recognition. With four N-linked glycosylation sites, the impact of glycosylation on CD38's recognition and function in malignant and therapy-resistant cells is not well-understood. We hypothesize that variations in CD38 glycosylation, influenced by leukemia type and disease stage, affect treatment responses, especially with monoclonal antibodies.
Methods: We analyzed CD38 expression and glycosylation across eight human leukemia cell lines (HL-60, THP-1, RPMI-8402, KG1a, Kasumi-1, LAMA-84, K-562, MV4-11), representing various stages and types of the disease. We employed flow cytometry and immunoblotting analyses using four distinct commercial anti-CD38 antibodies. The impact of all-trans retinoic acid (ATRA) on CD38 expression and glycosylation profile was assessed. Glycosylation patterns of total membrane lysates were explored using lectin arrays, and glycan structures were determined using mass spectrometry. To investigate the role of N-glycosylation in CD38 antibody recognition, we treated leukemia cell lines with PNGase F to enzymatically remove N-linked glycans, rendering CD38 in its unglycosylated form. Results: i) CD38 Expression and ATRA Response: Surface CD38 expression was detected in all leukemia cell lines, displaying significant heterogeneity in expression levels and glycan profiles, as assessed by flow cytometry. Treatment with ATRA variably increased CD38 levels, notably in HL-60 and MV4-11 cells, which showed low baseline expression but significant induction post-treatment. Conversely, ATRA failed to upregulate CD38 in LAMA-84 and K-562 cell lines. ii) Glycosylation Changes: The LAMA-84 cell line exhibited significant glycosylation changes without a corresponding increase in CD38 expression. In contrast, the HL-60 cell line showed no substantial glycosylation alterations. Mass spectrometry analysis in the MV4-11 cell line confirmed an overall increase in N-glycosylation intensity post-ATRA treatment, shifting towards increased, less complex high-mannose structures, indicating that ATRA significantly alters the glycosylation landscape in a cell type-dependent manner. iii) Antibody Recognition and Glycosylation: Immunoblotting revealed differential CD38 antibody recognition across various leukemia cell lines, with notable glycoprotein detection mainly in KG1a, HL-60, THP-1, and RPMI-8402 cell lines. The anti-CD38 HB7 clone showed minimal recognition in immunoblots but increased detection by flow cytometry, suggesting that CD38 membrane topology and potentially its glycosylation patterns affect antibody accessibility. PNGase F treatment shifted the CD38 band from an apparent molecular weight of 45 kDa to the expected 35 kDa in all cell lines. After PNGase F treatment, CD38 recognition varied: it was enhanced in THP-1 but reduced in HL-60 cell lines, indicating that glycosylation differentially affects antibody recognition across leukemia types. Conclusion: This study highlights significant heterogeneity in CD38 expression and glycosylation across leukemia types and stages, impacting the binding and efficacy of CD38-targeted antibodies.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal