The study compared surface sialylation on bone marrow mesenchymal stromal cells (BM‑MSCs) from healthy donors, patients with multiple myeloma (MM), treated MM (T‑MM) and MGUS. Using lectins SNA and MAL‑II in flow cytometry, the authors report that MM‑MSCs display significantly higher surface sialylation than HD‑, T‑MM‑ and MGUS‑MSCs. T‑MM and MGUS MSCs showed lectin binding levels that were statistically lower than MM‑MSCs and not significantly different from healthy donor MSCs, indicating that hypersialylation is a distinguishing feature of untreated MM‑MSC samples as measured in this dataset.
MM‑MSC sialylation levels were positively correlated with patient clinical variables, including the percentage of plasma cells and LDH levels, as reported by the authors. To interrogate mechanisms driving increased MSC sialylation, MSCs were cultured under several conditions: culture medium alone, conditioned medium from myeloma cell lines, transwell coculture, and direct coculture with myeloma cells. Direct MSC–myeloma cell contact produced a greater increase in MSC surface sialylation than conditioned medium or transwell culture, supporting a contact‑dependent enhancement of MSC hypersialylation in vitro.
An exploratory qPCR array targeting 84 glycosylation‑related genes was performed on MSC samples from healthy donors, MGUS and MM. The screening identified dysregulation in genes associated with the sialylation pathway, notably ST6Gal1, a sialyltransferase, and NEU2, a sialidase. The study reports these genes as implicated in the observed hypersialylation phenotype of MM‑MSCs, with further validation of their expression by qPCR described in the methods.
The authors evaluated how microenvironmental cues influence MSC glycosylation. Exposure to a pro‑inflammatory cytokine cocktail and culture under hypoxic conditions were associated with changes in MSC expression of Siglec ligands (Siglec‑7L, Siglec‑9L) and ST6Gal1, implicating inflammation and hypoxia as modulators of the sialylation machinery. MSCs were also cultured as 3D spheroids to model non‑adherent growth, and sialylation modulators (neuraminidase, P‑3FAX‑Neu5Ac inhibitor, and Ac4ManNAz activator) were used to manipulate MSC surface sialylation prior to functional assays.
Bone marrow aspirates were obtained with informed consent from 103 MM patients, 21 MGUS patients and 60 healthy donors. MSCs were isolated by standard adhesion methods and characterized per supplementary protocols. MSCs were cultured with RPMI 8226 or U266B1 myeloma cell lines in various coculture formats. Surface sialic acids were detected by SNA and MAL‑II lectins; Siglec ligands were labeled using recombinant Siglec‑7 and Siglec‑9‑Fc chimeras. Peripheral blood monocytes were isolated by CD14 selection and differentiated into macrophages using GM‑CSF. Macrophage polarization markers were assessed by flow cytometry: M1 associated markers CD40 and HLA‑DR were quantified as mean fluorescence intensity ratios (MFIRs), while M2 marker CD206 was reported as the percentage of positive cells. Cytokine levels (TNF‑α, IL‑10, TGF‑β1, CCL22) in co‑culture supernatants were measured by ELISA. Sialylation was modulated pharmacologically using neuraminidase, P‑3FAX‑Neu5Ac (inhibitor) and Ac4ManNAz (activator) with MSC pre‑treatment prior to macrophage coculture.
Functionally, MSCs with enhanced sialylation promoted macrophage polarization toward an M2‑like phenotype, evidenced by increased CD206 expression and elevated levels of M2‑associated cytokines in co‑culture supernatants. Conversely, inhibition of MSC sialylation shifted macrophages toward an M1‑like phenotype, with higher CD40 and HLA‑DR expression and increased pro‑inflammatory cytokine production (TNF‑α reported by ELISA). When MSC sialylation was reduced pharmacologically, their ability to induce an M2 phenotype in macrophages decreased, as reflected by lower M2 cytokine secretion.
The authors conclude that BM‑MSC hypersialylation in MM contributes to immune remodeling by promoting Siglec‑mediated macrophage polarization toward an immunosuppressive, tumor‑supportive M2 state. Identification of dysregulated sialylation enzymes such as ST6Gal1 and NEU2, along with the influence of inflammation and hypoxia, suggests stromal sialylation is a modifiable component of the tumor microenvironment. The data support the concept that targeting the stromal sialylation machinery could be explored as a strategy to limit immunosuppression and restore anti‑tumor immunity in MM. Details on specific effect sizes, statistical values and supplementary data are provided in the original article and its supplementary materials.