Fig. 6

Desialylation of MDSCs downregulates MDSC functional markers and cytokines at the transcript level. Suppressive MDSC-like cells were generated in vitro by coculture with A549 or A549-sia cancer cell lines as described in Fig. 5. CD33⁺ cells were isolated on Day 7 and processed for single-cell RNA sequencing (scRNAseq). A Seurat analysis of the scRNAseq dataset projected in UMAP colored by cluster. n = 4 donors per treatment group. B The dataset was subdivided into individual groups showing MDSC-like cells generated by A549 (green, left) or A549-sia (pink, right) coculture. C Stacked bar plots showing the frequency of each cluster annotated in (A) subclustered in A549- and A549-sia-generated MDSC-like cells. D Gene Ontology (GO) enrichment analysis of the top 10 upregulated gene sets found in Cluster 2. Dot plot showing the mean normalized enrichment score (NES) of the GO gene sets. The color coding indicates the adjusted p values, and the dot size is proportional to the gene count found in the listed pathway. E Heatmap of selected genes per patient divided into A549- and A549-sia-generated MDSC-like cells. The genes were functionally categorized into 5 groups: (i) chemokines and chemotaxis genes; (ii) MDSC and macrophage marker genes; (iii) protumor function MDSC genes; (iv) other genes; and (v) adhesion, attachment and ECM-related genes. F Gene expression of selected markers (IL1B, S100A9, S100A8, MKi67 and CCL2) as a density plot. The expression density is shown as a scale from blue (low) to yellow (high). G Expression of CCL2 was normalized to that of the donor.The data are presented as the mean ± SD. A two-tailed paired t test (F, E) was used. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001