Fig 2: Correlation between miR-22 expression, galectin-1 expression, CD3 expression and the clinicopathological features of HCCA. IHC analysis to measure expression of α-SMA and galectin-1 in consecutive sections of HCC samples (n=53). B-D. Galectin-1 expression in different TNM stages (TNM I-IV: A1-A4.) and CD3 expression in different TNM stages (TNM I-IV: B1-B4.) measured by IHC. The correlations between galectin-1 expression, CD3 expression, and TNM stage of patients with HCC were analysed using the χ2 test (TNM I, n = 13; TNM II, n = 55; TNM III, n = 58; TNM IV, n = 36). C-E. Expression of C1-C2. galectin-1 and D1-D2. CD3 in Ca-HSCs isolated from HCC patient liver samples, separated into low (n = 13) and high (n = 13) miR-22 expression groups, according to the median miR-22 expression. Magnification: ×200 for CD3 staining (B1-B4, B1-B2), ×100 for galectin-1 staining (a1-a4, c1, c2). *P < 0.05.

Fig 3: (A) The IHC assay of Gal-1 and GRP78 expression in GC tissues and corresponding adjacent normal tissues was performed, and the representative images were shown. (B) Qualification of GRP78 staining in TMAs described in panel A. The graph depicts the total score, the multiplication product of staining intensity and the percentage of stained cells. (C) Statistical analysis of the relationship between the expression level of GRP78 and Gal-1 (P-value: Spearman's correlation coefficient). (D) Kaplan-Meier overall survival curves for all 80 patients with GC stratified by high and low expression of GRP78. **P<0.01. Gal-1, galectin-1; GRP78, glucose-regulated protein 78; IHC, immunohistochemical; GC, gastric cancer.

Fig 4: Role of miR-22 in HSC-derived T cell apoptosis and Th cytokine balance skewingT cell apoptosis and cytokine (IFN-γ and IL-10) levels in the supernatant of 4 groups of HSCs isolated from normal liver samples and co-cultured with CD3+ T cells for 48 hours at a ratio of 1:10 (HSC:T) (n = 6 for each group), measured by A. flow cytometry and B. Elisa. Each of the 4 groups was subjected to different pre-treatments: transfection with 1) negative control mimic for miR-22 (mimic NC), 2) miRNA-22 of homo sapiens type mimic (hsa-miRNA-22 mimic), 3) hsa-miRNA-22 mimic plus galectin-1 overexpression plasmid (hsa-miRNA-22 mimic + Over), and 4) hsa-miRNA-22 mimic plus negative control for galectin-1 overexpression (hsa-miRNA-22 mimic + pcDNA3.1) Data are shown as the means (± SD) of triplicates. N.S. for P > 0.05; *P < 0.05.

Fig 5: The expression of galectin-1 in HSCs promotes HSC-induced T cell apoptosis and Th1/Th2 cytokine balance skewingA. Flow cytometry (annexin V-FITC apoptosis detection) analyses to detect T cell apoptosis in CD3+ T cells, cultured alone or co-cultured with HSCs subjected to different pre-treatments (cell transfection for galectin-1 knockdown and overexpression: sh-3 group versus Scr group; Over group versus pcDNA3.1 group) for 48 hours at a ratio of 10:1 (T:HSC), B. ELISA showing the levels of cytokines (IFN-γ and IL-10) in the supernatant. Data are shown as the means (± SD) of triplicates (n = 7). *P < 0.05. NC, negative control group; Scr, non-targeting scrambled sequence group; sh, small hairpin RNA sequence transfection group; pcDNA3.1, negative control group; Over, galectin-1 overexpression group; No HSCs, T cells cultured alone.