Fig 2: High frequency of Tfh cells in HBV-ACLF patients was associated with disease severity. (A) The frequencies of CD4+CXCR5+, CD4+CXCR5+ICOS+ and CD4+CXCR5+IL-21+ Tfh cells in the PBMCs from HBV-ACLF (n = 36), M-CHB (n = 21), S-CHB (n = 32) patients and HC (n = 10) subjects were demonstrated by flow cytometry. (B) The frequencies of CD4+CXCR5+, CD4+CXCR5+ICOS+ and CD4+CXCR5+IL-21+ Tfh cells in the PBMCs from HBV-ACLF, M-CHB, S-CHB and HC subjects were analyzed using Mann-Whitney U test. (C) The correlation between frequency of CD4+CXCR5+ICOS+ Tfh cells and MELD score was analyzed using Spearman correlation analysis. (D) The frequency of CD4+CXCR5+ICOS+ Tfh cells from ameliorated (n = 7) and non-ameliorated patients (n = 29) were analyzed by flow cytometry. (E) The frequencies of Tfh cells from ameliorated patients (n = 7) over 8- and 12-week treatment were analyzed by flow cytometry. Interclass comparison was made using Wilcoxon’s signed-rank test. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig 3: PLK1S1 knockdown prevents tumor growth in vivo. (A) Images of the tumors were obtained, and (B) tumor volumes and (C) the weight of the tumors in mice injected with either shNC or shPLK1S1 were determined. n=5. (D) PLK1S1 and (E) miR-653 mRNA expression levels were determined using reverse transcription-quantitative PCR. (F) The protein expression level of CXCR5 was examined in tumors using western blot assay, with GAPDH as the loading control. All data are represented as the mean ± SD. *P<0.05; **P<0.01; ***P<0.001. miRNA, microRNA; sh, short hairpin; NC, negative control; CXCR5, C-X-C motif chemokine receptors 5.

Fig 4: CD4+CXCR5+ICOS+ Tfh cells were induced by cytokines or serum of HBV-ACLF patients. (A, D) The naïve CD4+ T cells of HC subjects (n = 6) were stimulated by the RPMI complete medium as control (CTR), dynabeads® human T-activator CD3/CD28, IL-12 (10 ng/ml), IL-21 (20 ng/ml), IL-12+IL-21 and IL-17 (10 ng/ml) for 72 hours in vitro, and the frequencies of CD4+CXCR5+ICOS+ Tfh cells were demonstrated by flow cytometry, respectively. (B, E) The naïve CD4+ T cells of HC subjects were stimulated by CTR, dynabeads® human T-activator CD3/CD28, serum from HC subjects (n = 6), and serum from HBV-ACLF patients for 72 hours in vitro, and the frequencies of CD4+CXCR5+ICOS+ Tfh cells were demonstrated by flow cytometry, respectively. (C, F) The naïve CD4+ T cells of HC subjects were stimulated by CTR, dynabeads® human T-activator CD3/CD28, serum from HBV-ACLF patients (1:8) with or without IL-12 (1 µg/ml), IL-21 (1 µg/ml), IL-12/21 and IL-17(1 µg/ml) antibody (the antibody experiments were performed in the presence of 1:8 HBV-ACLF serum), for 72 hours in vitro, and the frequencies of CD4+CXCR5+ICOS+ Tfh cells were demonstrated by flow cytometry, respectively. Representative data of independent experiments are shown as median (range). *p < 0.05, **p < 0.01.

Fig 5: The frequency of Tfh cells was enhanced during the development of DR in STZ-induced mice. Representative images of immunofluorescence staining in cervical lateral lymph node cryosection (A), retinal flat-mount (B), and cryosection (C) showed wide distribution of PD-1+CXCR5+CD4+ Tfh cells in DR mouse, whereas lack of PD-1 and CXCR5 expressions on CD4+ T cells in LN (A) and few PD-1+CXCR5+CD4+ Tfh were observed in the retina from non-diabetic mice (B-C). GCL: ganglion cell layer; IPL: inner plexiform layer; INL: Inner nuclear layer; OPL: outer plexiform layer; ONL: outer nuclear layer; Scale bar, 50 µm.