Fig 1.
SOX9 is expressed in areas of HSC activation and fibrosis deposition in S. mansoni infection.
(A) Tissue sections from naïve and S. mansoni livers stained for SOX9 (brown), α-SMA (magenta), and PSR (red, counterstained with fast green). In naïve mice SOX9 is detected in cholangiocytes (open arrowhead) and occasional hepatocytes (black arrowheads). Α-SMA and PSR expression localise to vascular regions. In infected mice SOX9 expression is upregulated and localises in part to areas of α-SMA and PSR staining as progressive fibrosis develops in the granuloma (dotted black line). All scale bars 50µm. bd = bile duct, a = artery, pv = portal vein, * = granuloma egg. (B) SOX9 IHC quantification, (C). α-SMA IHC quantification and (D) PSR IHC quantification throughout the duration of S. mansoni infection. Data obtained by using 10 images per X20 field. 1 lobe per animal and n = 3 animals per time point. Statistical significance determined by one-way ANOVA and Dunnett’s multiple comparisons test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).
Fig 2.
SOX9 is expressed in three distinct cell types of the liver in S. mansoni infection.
(A) Naïve and S. mansoni infected livers underwent dual IHC for SOX9 (magenta) and CK19 (red) with toluidine blue as counterstain. SOX9 localises to CK19+ cholangiocytes in the naïve liver. In S. mansoni infection, SOX9 is detected in CK19+ ductular reaction cells. All scale bars 50µm. (B) Dual IF for SOX9 (red) and α-SMA (green) in naïve and S. mansoni infected livers, with DAPI (blue) as nuclear counterstain. SOX9 is expressed in nuclei of α-SMA+ cells around the schistosome egg (white asterisk), reflecting upregulation of SOX9 in activated HSCs. All scale bars 50µm. Bottom row is a zoom of the region demarcated by the dotted yellow box. (C) Dual IF for SOX9 (red) and HNF4α (green) in naïve and S. mansoni infected livers, with DAPI (blue) as nuclear counterstain. Dual expression of SOX9 and HNF4α+ hepatocytes are seen at margins of granulomas (white arrows). All scale bars 50µm. Bottom row is a zoom of the region demarcated by the dotted yellow box.
Fig 3.
Failure to generate robust granuloma results in interstitial fibrosis in SOX9KO.
(A) Spleen weight and (B) Liver weight as a proportion of body weight. (C) Area of granuloma as a proportion of total liver area. (D) Representative H&E, PSR and aSMA IHC of naïve tissue, granulomas of infected mice and interstitial tissue between granulomas. SOX9KO granulomas are smaller and less well defined. SOX9KO mice have increased interstitial fibrosis. (E) Significantly reduced COL1 protein in infected SOX9KO than WT controls at the tissue level as detected by WB. Statistical analysis using (A, B, E) Two-way ANOVA with Tukey’s multiple comparisons and (C) unpaired Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. n = 5–7 animals per group.
Fig 4.
Sox9 depletion reconfigures hepatic immune profiles during S. mansoni infection.
The frequency of indicated myeloid cell populations (A) total CD4+ T cells and (B) their expression of ST2, CD25 and RORγt in the livers of naïve and infected, Cre-ve or RosaCre Sox9 KO mice. (C) Representative flow plots for ST2+ gating, pre-gating on liver CD45+TCRβ+CD4+ cells. (D) Cytokine secretion in PMA-stimulated hepatic CD4+ T cells. Data presented as mean + /- SEM. Data shown are from 2 pooled experiments. N = 1-4 per experimental group per experiment. Significant differences were determined by one-way ANOVA with Tukey’s post-hoc testing. * p < 0.05, ** p < 0.01, *** p < 0.001.
Fig 5.
Aberrant spatial organisation of immune cells in SOX9 deficient animals.
(A) F4/80 IHC staining of infected control and SOX9KO animals and quantification of stained area as a proportion of tissue area (n = 5–7 animals per group). Significant differences were determined by one-way ANOVA with Tukey’s post-hoc testing. *** = p < 0.001. (B) Representative mass cytometry images of liver from infected control and SOX9KO animals. Staining targets are identified by the colour key.