Lai, Jinping

Transforming growth factor-β (TGF-β) activates hepatic stellate cells (HSCs), which drive liver fibrosis via the production and deposition of extracellular matrix (ECM). We aimed to elucidate the mechanistic role of sulfatase-2 (SULF2) in liver fibrosis. To this end, we induced liver fibrosis in wild-type (WT) and SULF2 knockout (Sulf2-KO) mice (6–8 weeks-old) via bile duct ligation (BDL), intraperitoneal injection of carbon tetrachloride (CCl4) or thioacetamide (TAA). The levels of fibrosis in the liver sections were assessed via Sirius red and Masson’s trichrome staining, immunohistochemistry and immunoblotting for α-smooth muscle actin (α-SMA) and hydroxyproline. To evaluate the interaction between TGF-β and SULF2, we transfected human HSCs with scrambled control shRNA and shRNA constructs targeting SULF2 and measured α-SMA expression following treatment with TGF-β1 ligand. We show here that knockout of SULF2 significantly decreases collagen content, as well as bands of bridging fibrosis, as demonstrated by Sirius red, Masson’s trichrome and α-SMA staining after BDL, CCl4 and TAA injection in Sulf2-KO versus WT mice. In all three models of liver fibrosis, we observed significantly lower levels of hydroxyproline in the Sulf2-KO mice compared to the WT mice. HSCs with reduced levels of SULF2 failed to significantly express α-SMA and collagen type I following treatment with TGF-β1. Furthermore, SULF2 co-localizes with TGFBR3 and the in vitro knockdown of SULF2 in HSCs decreases the release of TGF-β1 from TGFBR3. Together, these data suggest that SULF2 regulates liver fibrosis via the TGF-β signaling pathway. Pharmacologic inhibition of SULF2 may represent a novel therapeutic approach to improve liver fibrosis.

Background: The human sulfatase 1 (SULF1) and sulfatase 2 (SULF2) genes modulate cell signaling and homeostasis in many tissues. Gene expression analyses have implicated SULF2 in disease pathogenesis, including Alzheimer’s disease (AD), but changes in brain SULF2 expression have not been directly established. Objective: To investigate the expression of SULF1 and SULF2 in brain tissues from AD cases and cognitively normal controls. Methods: Autopsy tissue from AD cases ( n = 20) and age-and gender-matched cognitively normal controls ( n = 20) were identified from the Mayo Clinic Alzheimer’s Disease Patient Registry neuropathology database. Tissue slides were stained for SULF1 and SULF2 protein expression in the hippocampus and frontal lobe and an expression score computed from the proportion of cells stained and the intensity of staining (range 0 [no expression] to 9 [marked expression]). Results: SULF2 expression was reduced in AD cases. Compared to cognitively normal controls, SULF2 expression in AD cases was significantly decreased in the hippocampal Cornu Ammonis (CA) (mean score of 6.5 in cases versus 8.3 in controls; p = 0.003), in the gray matter of the parahippocampal gyrus (5.6 in cases versus 7.6 in controls; p = 0.003), and in the frontal lobe gray matter (5.4 in cases versus 7.4 in controls; p = 0.002). There was no difference in SULF1 expression in the hippocampus or frontal lobe of AD cases and controls. As expected there were no differences in SULF1 or SULF2 expression in white matter in AD cases compared to cognitively normal controls. Conclusion: Decreased SULF2 in specific regions of the brain occurs in AD.

Abstract Introduction: Liver metastasis of pancreatic or colorectal tumors is a major cause of cancer-related death worldwide. When cancer cells reach the liver parenchyma and make a direct contact with the surrounding extracellular matrix (ECM), β1-integrin-FAK signaling is activated allowing the cells attach, migration, and proliferate. Thus, the β1-integrin-FAK signaling axis is key for cancer cells to colonize the liver. However, mechanisms governing β1-integrin activation and how this pathway promotes liver metastasis remain incompletely understood. Aim: We sought to test the hypothesis that vasodilator-stimulated phosphoprotein (VASP), an actin binding protein, regulates the β1-integrin-FAK signaling axis, thereby, promoting liver metastasis of GI cancer. Methods: We employed lentiviruses encoding different VASP shRNAs to knockdown (KD) VASP in human GI cancer cells. We used a Matrigel-based 3-dimensional (3D) cell culture model and intrasplenic tumor injection mouse model to assess the role of VASP for cancer cells in vitro and in vivo. Immunofluorescence (IF) was used to study localization of VASP and active β1-integrin (using Huts-4 antibody), and Western blot analysis (WB) was used to quantitate protein levels of active β1-integrin and YAP1/TAZ. Additionally, we performed VASP immunohistochemistry (IHC) to analyze VASP immunoreactivity in 116 colorectal or pancreatic adenocarcinomas of patients. Results: In the Matrigel-based 3D-culture model, a single cancer cell gave arise to a 3D-cancer spheroid after being cultured for 6-8 days. Using KM12L4, HCT116, or HT29 colorectal cancer cells, and L3.6 or MIA PaCa-1 pancreatic cancer cells, we found that VASP KD reduced the sizes of the 3D-cancer spheroids grown on Matrigel (P<0.05 by t-test for each cell line). IF and WB show that VASP KD reduced levels of Huts-4 (active β1-integrin) in KM12L4, HCT116 colorectal cancer cells and in L3.6, MIA PaCa-1 pancreatic cancer cells. VASP KD also reduced the protein levels of P-FAK and YAP1/TAZ in the 4 cell lines tested. Additionally, YAP1/TAZ protein levels were reduced by PF-228, a FAK inhibitor that blocks the β1-integrin-FAK signaling. In the intrasplenic tumor implanation mouse model, VASP KD suppressed liver metastasis of GI cancer as well as reduced Huts-4 and YAP1 signals of their liver metastases in mice (P<0.05 by t-test). IHC of patient's samples reveals that VASP protein levels were upregulated in cancer cells in 81% of colorectal tumors (51/63) and pancreatic tumors (43/53), with higher VASP levels being associated with liver metastasis of colorectal cancer (P<0.01 by Fisher exact test). Conclusions: VASP promotes ECM-mediated β1-integrin-FAK-YAP1/TAZ signaling axis of GI cancer. We propose that VASP and the β1-integrin-FAK-YAP1/TAZ signaling axis represent potential new therapeutic targets for liver metastasis of pancreatic or colorectal cancer. Citation Format: Xiaoyu Xiang, Jinhua Piao, Selvaraj Muthusamy, Lei Wang, Yibin Deng, Shengbing Huang, Raul Urrutia, Jinping Lai, Ningling Kang. Vasodilator-stimulated phosphoprotein promotes β1-integrin-FAK-YAP1/TAZ signaling axis that is required for liver metastasis of GI cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1633.