Role of connective tissue growth factor in oval cell response during liver regeneration after 2-AAF/PHx in rats

Abstract

Background & aims: Recruitment and proliferation of Thy-1+ oval cells is a hallmark of liver regeneration after 2-acetylaminofluorene (2-AAF)/partial hepatectomy (PHx) in rats. To understand the molecular mechanism underlying this process, we investigated the role of connective tissue growth factor (CTGF), one of the candidate genes differentially expressed in Thy-1+ oval cells, in this liver injury model.

Methods: Northern and Western analyses were performed to examine the induction of CTGF in total liver homogenate. Quantitative real-time polymerase chain reaction (PCR), immunofluorescent staining, and in situ hybridization were performed to confirm the expression and localization of CTGF in Thy-1+ oval cells. Finally, a known inhibitor of CTGF synthesis, Iloprost, was administered to 2-AAF/PHx treated rats to investigate the effect of Iloprost on oval cell response.

Results: CTGF was found to be up-regulated at both the RNA and protein levels and occurred concurrently with an up-regulation of transforming growth factor beta1 (TGF-beta1). Sorted Thy-1+ oval cells expressed a high level of CTGF gene in a quantitative PCR assay. Colocalization of Thy-1 antigen and ctgf signals by in situ hybridization further confirmed that Thy-1+ oval cells were a source of CTGF. Iloprost administration blocked CTGF induction in treated animals but did not affect TGF-beta1 expression. The inhibition of CTGF induction by Iloprost was associated with a significant decrease in oval cell proliferation and a lower level of alpha-fetoprotein expression as compared with control animals.

Conclusions: These results show that CTGF induction is important for robust oval cell response after 2-AAF/PHx treatment in rats.

Figure 1

Massive proliferation of Thy-1⁺ oval cells in portal region during liver regeneration after 2-AAF/PHx. Frozen liver sections were examined by immunohistochemistry for Thy-1 antigen. Blue color indicates DAPI stained DNA, whereas red indicates Texas red signal for Thy-1 antigen. (A) Normal. (B) 2-AAF/PHx treated liver (day 9). Both images were taken at 200×. Scale bar: 10 μm.

Figure 2

Induction of CTGF expression in the 2-AAF/PHx model. (A) Northern blot analysis of whole liver homogenate shows induction of CTGF message beginning at day 5 and reaching a shoulder peak around day 7 to 9 post PHx. N represents normal liver. (B) Western blot analysis of whole liver homogenate shows accumulation of CTGF protein. Actin was used as a loading control.

Figure 3

Thy-1⁺ oval cells express CTGF transcript. (A) Real-time PCR shows induction of CTGF in both whole rat liver and Thy-1 sorted oval cells after 2-AAF/PHx compared with those of normal livers. N indicates normal livers. Values in normal livers are arbitrarily assigned to 1 unit. Data represent the mean value ± SD (n ≥ 3). *P < .05. (B) In situ hybridization negative control probed with sense CTGF (200×). Thy-1⁺ oval cells obtained at day 9 after 2-AAF/PHx treatment shown in (C) and (E) were stained in red. In situ hybridization of the same section with CTGF antisense probe in (D) and (F) shows dark brown signals for CTGF transcripts. Blue arrows indicate the colocalization of CTGF transcripts in Thy-1⁺ oval cells. Images of B, C, and D were taken at 200×. E and F were taken at 400×. E and F were the large magnifications of C and D, respectively.

Figure 4

Coordinated expression of TGF-β1, pro-collagen type I, and fibronectin in oval cell–aided liver regeneration. Northern blot analysis shows TGF-β1 and ECM proteins are induced simultaneously with CTGF after 2-AAF/PHx treatment in rats. GADPH was used as a loading control.

Figure 5

Iloprost blocks up-regulation of CTGF but does not affect TGF-β1 induction after 2-AAF/PHx treatment. Northern blot analysis showed that CTGF up-regulation is blocked, whereas TGF-β1 expression remains unaffected by Iloprost in 2-AAF/PHx-treated rats. GADPH was used as a loading control.

Figure 6

Decrease in the number of histologically evident oval cells in Iloprost-treated rat liver after 2-AAF/PHx. (A) Portal regions of H&E-stained liver sections from 2-AAF/PHx-treated rats contain an abundance of cells with high nuclear to cytoplasm ratio (40×). (B) Far fewer of these cells are seen in sections from rats that received Iloprost (40×). (C) Magnification of the area outlined in A (400×). (D) Magnification of the area outlined in B (400×).

Figure 7

Decrease in the number of proliferating cells in Iloprost-treated rat livers after 2-AAF/PHx. (A) Portal regions of BrdU-stained liver sections from day 7; 2-AAF/PHx-treated rats contain a high number of proliferating cells (40×). (B) A dramatic decrease in proliferation is seen in sections from rats that received Iloprost (40×). (C) Magnification of the area outlined in A (400×). (D) Magnification of the area outlined in B (400×). (E) Quantitation of BrdU-positive cells in periportal regions at day 5, 7, and 9 after 2-AAF/PHx in Iloprost-treated (black bars) as compared with control (open bars) rats. The number of BrdU-positive cells was counted in >10 periportal areas selected randomly in each specimen stained with hematoxylin. Each value represents the mean ± SD.

Figure 8

AFP message is decreased in Iloprost-treated rat liver after 2-AAF/PHx. (A) Quantitative, real-time PCR shows a substantial reduction in liver AFP message after 2-AAF/PHx treatment in rats that received Iloprost (second set of bars) as compared with control rats (first set of bars). Data are normalized to the Iloprost-treated sample at day 5 after 2-AAF/PHx. Data represent the mean value ± SD (n ≥3). (B) AFP expressions at day 5, 7, and 9 in Iloprost-treated animals were significantly reduced compared with those of control animals in a Northern analysis.