Interplay of neuropilin-1 and semaphorin 3A after partial hepatectomy in rats.

aimTo elucidate the role of neuropilin-1 (Nrp-1) and semaphorin 3A (Sema3A) in sinusoidal remodeling during liver regeneration in rats.

methodsMale Wistar/ST rats at 7 wk of age, weighing about 200 g, were used for all animal experiments. In vivo, at 24, 48, 72, 96, 144 and 192 h after two-thirds partial hepatectomy (PHx), the remnant livers were removed. Liver tissues were immunohistochemically stained for Nrp-1, Sema3A and SE-1, a liver sinusoidal endothelial cell (SEC) marker. Total RNA of the liver tissue was extracted and reversely transcribed into cDNA. The mRNA expression of Sema3A was analyzed by quantitative real-time polymerase chain reaction and normalized to that of ribosomal protein S18. In vitro, SECs were isolated from rat liver and cultured in endothelial growth medium containing 20 ng/mL vascular endothelial cell growth factor. Migration of SECs in primary culture was assessed by cell transwell assay with or without recombinant Sema3A. Apoptotic cells were determined by a terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling method.

resultsIn vitro, immunohistochemistry study revealed that Sema3A and Nrp-1 were constitutively expressed in hepatocytes and SECs, respectively, in normal rat liver tissues. Nrp-1 expression in SECs was quantified by the percentage of immunostained area with anti-Nrp-1 antibody in relation to the area stained with SE-1. Between 24 h and 96 h following resection of liver, Nrp-1 expression in SECs was transiently increased. Compared with the baseline (5.2% ± 0.1%), Nrp-1 expression in SECs significantly increased at 24 h (17.3% ± 0.7%, P < 0.05), 48 h (39.1% ± 0.6%, P < 0.01), 72 h (46.9% ± 4.5%, P < 0.01) and 96 h (29.9% ± 3.8%, P < 0.01) after PHx, then returned to the basal level at termination of liver regeneration. Interestingly, the expression of Sema3A was inversely associated with that of Nrp-1 in liver after PHx. Sema3A mRNA expression was significantly reduced by about 75% over the period 24-144 h after PHx (P < 0.05), and returned to basal levels at 192 h after PHx. In vitro, SECs isolated from rats after PHx (PHx-SECs) were observed to migrate to the lower chamber of the cell transwell system after incubation for 24 h, but not cells from normal rats (CONT-SECs), indicating that mobility of PHx-SECs increases as compared with that of CONT-SECs. Moreover, recombinant Sema3A significantly attenuated migration in PHx-SECs in primary culture (vehicle-treated 100% ± 7.9% vs Sema3A-treated 42.6% ± 5.4%, P < 0.01), but not in CONT-SECs. Compared with CONT-SECs, the apoptotic rate of PHx-SECs decreased by 78.3% (P < 0.05). There was no difference in apoptosis between CONT-SECs that were treated with vehicle and Sema3A. However, in PHx-SECs, apoptosis was induced by the presence of 5 nmol Sema3A for 24 h (vehicle-treated 21.7% ± 7.6% vs Sema3A-treated 104.3% ± 8.9%, P < 0.05). In addition, immunohistochemistry confirmed the increased expression of Nrp-1 in PHx-SECs, while it was noted to a lesser extent in CONT-SECs.

conclusionThe interplay of Nrp-1 and Sema3A shown in our results may lead to a better understanding of interaction between sinusoidal remodeling and SECs during liver regeneration.