Nobuyasu Hayashi

Thiazolidinediones inhibit growth of gastrointestinal, biliary, and pancreatic adenocarcinoma cells through activation of the peroxisome proliferator-activated receptor γ/retinoid X receptor α pathway

Peroxisome prolixferator-activated receptor gamma (PPARgamma), a ligand-activated transcription factor, forms a heterodimer with retinoid X receptor alpha (RXRalpha), and its transcriptional activity is thought to be maximal in the presence of both PPARgamma and RXRalpha ligands. Although previous studies suggested that thiazolidinediones (TZDs), known as PPARgamma ligands, inhibit the growth of several types of tumor cells, the precise mechanism still remains obscure. The present study was designed to examine the effects of PPARgamma/RXRalpha transcriptional activation on cell growth in cancer cells. We compared the effects of six types of TZDs (troglitazone, RS-1303, RS-1330, RS-1387, RS-1455, and RS-1456) and 9-cis RA, an RXRalpha ligand, on the activation of PPARgamma/RXRalpha and the growth inhibition of six types of adenocarcinoma cell lines (MKN45, HT-29, HCT116, HuCCT1, KMP-2, and BxPC3) established from abdominal malignancies. PPARgamma was expressed in all six tumor cell lines and transcriptionally functional in five of the six lines. The stronger PPARgamma activator showed the stronger growth inhibitor in these five cell lines. However, no significant growth inhibitory effect of six types of PPARgamma activators was observed in BxPC3 cells, which showed no significant PPARgamma transactivation by these activators. Simultaneous addition of troglitazone and 9-cis RA enhanced both activation of PPARgamma/RXRalpha and growth inhibition in several types of cancer cells. The degree of PPARgamma/RXRalpha activation correlated with the extent of growth inhibition (r > 0.70, P < 0.05). This growth inhibition was associated with G1 cell cycle arrest and cell differentiation. These findings suggest that activation of the PPARgamma/RXRalpha pathway plays an important role in the growth inhibition of tumor cells and that this nuclear hormone receptor may be a possible novel molecular target for treatment of tumors in humans.

Suppression of pancreatic cancer cell invasion by a cyclooxygenase-2-specific inhibitor.

Pancreatic cancer is characterized by invasive and metastatic potential. In this study, effects of the COX-2 inhibitor JTE-522 on cell viability, invasion, and invasion-related cellular properties were determined. JTE-522 (10 μM) induced a 75–90% reduction in invasion, compared to cells treated with a vehicle only, in the COX-2-expressing cells. In contrast, this inhibitor caused no significant reduction in cells lacking COX-2. Determinants of cell invasion, such as cell motility, adhesion to the extracellular matrix, and gelatinolytic activity of metalloproteinase, were also modulated in COX-2-positive pancreatic cancer cells. Thus, COX-2-specific inhibitors may be a useful anti-invasive therapeutic option in pancreatic cancer.

Schedule-dependent therapeutic effects of gemcitabine combined with uracil-tegafur in a human pancreatic cancer xenograft model.

Objectives: Gemcitabine is taken up by cells mainly via human equilibrative nucleoside transporter 1 (hENT1). Pretreatment of cancer cell lines with 5-fluorouracil (5-FU) leads to an increase in the expression of hENT1 and augments the effect of single-agent gemcitabine treatment in vitro. The purpose of the present study was to evaluate the relationship between the schedules of gemcitabine/uracil-tegafur (UFT) combination therapy and their effects in pancreatic cancer in vivo. Methods: The expression level of hENT1 mRNA was examined using 6 types of human pancreatic cancer cell lines treated with 5-FU and MiaPaCa-2 xenograft tumors in BALB/c nu/nu mice treated with UFT. A [3H] gemcitabine cellular uptake assay was performed using MiaPaCa-2 cells treated with 5-FU. We compared the effects of 6 different schedules of treatment using UFT and/or gemcitabine on MiaPaCa-2 xenograft tumors. Results: MiaPaCa-2 cell line was one of the lines that showed the highest rate of 5-FU-induced increase in the hENT1 mRNA level. Gemcitabine uptake was significantly increased when cells were treated with 5-FU. Treatment with UFT significantly increased the hENT1 mRNA expression in MiaPaCa-2 tumors. A significant growth inhibition of MiaPaCa-2 tumors was observed in the mice treated with UFT followed by gemcitabine as compared with either untreated mice or UFT alone-treated mice. Conclusions: Our results suggest that the schedule in which the gemcitabine is administered after UFT may be the optimal combination for gemcitabine/UFT treatment in pancreatic cancer.

Association between Expression Levels of CA 19-9 and N-Acetylglucosamine-β1,3-Galactosyltransferase 5 Gene in Human Pancreatic Cancer Tissue

Objective: CA 19-9, equivalent to Sialyl Lewisa antigen, is a well-known tumor marker in pancreatic cancer. At the initial step of the biosynthesis of CA 19-9, N-acetylglucosamine-β1,3-galactosyltransferase (β3Gal-T) transfers galactose to N-acetylglucosamine (GlcNAc). Recently, β3Gal-T5 has been presumed to be related to the formation of the type 1 chain in an in vitro experiment in terms of kinetic enzyme characterization. The purpose of this study was to investigate which β3Gal-T is related to the synthesis of CA 19-9 in human pancreatic cancer tissues. Methods: We examined β3Gal-T1, T2, T3, T4, and β3Gal-T5 mRNA expressions in 13 noncancerous and cancerous tissues of the human pancreas using real-time polymerase chain reaction, and compared those gene expression levels with the immunoreactivity of CA 19-9 and its precursor DUPAN-2 in cancerous tissues. Results:β3Gal-T5 gene expression significantly augmented in cancerous tissues, when compared with the adjacent noncancerous tissues. Additionally, there was a good correlation between β3Gal-T5 gene transcription levels and immunohistochemical grades of CA 19-9 or its precursor DUPAN-2 in cancerous tissues. However, no correlation was observed between β3Gal-T1, T2, T3, and β3Gal-T4 gene expression levels and CA 19-9 or DUPAN-2 immunoreactive grades in cancerous tissue. Conclusion: β3Gal-T5 is presumed to be responsible for the synthesis of CA 19-9 in pancreatic cancer tissue.