Adenoviral-mediated transfer of Escherichia coli uracil phosphoribosyltransferase (UPRT) gene to modulate the sensitivity of the human colon cancer cells to 5-fluorouracil

5-Fluorouracil (5-FU) has been used as a chemotherapeutic drug for colorectal cancer. Escherichia coli uracil phosphoribosyltransferase (UPRT), a pyrimidine salvage enzyme, converts 5-FU into 5-fluorouridine monophosphate (5-FUMP) at the initial step of 5-FU activation. We investigated the effects of adenoviral-mediated transfer of the E. coli UPRT gene into human colon cancer cells on 5-FU metabolism and 5-FU chemosensitivity. Three cell lines were used (HT29, KM12 and SW1116). The intracellular levels of 5-fluorodeoxyuridine monophosphate (5-FdUMP) and 5-FU incorporated into RNA after 5-FU treatment in cells infected with adenovirus containing the UPRT gene (AdCA-UPRT) were significantly higher than those of non-infected cells. This was accompanied by marked inhibition of thymidylate synthase (TS) in all cell lines. Furthermore, HT29, KM12 and SW1116 infected with AdCA-UPRT were, respectively, 13.1-, 30.2- and 70.5-fold more sensitive to 5-FU than non-infected cells. Most importantly, treatment with AdCA-UPRT and 5-FU effectively inhibited the growth of HT29-xenografted subcutaneous tumours in nude mice. Therefore, AdCA-UPRT/5-FU treatment had the potential to enhance the actions of 5-FU at both the DNA and RNA levels. Treatment augmented the sensitivity of human colon cancer cells to 5-FU both in vitro and in vivo. We conclude that adenoviral-mediated transfer of the E. coli UPRT gene into colon cancer cells can achieve biochemical modulation of 5-FU and this provides a new approach in the treatment of colorectal cancer.     

Combined suicide gene therapy for human colon cancer cells using adenovirus-mediated transfer of escherichia coli cytosine deaminase gene and Escherichia coli uracil phosphoribosyltransferase gene with 5-fluorocytosine

The virus-directed enzyme/prodrug system using the Escherichia coli cytosine deaminase (CD) gene and 5-fluorocytosine (5-FC) suffers from a sensitivity limitation in many tumor cells. The E. coil uracil phosphoribosyltransferase (UPRT), which is a pyrimidine salvage enzyme, directly converts 5-fluorouracil (5-FU) to 5-fluorouridine monophosphate at the first step of its activating pathway. To improve the antitumoral effect of the CD/5-FC system, we investigated a combined suicide gene transduction therapy for human colon cancer cells using two separate adenovirus vectors expressing the E. coli CD and E. coli UPRT genes and systemic 5-FC administration (the CD, UPRT/5-FC system). The present study demonstrates that the CD, UPRT/5-FC system generates a co-operative effect of CD and UPRT, resulting in dramatic increases in both RNA- and DNA-directed active forms, including 5-fluorouridine triphosphate incorporated into RNA, 5-fluorodeoxyuridine monophosphate, and the thymidylate synthase inhibition rate, compared with the CD/5-FC system. Furthermore a significant increase in the 5-FC sensitivity of colon cancer cells was demonstrated in the CD, UPRT/5-FC system compared with the CD/5-FC system in vitro and in vivo. These results suggest that the CD, UPRT/5-FC system is a powerful approach in gene therapy for colorectal cancer.     

Circumvention of 5-fluorouracil resistance in human stomach cancer cells by uracil phosphoribosyltransferase gene transduction.

A human stomach cancer cell line with acquired resistance to 5-fluorouracil (5-FU), NUGC-3/5FU/ L, has been found to possess reduced ability to convert 5-FU into active metabolites. We attempted in vitro gene therapy for this 5-FU-resistant cell line. NUGC-3 and NUGC-3/5FU/L cells were infected with recombinant adenovirus (Ad) containing Escherichia coli uracil phosphoribosyltransferase (UPRT) gene driven by CAG promoter (CA), AdCA-UPRT, and changes in their 5-FU metabolism and sensitivity were investigated. Activities of orotate phosphoribosyltransferase increased from 10.2 and 1.56 (nmol/mg protein/30 min) in the uninfected cells of NUGC-3 and NUGC-3/5FU/L to 216 and 237, respectively, after the transfection of UPRT gene. The 5-FU nucleotide level in the acid-insoluble fraction increased from 7.32 to 15.9 (pmol/mg protein) in NUGC-3 cells on infection with AdCA-UPRT, and in NUGC-3/5FU/L cells it increased from 1.91 to 21.4. The 50% growth-inhibition concentration (IC50) was 12.7 micromol/liter for NUGC-3 and much higher than 100 micromol/liter for NUGC-3/5FU/L, indicating over 8-fold resistance. NUGC-3/ SFU/L transfected with the UPRT gene showed very high sensitivity to 5-FU with an IC50 of 3.2 micromol/liter. The high resistance in this metabolic activation-deficient cell line was thus completely reversed by transduction of an exogenous gene coding for a 5-FU-anabolizing enzyme.     

Expression of the interleukin-21 gene in murine colon carcinoma cells generates systemic immunity in the inoculated hosts

Interleukin-21 (IL-21) is a novel cytokine that can induce proliferation of activated T cells and maturation of natural killer (NK) cells. We therefore examined whether expression of the IL-21 gene in tumor cells could generate antitumor responses. Murine colon carcinoma Colon 26 cells that were transduced with the mouse IL-21 gene (Colon 26/IL-21) were rejected in syngeneic mice and the mice subsequently acquired protective immunity. The growth of Colon 26/IL-21 tumors developed in nude mice was retarded compared with that of parent tumors, and this growth suppression was not observed in nude mice that were treated with anti-asialo GM(1) antibody. Spleen cells from the mice that had rejected Colon 26/IL-21 cells showed cytotoxic activity to Colon 26 but not to irrelevant tumor cells, and produced larger amounts of interferon-gamma upon stimulation with irradiated Colon 26 cells. Spleen cells from Colon 26/IL-21-tumor- but not parent-tumor-bearing mice had lytic activity to YAC-1 cells. These data suggest that expression of IL-21 in tumors induces T- and NK-cell-dependent antitumor effects.     

Expression of IL-27 in murine carcinoma cells produces antitumor effects and induces protective immunity in inoculated host animals

A novel cytokine interleukin-27 (IL-27), composed of p28 and Epstein-Barr virus-induced gene 3 (EBI3), is produced from activated dendritic cells and is involved in an early phase of T-helper type I differentiation. We examined whether Colon 26 murine colon carcinoma cells that were retrovirally transduced with the p28-linked EBI3 gene (Colon 26/IL-27) could produce antitumor effects in inoculated mice. Although proliferation in vitro of Colon 26/IL-27 cells was not different from that of parent cells, syngeneic BALB/c mice rejected Colon 26/IL-27 tumors inoculated and subsequently acquired tumor-specific protective immunity. In contrast, mice inoculated with Colon 26 cells transduced with either the p28 or EBI3 gene developed tumors and survival of the mice remained the same as that of the mice inoculated with parent cells. Syngeneic nude mice developed Colon 26/IL-27 tumors, but the growth was retarded compared to that of parent tumors. Depletion of natural killer cells from nude mice with anti-asialo GM(1) antibody diminished the growth retardation of Colon 26/IL-27 tumors. Survival of severe combined immunodeficient mice that received subcutaneous inoculation of Colon 26/IL-27 cells was not different from that of the immunodeficient mice inoculated with parent cells. Interferon-gamma was produced from CD4(+) and CD8(+) T, and natural killer cells of the mice that rejected Colon 26/IL-27 tumors and cytotoxic activity against Colon 26 cells were also detected from the mice. These data collectively suggest that expressed IL-27 in tumors produces T cell-dependent and-independent antitumor effects and is a possible therapeutic strategy for cancer.     

Secretion of interleukin-10 from murine colon carcinoma cells suppresses systemic antitumor immunity and impairs protective immunity induced against the tumors

Interleukin-10 (IL-10) is a T helper type 2 (Th2) cytokine that suppresses Th1-mediated, cell-mediated immune responses and reciprocally enhances antibody-mediated responses. Previous studies, however, demonstrated that forced expression of the IL-10 gene in tumor cells could unexpectedly produce antitumor effects. We then examined whether tumor-derived IL-10 could modulate systemic immune responses. Murine colon carcinoma (Colon 26) cells that were retrovirally transduced with the murine IL-10 gene (Colon 26/IL-10) were inoculated in syngeneic immunocompetent or T cell-defective nude mice. Growth of Colon 26/IL-10 tumors was augmented in immunocompetent and, to less extent, in nude mice compared with that of wild-type tumors developed in respective mice. Growth of wild-type tumors was accelerated to the same level as that of Colon 26/IL-10 tumors when wild type and Colon 26/IL-10 cells were respectively inoculated in different flanks of the same immunocompetent mice. This enhanced growth of wild-type tumors was not observed in nude mice. Immunocompetent mice that had rejected IL-2- or IL-12-secreting Colon 26 cells developed protective immunity and became completely resistant to wild-type Colon 26 cells subsequently challenged. However, some of the mice that had rejected IL-2 or IL-12 producers developed Colon 26/IL-10 tumors inoculated thereafter. The present study showed that production of IL-10 from tumor cells impaired T cell- and non-T cell-mediated systemic antitumor immunity in hosts.     

Toxicity of levamisole and 5-fluorouracil in human colon carcinoma cells

The combination of 5-fluorouracil (5-FU) with the immunomodulator levamisole (Lev) has been clinically tested in patients with metastatic colorectal carcinoma and as adjuvant therapy following primary tumor surgery. In some studies in advanced disease, the addition of Lev to 5-FU improved the median duration of response; in the adjuvant setting, the combination was associated with improvement in the disease-free survival. We studied whether Lev was directly toxic to three human colorectal carcinoma cell lines (HCT 116, SNU-C4, and NCI-H630). We also evaluated the toxicity of Lev in combination with 5-FU in these three cell lines. Lev inhibited the growth of all three colorectal cell lines, but only at concentrations two logs above that achieved with a standard 150-mg oral dose of Lev. In cell growth studies, 500 and 1,000 microM Lev increased the toxicity of 5-FU in HCT 116 cells in an additive fashion. In clonogenic assays, continuous exposure to 10 or 100 microM Lev was minimally toxic and did not enhance the lethality associated with a 24-hour exposure to 5-FU in any of the cell lines. Lev alone at 1,000 microM decreased colony formation by 45% in HCT 116 cells. A combination of 1,000 microM Lev with 10 microM 5-FU resulted in a decrease in HCT 116 colony formation from 54% to 6% of control levels. Continuous exposure of NCI-H630 cells to 500 microM Lev decreased colony formation to 76.5% of control levels; when Lev was combined with 50 microM 5-FU, colony formation was decreased from 59.5% to 27.5% of control levels. We conclude that at concentrations achievable with conventional doses of Lev, there was no evidence of direct toxicity in these colorectal cell lines. Furthermore, an additive interaction with 5-FU was evident only at suprapharmacologic doses of Lev.     

Angiogenic effect of naive and 5-fluorouracil resistant colon carcinoma on endothelial cells in vitro

Tumour associated neovascularisation is a complex interplay between inhibitory and stimulatory angiogenic factors. Despite intense research in this field, little is known about the interaction between endothelial and chemoresistant cancer cells. For this purpose, we assessed the impact of cellular supernatants of the primary adenocarcinoma cell line CCL228, its lymph node metastasis CCL227, and four subclones resistant to different levels of 5-fluorouracil on the growth of microvascular and macrovascular endothelial cells. The growth of endothelial cells in vitro was affected to a moderate degree by supernatants from colon cancer cell lines. This effect was independent of the degree of chemoresistance. The stimulation of endothelial cells by the growth factors VEGF, bFGF, and PD-ECGF in the presence of supernatants from cancer cell lines was generally higher in macrovascular endothelial cells when compared with microvascular cells. The secretion of VEGF from colon cancer cells in vitro was inversely related to the degree of chemoresistance with the low chemoresistance phenotype producing VEGF 8.7-fold higher than the high resistance subclone. With a maximal secretion of 1500 pg VEGF/ml cell supernatant, the concentration necessary to directly stimulate the growth of endothelial cells was not achieved. In conclusion, chemoresistance affects the interaction between colon cancer cells and endothelial cells dependant on the endothelial cell type. Although the level of chemoresistance has a profound impact on the production of VEGF by cancer cells with low, intermediate or high resistance, it does not differentially affect growth stimulation or inhibition of endothelial cells in vitro.     

Survivin-3B gene decreases the invasion-inhibitory effect of colon cancer cells with 5-fluorouracil

The expression of survivin molecules has been confirmed in many types of cancer cells, including colon cancer cells, and they are considered important antiapoptotic molecules. Recent studies have revealed the existence of different splicing forms of survivin molecules; however, no studies have examined their expression in gastrointestinal cancers. In 2004, we reported the existence of the survivin-3B gene, a novel splice variant of survivin. In this study, we investigated the relationship between human colon cancer and our recently cloned survivin-3B gene with a coding region of 594 bp. In the first examination, survivin-3B expression was analyzed by RT-PCR in human colon cancer and adjacent normal mucosal tissues. The associations of its expression status with clinicopathological parameters and the prognosis were also examined. Survivin-3B mRNA expression was observed in 37 (46.3%) of 80 primary colon cancers, but not in the adjacent normal colonic mucosal tissue. The rate of survivin-3B gene expression was significantly higher in colon cancer with serosal invasion. The 5-year survival rate of patients with survivin-3B gene-positive primary colon cancer was significantly poorer, at 48.7%, than that (75.4%) of survivin-3B gene-negative patients. In the second examination, after the introduction of the survivin-3B gene into cells of the colon cancer cell line DLD-1, 5-fluorouracil-induced changes in their invasive capacity was examined. The invasion-inhibitory effect of 5-fluorouracil on survivin-3B gene-transfected DLD-1 cells was significantly lower than their empty vector gene-transfected counterparts. We speculate that survivin-3B expression in colon cancer is an important factor involved in the invasive capacity of cancer cells in the presence of anticancer drug.     

Effect of carbogen breathing on the pharmacodynamics of 5-fluorouracil in a murine colon carcinoma

To determine whether carbogen breathing has an effect on 5-fluorouracil (5-FU) uptake, retention and metabolism in C38 murine colon tumours grown in C57Bl/6 mice, we used in vivo 19F nuclear magnetic resonance (NMR) spectroscopy. Eleven tumour-bearing mice were treated with 150 mg/kg of 5-FU given intraperitoneally (i.p.). Five mice received carbogen gas (95% O(2) and 5% CO(2)) for 9.5 min, starting 1 min before 5-FU administration. We found increased levels of 5-FU and its anabolites and catabolites by sequential ?19F NMR spectroscopy in the group treated with 5-FU in combination with carbogen compared with the group treated with 5-FU alone. The maximum of normalised values of 5-FU and its metabolites, reached after carbogen breathing, was almost 2-fold higher than after treatment with 5-FU alone. Despite these increased concentrations no significant effect of carbogen on growth inhibition of the tumour by 5-FU was observed, which may be related to the size as well as the well vascularised and perfused conditions of the tumours studied.     

Orthotopic and ectopic organ environments differentially influence the sensitivity of murine colon carcinoma cells to doxorubicin and 5-fluorouracil.

We determined the effects of organ environment on the response of murine CT-26 colon carcinoma cells to 2 structurally and pharmacologically distinct chemotherapeutic agents. CT-26 cells were injected i.v. (to produce lung lesions), s.c., into the cecal wall, and into the spleen (to produce spleen and liver lesions). Doxorubicin (DXR) at 10 mg/kg, 5-fluorouracil (5-FU) at 20 mg/kg, or saline (control) was injected intravenously on different schedules after tumor-cell implantation. The in vivo responses of the tumors growing in the cecum, spleen, liver, lung and subcutis were compared. Colon carcinomas growing in the subcutis were most sensitive to DXR. Tumors growing in the spleen and cecum were most sensitive to 5-FU and less so to DXR. Tumors in the liver were highly resistant to both drugs, whereas experimental lung metastases were sensitive to 5-FU but resistant to DXR. The differential responses of the tumors to the drugs were not due to drug distribution. The level of protein-kinase-C activity was elevated in the spleen, liver and cecum tumors as compared with s.c. tumors and correlated with the in vivo DXR resistance of the tumor cells. This correlation suggested that organ environment may modulate the chemosensitivity of tumor cells, at least in part, by perturbing signal transduction pathways. Collectively, the data indicate that the organ environment has profound effects on the response of tumor cells to chemotherapy. A molecular understanding of this phenomenon should facilitate the design of more effective systemic chemotherapy for cancer metastases.     

Transduction of murine colon carcinoma cells with interleukin-15 gene induces antitumor effects in immunocompetent and immunocompromised hosts

We examined the antitumor effects caused by murine colon carcinoma cells (Colon 26) transduced with interleukin-15 (IL-15) gene. Although the in vitro proliferation rate of IL-15-secreting Colon 26 (Colon 26/IL-15) cells was not different from that of wild-type (wt) cells, small subcutaneous tumors of Colon 26/IL-15 cells that developed in syngeneic immunocompetent mice regressed spontaneously in contrast to tumors of wt cells. The mice that had eliminated tumors of Colon 26/IL-15 cells rejected wt cells when subsequently challenged. The survival of the mice that had been inoculated intraperitoneally with Colon 26/IL-15 cells was significantly prolonged compared with that of the mice injected with wt cells. However, in an experimental lung metastasis model, the survival of the mice inoculated with Colon 26/IL-15 cells remained the same as that of the mice inoculated with wt cells. The inoculation of Colon 26/IL-15 cells into immunocompromised nude or severe combined immunodeficient mice produced tumors, but the survival of the immunocompromised mice was significantly longer than that of the mice inoculated with wt cells. The nude mice inoculated with Colon 26/IL-15 cells also survived longer than the severe combined immunodeficient mice with Colon 26/IL-15 cells. Depletion of natural killer cells in nude mice with anti-asialo GM1 antibody did not influence the survival of the mice injected with Colon 26/IL-15 cells. Immunohistological examination revealed that CD31+ cells migrated into tumors of Colon 26/IL-15 cells that developed in immunocompetent and immunocompromised mice. Taken together, our results indicate that an inoculation of IL-15-producing tumor cells can produce antitumor effects that are mediated by a variety of immunocompetent cells.     

Bystander effect in uracil phosphoribosyltransferase/5-fluorouracil-mediated suicide gene therapy is correlated with the level of intercellular communication

We examined whether a suicide gene/prodrug system using the uracil phosphoribosyltransferase (UPRT) of E. coli origin and 5-fluorouracil (5-FU) could achieve a bystander effect in two rodent tumor cell lines, murine colon carcinoma (Colon 26) and rat gliosarcoma (9L) cells. Cytotoxicity tests of mixed populations consisting of parent and transduced cells showed that the bystander effect was not produced in Colon 26 cells in either the UPRT/5-FU system or the herpes simplex virus-thymidine kinase/ganciclovir system but a strong bystander effect was evidenced by both suicide gene systems in 9L cells. The expression level of connexin 43, a protein that constitutes gap junctions, was high in 9L but low in Colon 26 cells. A gap junction-permeable fluorescein dye could be transferred among 9L cells but hardly at all among Colon 26 cells. Taken together, the efficacy of the bystander effect in the UPRT/5-FU system can be affected by gap junction-mediated intercellular communication.     

Induction of systemic immunity by expression of interleukin-23 in murine colon carcinoma cells

Interleukin-23 (IL-23), a novel cytokine composed of a newly identified p19 molecule and the p40 subunit of IL-12, can stimulate the proliferation in vitro of memory T cells. We examined whether Colon 26 murine colon carcinoma cells that were retrovirally transduced with the p19-linked p40 gene (Colon 26/IL-23) could produce antitumor effects in inoculated mice. The growth of Colon 26/IL-23 tumors developed in immunocompetent mice was significantly retarded and the tumors disappeared thereafter. Spleen cells from the mice that received Colon 26/IL-23 cells produced significant amounts of interferon-gamma, when they were cultured with irradiated Colon 26 but not irrelevant cells. Depletion of CD8(+) T cells suppressed the production of interferon-gamma. The mice that had rejected Colon 26/IL-23 tumors were resistant to subsequent challenge of parent but not irrelevant tumor cells. Colon 26/IL-23 tumors were not rejected in nude mice but the growth was retarded compared to parent tumors. Treatment of nude mice with anti-asialo GM(1) antibody did not influence the growth of Colon 26/IL-23 tumors. These data suggest that expression of IL-23 in tumors produces T cell-dependent antitumor effects and induces systemic immunity.     

Correlation between 5-fluorouracil metabolism and treatment response in two variants of C26 murine colon carcinoma

Following an i.p. dose of 150 mg x kg(-1) 5-fluorouracil (5-FU), drug uptake and metabolism over a 2-h period were studied by in vivo (19)F magnetic resonance spectroscopy (MRS) for the murine colon carcinoma lines C26-B (5-FU-insensitive; n=11) and C26-10 (5-FU-sensitive; n=15) implanted s.c. in Balb/C mice. Time courses for tumour growth, intracellular levels of FdUMP, thymidylate synthase (TS) activity, and 5-FU in RNA were also determined, and the effects of a 9.5-min period of carbogen breathing, starting 1 min before drug administration, on MRS-detected 5-FU metabolism and tumour growth curves were examined. Both tumour variants generated MRS-detectable 5-FU nucleotides and showed similar initial growth inhibition after treatment. However, the growth rate of C26-B tumours returned to normal, while the sensitive C26-10 tumours, which produced larger fluoronucleotide pools, still showed moderate growth inhibition. Carbogen breathing did not significantly influence 5-FU uptake or fluoronucleotide production but did significantly enhance growth inhibition in C26-10 tumours. While both tumour variants exhibited incorporation of 5-FU into RNA and inhibition of TS via FdUMP, clearance of 5-FU from RNA and recovery of TS activity were greater for the insensitive C26-B line, indicating that these processes, in addition to 5-FU uptake and metabolism, may be important determinants of drug sensitivity and treatment response.     

Antagonistic effect of polyinosinic-polycytidylic acid on the cell lethality produced by 5-fluorouracil in human colon carcinoma cells in vitro

The effect of polyinosinic X polycytidylic acid [poly(I) X poly(C)] on the cell lethality produced by 5-fluorouracil [FUra] and 5-fluorouridine [FUrd] was examined in human colon carcinoma cell line HT-29. Pretreatment of cells with poly(I) X poly(C) as well as during exposure to FUra or FUrd resulted in antagonism of the cell lethality generated by the fluoropyrimidines. Antagonism of FUra cytotoxicity was also produced by the 2'-O-methylated analogue, polyinosinic X poly-2'-O-methylcytidylic acid, but not by the individual single-stranded polynucleotides or by the component mononucleotides, cytidine 5'-phosphate and inosine 5'-phosphate. In contrast, cytidine 5'-phosphate blocked the toxicity of FUrd. The antagonism by poly(I) X poly(C) of FUra and FUrd cytotoxicity was related to the inhibition of their metabolism to fluorouridine triphosphate and their incorporation into RNA and not to inhibition of the synthesis of RNA. Antibodies to leukocyte and fibroblast interferons did not reverse the antagonistic activity of poly(I) X poly(C). These results indicate that poly(I) X poly(C) may be interfering with the transport and/or initial metabolism of FUra and FUrd to fluorouridine monophosphate which is independent of the ability of the double-stranded RNA to induce interferon.     

A Fas-dependent component in 5-fluorouracil/leucovorin-induced cytotoxicity in colon carcinoma cells.

We have shown previously (J. A. Houghton et al., Proc. Natl. Acad. Sci. USA, 94: 8144-8149, 1997) that thymineless death in thymidylate synthase-deficient (TS-) colon carcinoma cells is mediated via Fas/FasL interactions after deoxythymidine (dThd) deprivation, and that Fas-dependent sensitivity of human colon carcinoma cell lines may be dependent upon the level of Fas expressed. The objective of this study was to elucidate whether a Fas-dependent component exists in 5-fluorouracil (FUra)/leucovorin (LV)-induced cytotoxicity of colon carcinoma cells, and whether this may be potentiated by IFN-gamma-induced elevation in Fas expression, using the HT29 cell line as a model. The cytotoxic activity of FUra/LV was inhibited by dThd in HT29 cells and also, in part, by NOK-1+NOK-2 MoAbs that prevent Fas/FasL interactions. FUra/LV-induced cytotoxicity was significantly potentiated by IFN-gamma, reversed by exposure to NOK-1+NOK-2 antibodies, and correlated with a 4-fold induction of Fas expression in the presence of IFN-gamma and significant elevation in expression of FasL. Using five additional human colon carcinoma cell lines, FUra/LV-induced cytotoxicity was dThd-dependent in GC3/c1, VRC5/c1, and Caco2 but not in HCT8 or HCT116 cells. Like HT29 cells, this cytotoxicity was potentiated by IFN-gamma in GC3/c1 and VRC5/c1 but not in Caco2, which fails to express Fas, nor in HCT8 and HCT116, in which no dThd-dependent FUra-induced cytotoxicity was demonstrated. Data suggest that a Fas-dependent component, potentiated by IFN-gamma, exists in FUra/LV-induced cytotoxicity but requires FUra/LV-induced DNA damage for IFN-gamma-induced potentiation to occur.     

Effective gene therapy of biliary tract cancers by a conditionally replicative adenovirus expressing uracil phosphoribosyltransferase: significance of timing of 5-fluorouracil administration

In order to enhance the efficacy of conditionally replicating adenoviruses (CRAd) in the treatment of cancers of the biliary tract, we studied the efficacy in vitro and in vivo of AxE1CAUP, a CRAd vector that carries a gene for uracil phosphoribosyltransferase (UPRT), which converts 5-fluorouracil (5-FU) directly to 5-fluorouridine monophosphate and greatly enhances the cytotoxicity of 5-FU. AxE1CAUP replicated and induced an increased UPRT expression in biliary cancer cells more efficiently than AxCAUP, a nonreplicative adenovirus carrying the UPRT gene. Whereas AxCAUP and AxE1AdB, a CRAd without the UPRT gene, modestly increased the sensitivity of BC cells to 5-FU, AxE1CAUP markedly increased the sensitivity, especially when the timing of 5-FU administration was appropriately chosen. AxE1CAUP replicated much less efficiently in normal WI-38 fibroblasts without any change in the sensitivity to 5-FU. In nude mice with s.c. biliary cancer xenografts, i.t. AxE1CAUP/5-FU therapy inhibited tumor growth significantly more strongly than AxCAUP/5-FU or AxE1AdB/5-FU therapy. Furthermore, in mice with peritoneally disseminated biliary cancer, i.p. AxE1CAUP efficiently proliferated in the tumors, decreased the tumor burden, and prolonged the survival of the mice when 5-FU was started 10 or 15 days after the vector inoculation, whereas earlier initiation of 5-FU resulted in early eradication of the vector and no survival benefit. The present study shows that the CRAd expressing UPRT was a more potent sensitizer of biliary cancer to 5-FU, than was a nonreplicative UPRT-encoding vector or a CRAd without UPRT gene, even at a lower dose of the vector, and that timing of 5-FU administration was a key factor to maximize the efficacy. This gene therapy with appropriately timed administration of 5-FU should be useful in overcoming the resistance of biliary cancers to 5-FU.     

Liposomal prednisolone phosphate potentiates the antitumor activity of liposomal 5-fluorouracil in C26 murine colon carcinoma in vivo

The antitumor efficacy of 5-fluorouracil (5-FU) in advanced colorectal cancer (CRC) is hindered not only by the low therapeutic index, but also by tumor cell resistance to this cytotoxic drug. Therefore, to enhance the 5-FU antitumor activity, the present research used a novel tumor-targeted therapy based on the co-administration of 5-FU encapsulated in long-circulating liposomes (LCL-5-FU) together with liposomal prednisolone phosphate (LCL-PLP), a formulation with known anti-angiogenic actions on C26 murine colon carcinoma cells. Thus, we assessed the in vivo effects of the combined liposomal drug therapy on C26 carcinoma growth as well as on the production of molecular markers with key roles in tumor development such as angiogenic, inflammatory, and oxidative stress molecules. To get further insight into the polarization state of tumor microenvironment after the treatment, we determined the IL-10/IL-12p70 ratio in tumors. Our results showed that combined liposomal drug therapy inhibited almost totally tumor growth and was superior as antitumor activity to both single liposomal drug therapies tested. The antitumor efficacy of the combined therapy was mainly related to the anti-angiogenic and anti-inflammatory actions on C26 carcinoma milieu, being favored by its controlling effect on intratumor oxidative stress and the skewing of polarization of tumor microenvironmental cells toward their antineoplastic phenotypes. Thus, our study unveils a promising treatment strategy for CRC that should be furthermore considered.     

Gene expression profiling-based prediction of response of colon carcinoma cells to 5-fluorouracil and camptothecin

5-Fluorouracil (5-FU) is the most common chemotherapeutic agent used in the treatment of colorectal cancer, yet objective response rates are low. Recently, camptothecin (CPT) has emerged as an effective alternative therapy. Decisive means to determine treatment, based on the likelihood of response to each of these agents, could greatly enhance the management of this disease. Here, the ability of cDNA microarray-generated basal gene expression profiles to predict apoptotic response to 5-FU and CPT was determined in a panel of 30 colon carcinoma cell lines. Genes whose basal level of expression correlated significantly with 5-FU- and CPT-induced apoptosis were selected, and their predictive power was assessed using a "leave one out" jackknife cross-validation strategy. Selection of the 50 genes best correlated with 5-FU-induced apoptosis, but not 50 randomly selected genes, significantly predicted response to this agent. Importantly, this gene expression profiling approach predicted response more effectively than four previously established determinants of 5-FU response: thymidylate synthase and thymidine phosphorylase activity; and p53 and mismatch repair status. Furthermore, reanalysis of the database demonstrated that selection of the 149 genes best correlated with CPT-induced apoptosis maximally and significantly predicted response to this agent. These studies demonstrate that the basal gene expression profile of colon cancer cells can be used to predict and distinguish response to multiple chemotherapeutic agents and establish the potential of this methodology as a means by which rational decisions regarding choice of therapy can be approached.