Human Soluble TRAIL Secreted by Modified Lactococcus lactis Bacteria Promotes Tumor Growth in the Orthotopic Mouse Model of Colorectal Cancer.

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) selectively induces apoptosis of sensitive cancer cells, including colorectal cancer (CRC). Due to its short biological half-life after intravenous administration and related clinical ineffectiveness, novel formulations of TRAIL need to be developed. Here we propose Lactococcus lactis bacteria as a vehicle for local delivery of human soluble TRAIL (hsTRAIL) in CRC. The use of common probiotics targeting guts as carriers for TRAIL could ensure its sustained release at the tumor site and extend the duration of its activity. We have already engineered hsTRAIL-secreting L.lactis bacteria and showed their effectiveness in elimination of human CRC cells in vitro and in vivo in a mouse subcutaneous model. Here, L.lactis(hsTRAIL+) were administered by gastric gavage to SCID mice with orthotopically developed HCT116 tumor in cecum, in monotherapy or in combination with metformin (MetF), already shown to enhance the hsTRAIL anti-tumor activity in subcutaneous CRC model. Oral administration of L.lactis(hsTRAIL+) resulted in significant progression of HCT116 tumors and shortening of the colon crypts. Secretion of hsTRAIL in the colon was accompanied by infiltration of the primary tumor with M2-macrophages, while MetF promoted transient colonization of the gut by L.lactis. Our study indicates that L.lactis bacteria after oral administration enable delivery of biologically active hsTRAIL to colon, however its potential therapeutic effect in CRC treatment is abolished by its pro-tumorigenic signalling, leading to the recruitment of M2-macrophages and tumor growth promotion.

The Anti-Tumor Effect of Lactococcus lactis Bacteria-Secreting Human Soluble TRAIL Can Be Enhanced by Metformin Both In Vitro and In Vivo in a Mouse Model of Human Colorectal Cancer.

Tumor Necrosis Factor-Related Apoptosis-Inducing Ligand (TRAIL) induces apoptosis of many cancer cells, including CRC cells, being non-harmful for normal ones. However, recombinant form of human TRAIL failed in clinical trial when administered intravenously. To assess the importance of TRAIL in CRC patients, new form of TRAIL delivery would be required. Here we used genetically modified, non-pathogenic Lactococcus lactis bacteria as a vehicle for local delivery of human soluble TRAIL (hsTRAIL) in CRC. Operating under the Nisin Controlled Gene Expression System (NICE), the modified bacteria (L. lactis(hsTRAIL+)) were able to induce cell death of HCT116 and SW480 human cancer cells and reduce the growth of HCT116-tumor spheres in vitro. This effect was cancer cell specific as the cells of normal colon epithelium (FHC cells) were not affected by hsTRAIL-producing bacteria. Metformin (MetF), 5-fluorouracil (5-FU) and irinotecan (CPT-11) enhanced the anti-tumor actions of hsTRAIL in vitro. In the NOD-SCID mouse model, treatment of subcutaneous HCT116-tumors with L. lactis(hsTRAIL+) bacteria given intratumorally, significantly reduced the tumor growth. This anti-tumor activity of hsTRAIL in vivo was further enhanced by oral administration of MetF. These findings indicate that L. lactis bacteria could be suitable for local delivery of biologically active human proteins. At the same time, we documented that anti-tumor activity of hsTRAIL in experimental therapy of CRC can be further enhanced by MetF given orally, opening a venue for alternative CRC-treatment strategies.

Secretion of tumoricidal human tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) by recombinant Lactococcus lactis: optimization of in vitro synthesis conditions.

BACKGROUND: Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) selectively eliminates tumor cells. However, the short biological half-life of this molecule limits its potential use in the clinic. Our aim was to construct a recombinant strain of nonpathogenic Lactococcus lactis bacteria as a vector for effective and prolonged human TRAIL production. Herein, we examined the expression and secretion conditions leading to the production of biologically active protein in vitro. RESULTS: The human soluble TRAIL-cDNA (hsTRAIL-cDNA) with optimized codons was designed to fit the codon usage pattern (codon bias) of the L. lactis host. This cDNA construct was synthesized and cloned in lactococcal plasmid secretion vector pNZ8124 under the control of the nisin-induced PnisA promoter. The pNZ8124-hsTRAIL plasmid vector was transformed into the L. lactis NZ9000 host strain cells by electroporation. Secretion of the protein occurred at the neutral pH during induction, with optimized concentration of the inducer and presence of serine proteases inhibitor. Using Western blotting and amino acid sequencing method we found that TRAIL was secreted in two forms, as visualized by the presence of two distinct molecular size bands, both deprived of the usp45 protein, the bacterial signal peptide. By the use of MTS assay we were able to prove that hsTRAIL present in supernatant from L. lactis (hsTRAIL+) broth culture was cytotoxic to human HCT116 colon cancer cells but not to normal human fibroblasts. Flow cytometry analysis revealed TRAIL-induced apoptosis of cancer cells. CONCLUSIONS: We designed recombinant L. lactis bacteria, which efficiently produce biologically active, anti-tumorigenic human TRAIL in vitro. Further studies in tumor-bearing NOD-SCID mice will reveal whether the TRAIL-secreting L. lactis bacteria can be used as a safe carrier of this protein, capable of inducing effective elimination of human colon cancer cells in vivo.

Induction of monocyte antitumor response by human cancer cells transduced with TNF-GFP fusion gene: possible implications for immunotherapy of cancer.

This study was undertaken to determine how human pancreatic cancer (HPC-4) cells transduced with the TNF-GFP fusion gene (TLG) alter the antitumor response of human monocytes in vitro and whether they could act as an antitumor vaccine. In our model, HPC-4 cells were transduced with retroviral vector harboring TLG gene and designated as HPC-4(TLG). The TLG protein expression was confirmed by Western blot and flow cytometry analysis. Monocytes were co-cultured with transduced and control HPC-4 cells. The secretion of TNF, IL-10 and IL-12 was measured by ELISA. The cytotoxicity of monocytes against HPC-4 cells was determined by MTT test. The results show that the HPC-4(TLG) cells expressed membrane-bound, intracellular and secretory TLG protein. When cultured with HPC-4(TLG) cells, monocytes released a higher amount of TNF, but IL-10 and IL-12 secretion was inhibited. The pre-exposure of monocytes to HPC-4(TLG), but not to HPC-4, cells did not decrease TNF nor increase IL-10 production, thus not leading to monocyte deactivation. Also, the antitumor cytotoxicity of monocytes stimulated with HPC-4(TLG) was not downregulated, which occurred when non-transduced HPC-4 cells were used. In conclusion, compared to parental HPC-4 cells, TLG gene transduced HPC-4 cells induced stronger antitumor response of monocytes in vitro and prevented deactivation of monocytes.

Tumor cell-induced deactivation of human monocytes.

Although blood monocytes exhibit significant cytotoxic activity against tumor cells, the function of tumor infiltrating macrophages (TIM) is depressed in cancer patients. This study addresses the question of how the antitumor response of human monocytes, assessed by production of cytokines (tumor necrosis factor alpha, TNF; IL-10; IL-12p40) and cytotoxicity, is altered by exposure to cancer cells. Tumor cell--pre-exposed monocytes restimulated with tumor cells showed significantly decreased production of TNF, IL-12, increased IL-10 (mRNA and release) and inhibition of IL-1 receptor-associated kinase-1 (IRAK-1) expression. This down-regulation of cytokine production was selective, as the response of pre-exposed monocytes to lipopolysaccharide (LPS) was unaffected. Treatment of tumor cell--pre-exposed monocytes with hyaluronidase (HAase) improved their depressed production of TNF, while HAase-treated cancer cells did not cause monocyte dysfunction. The response of hyaluronan (HA)--pre-exposed monocytes to stimulation with tumor cells was also inhibited. Cytotoxic activity of monocytes pretreated with cancer cells was also decreased. This study shows that tumor cells selectively deactivate monocytes and suggests that tumor cell-derived HA by blocking CD44 on monocytes inhibits their antitumor response. These observations may provide some explanation for the depressed function of TIM in human malignancy.