Generation and expansion of multipotent mesenchymal progenitor cells from cultured human pancreatic islets.

Cellular models and culture conditions for in vitro expansion of insulin-producing cells represent a key element to develop cell therapy for diabetes. Initial evidence that human beta-cells could be expanded after undergoing a reversible epithelial-mesenchymal transition has been recently negated by genetic lineage tracing studies in mice. Here, we report that culturing human pancreatic islets in the presence of serum resulted in the emergence of a population of nestin-positive cells. These proliferating cells were mainly C-peptide negative, although in the first week in culture, proliferating cells, insulin promoter factor-1 (Ipf-1) positive, were observed. Later passages of islet-derived cells were Ipf-1 negative and displayed a mesenchymal phenotype. These human pancreatic islet-derived mesenchymal (hPIDM) cells were expanded up to 10(14) cells and were able to differentiate toward adipocytes, osteocytes and chondrocytes, similarly to mesenchymal stem/precursor cells. Interestingly, however, under serum-free conditions, hPIDM cells lost the mesenchymal phenotype, formed islet-like clusters (ILCs) and were able to produce and secrete insulin. These data suggest that, although these cells are likely to result from preexisting mesenchymal cells rather than beta-cells, hPIDM cells represent a valuable model for further developments toward future replacement therapy in diabetes.

Inhibition of B16 melanoma metastases with the ruthenium complex imidazolium trans-imidazoledimethylsulfoxide-tetrachlororuthenate and down-regulation of tumor cell invasion.

The antimetastatic ruthenium complex imidazolium trans-imidazoledimethylsulfoxide-tetrachlorouthenate (NAMI-A) is tested in the B16 melanoma model in vitro and in vivo. Treatment of B6D2F1 mice carrying intra-footpad B16 melanoma with 35 mg/kg/day NAMI-A for 6 days reduces metastasis weight independently of whether NAMI-A is given before or after surgical removal of the primary tumor. Metastasis reduction is unrelated to NAMI-A concentration, which is 10-fold lower than on primary site (1 versus 0.1 mM), and is correlated to the reduction of plasma gelatinolitic activity and to the decrease of cells expressing CD44, CD54, and integrin-beta(3) adhesion molecules. Metastatic cells also show the reduction of the S-phase cells with accumulation in the G(0)/G(1) phase. In vitro, on the highly metastatic B16F10 cell line, NAMI-A reduces cell Matrigel invasion and its ability to cross a layer of endothelial cells after short exposure (1 h) to 1 to 100 microM concentrations. In these conditions, NAMI-A reduces the gelatinase activity of tumor cells, and it also increases cell adhesion to poly-L-lysine and, in particular, to fibronectin, and this effect is associated to the increase of F-actin condensation. This work shows the selective effectiveness of NAMI-A on the metastatic melanoma and suggests that metastasis inhibition is due to the negative modulation of tumor cell invasion processes, a mechanism in which the reduction of the gelatinolitic activity of tumor cells plays a crucial role.

Influence of chemical stability on the activity of the antimetastasis ruthenium compound NAMI-A.

The influence of chemical stability on the antimetastatic ruthenium(III) compound imidazolium trans-imidazoletetrachlorodimethylsulphoxideruthenium(III) (NAMI-A) in aqueous solution was studied both in vitro and in vivo. The loss of dimethyl-sulphoxide (DMSO) ligand from the compound was tested by using a NAMI-A solution acidified with HCl at pH 3.0 and aged for 0, 4, 8 and 24 h prior to intraperitoneal (i.p.) injection into CBA mice bearing advanced MCa mammary carcinoma. The activity of NAMI-A on lung metastases showed no change even after the loss of DMSO ligand from up to 50% of the molecules. The reduction of NAMI-A did not modify the number of KB cells blocked in the S+G2M phases, independent of whether the reduction occurred outside the cells or after loading the cells with the compound prior to treatment with the reductants (ascorbic acid, glutathione or cysteine). In vivo, the complete reduction of NAMI-A with equivalent amounts of ascorbic acid, glutathione or cysteine prior to administration to mice bearing advanced MCa mammary carcinoma was more active than NAMI-A alone. The data show that NAMI-A, although undergoing a series of chemical modifications, maintains its antimetastatic activity in a broad range of experimental conditions.

Lack of In vitro cytotoxicity, associated to increased G(2)-M cell fraction and inhibition of matrigel invasion, may predict In vivo-selective antimetastasis activity of ruthenium complexes.

The ruthenium complexes trans-dichlorotetrakisdimethylsulfoxide ruthenium(II) (trans-Ru), imidazolium trans-imidazoletetrachlororuthenate (ICR), sodium trans-tetramethylensulfoxideisoquinolinetetrachlororuthenate (TEQU), and imidazolium trans-imidazoledimethylsulfoxidetetrachlororuthenate (NAMI-A) are tested in vitro by short exposure of MCF-7, LoVo, KB, and TS/A tumor cells to 10(-4) M concentration, and in vivo on Lewis lung carcinoma by a daily i.p. treatment for 6 consecutive days using equitoxic and maximum tolerated doses. NAMI-A 1) inhibited tumor cell invasion of matrigel, 2) induced a transient accumulation of cells in the G(2)-M phase, 3) did not modify in vitro cell growth, and 4) markedly reduced lung metastasis formation. TEQU showed significant cytotoxicity in vitro and was not antimetastatic in vivo. ICR and trans-Ru did not modify cell cycle distribution of in vitro tumor cells nor did they inhibit matrigel invasion; ICR was also devoid of antimetastasis effects in vivo. Ruthenium uptake by tumor cells did account for in vitro cytotoxicity but not for other in vitro actions or for in vivo antimetastasis activity. The contemporary absence of cytotoxicity, associated to inhibition of matrigel crossing and to transient block in the premitotic G(2)-M phase, appears to be prerequisites for a ruthenium compound to show in vivo-selective antimetastasis effect. The validation of this model for other classes of compounds will allow an understanding of the combined weight of the above-mentioned phenomena for tumor metastasis growth and control.

Effects of NAMI-A and some related ruthenium complexes on cell viability after short exposure of tumor cells.

A series of three ruthenium complexes, i.e. trans-dichlorote-trakisdimethyl-sulfoxide ruthenium(ll) (trans-Ru), imidazolium trans-imidazoletetra-chlororuthenate (ICR) and sodium trans-tetramethylensulfoxideisoquinoline-tetrachlororuthenate (TEQU), were studied in vitro in comparison to NAMI-A, a potent ruthenium-based antimetastasis agent. In vitro challenge of TS/A adenocarcinoma or KB oral carcinoma tumor cells with 10(-4) M concentration for 1 h evidenced the lack of cytotoxicity of NAMI-A, ICR and trans-Ru, the accumulation of cells in the G2/M pre-mitotic cell phase by NAMI-A and the attachment of tumor cells to the plastic substrate was significantly greater for NAMI-A than for ICR. These data stress that in vitro cytotoxicity is not necessary for in vivo activity of ruthenium antitumor complexes: NAMIA, ICR and trans-Ru, are in fact known to be active against murine tumors in the mouse system. Rather, TEQU, the compound free of in vivo activity, was the only one to reduce cell growth of in vitro cultured cells. In conclusion, the data on the effects of NAMI-A on in vitro cultured cells show that the increase of cell adhesion properties and the transient cell cycle arrest in the G2/M phase are much more relevant than the effects on cell properties relevant to cell growth (i.e. on CD44, CD54 or CD71 antigens) for determining in vivo antimetastasis activity.