Karen M. Holdaway

Inhibition of growth of colon 38 adenocarcinoma by vinblastine and colchicine: Evidence for a vascular mechanism

Vinblastine or colchicine, administered intraperitoneally to B6D2F1 mice with advanced subcutaneous colon 38 tumours, induced substantial tumour growth delays with progressive development of haemorrhagic necrosis beginning within 8 hours of treatment. Two multidrug-resistant P388 leukaemia sublines, refractory to vinblastine and vincristine when grown as intraperitoneal ascites, were sensitive to necrosis induction when grown as subcutaneous tumours. Vascular labelling with two fluorescent markers indicated that vincristine substantially reduced tumour blood flow within 4 hours after treatment. The effects of vinblastine, vincristine and colchicine were similar to those of tumour necrosis factor alpha in that: (a) similar tumour necrosis and blood flow changes were induced, (b) coadministration of the serotonin antagonist cyproheptidine prevented tumour necrosis and (c) plasma nitrate levels were elevated, indicative of the stimulation of oxidation of L-arginine to nitric oxide. The results suggest that vinca alkaloids and colchicine act on solid tumours by host cell-mediated vascular effects as well as by direct tubulin-mediated cytotoxicity.

Inhibition of growth of primary human tumour cell cultures by a 4-anilinoquinazoline inhibitor of the epidermal growth factor receptor family of tyrosine kinases

The epidermal growth factor receptor (EGFR) is thought to mediate the action of the mitogens EGF and tumour growth factor-alpha (TGF-alpha) in a variety of cancers, including those of the lung, breast and ovary. A number of new selective inhibitors of EGFR tyrosine kinase have now been developed as potential new antitumour agents. We used a potent inhibitor of this tyrosine kinase, 6-amino-4-[(3-bromophenyl)amino]-7-(methylamino)quinazoline (SN 25531; PD 156273), to determine the responses of primary cultures derived from patients with cancer of the lung, ovary, breast, cervix and endometrium. Cells were cultured in 96-well plates and proliferation assessed by incorporation of 3H-thymidine. Measured growth inhibitory concentrations IC50 values) varied from 1 nM to 14 microM with a 1000-fold differential between sensitive and resistant cultures. Results were compared with rates of proliferation, estimated using a paclitaxel-based method. We also measured the IC50 values for the tyrosine kinase inhibitor using a number of established human cell lines, and compared them with EGFR content using fluorescent antibody staining and flow cytometry. The presence of EGFR was found to be necessary, but not sufficient, for in vitro response. Only a small number of cell lines (3 of 7 for lung, 1 of 7 for ovarian, 2 of 3 squamous cell and 0 of 12 for melanoma) were sensitive to the tyrosine kinase inhibitor. In contrast, 40 of the 50 primary cultures (including 14 of 15 lung cancer samples and 14 of 19 ovarian cancer samples) were sensitive.

Application of fluorescence in situ hybridisation to study the relationship between cytotoxicity, chromosome aberrations, and changes in chromosome number after treatment with the topoisomerase II inhibitor amsacrine.

Amsacrine (4′‐(9‐acridinylamino)methanesulphon‐m‐anisidide) is an antileukemic drug which inhibits topoisomerase II (topo II) enzymes. We studied effects of two concentrations of amsacrine on the GM10115A cell line. This is a Chinese hamster line containing a single human chromosome 4, which can be readily visualised using fluorescence in situ hybridisation (FISH). The low amsacrine concentration slowed cell growth but did not cause significant arrest in the G2 phase of the cell cycle, while a higher concentration caused more long‐term effects on the growth of the cells and caused G2 arrest. Either concentration led to chromosomal fragments which were lost with increasing time after treatment, and chromosomal translocations which appeared stable for at least 8 days after treatment. At the low concentration, the loss or gain of a single chromosome was a common event. The higher concentration led to polyploid cells, usually containing an uneven number of chromosome 4.

Radiation-induced cell cycle delays and p53 status of early passage melanoma cell lines.

Cell cultures exposed to DNA-damaging agents such as gamma radiation respond by arresting at cell cycle checkpoints, and the p53 tumor suppressor protein is strongly implicated in this behavior. We have investigated the TP53 status and cell cycle response to ionizing radiation of a series of early passage cell lines (designated NZM1 to NZM15) previously developed from patients with metastatic melanoma. The TP53 status of each of the cell lines was determined by single-strand conformation polymorphism and DNA sequence analysis. The majority of the lines appeared to have a wild-type TP53 gene sequence, consistent with published studies. Two lines (NZM4 and NZM7.2) were found to have an identical T-->C transition mutation in nucleotide 721 (exon 7) of the coding region. NZM7.2 (mutant) and NZM7.4 (wild-type) were clonally derived from the same line (NZM7). The existence of radiation-induced cell cycle arrest in G and/or G2M phase was determined 16 h after irradiation (6.3 Gy) by DNA staining and flow cytometric analysis. The mitotic inhibitor paclitaxel was used as a reference compound, with or without irradiation, to assess the efficiency of radiation-induced cell cycle arrest. G1 phase arrest was associated only with the presence of the wild-type TP53 gene, but the efficiency of induced arrest varied among the cell lines and the period of G phase arrest appeared to be short. A significant difference (P < 0.002) was also found between the efficiency of induction of G2 phase arrest and the presence of wild-type TP53 gene. The results provide evidence that although the melanoma cell lines generally had an intact TP53 gene, the efficiency of p53-mediated cycle arrest might be deficient and contribute to the resistance of this tumor to treatment.

Cytokinetic differences in the action ofN-[2-(dimethylamino)ethyl]acridine-4-carboxamide as compared with that of amsacrine and doxorubicin

N-[2-(Dimethylamino)ethyl]acridine-4-carboxamide dihydrochloride (DACA) is a topoisomerase II-directed DNA intercalator with high experimental solid-tumour activity. The effect of DACA on the cytokinetics of cultured Lewis lung adenocarcinoma cells was compared with those of two clinical drugs of this class, doxorubicin and amsacrine. Cells were exposed to drugs for a 1-h period at concentrations that reduced viability by approximately 99% as measured by clonogenic assays. Subsequent progress through the cell cycle was monitored by propidium staining of fixed cells and flow cytometry. DACA, amsacrine and doxorubicin did not inhibit the G1-to S-phase transition but did delay progression through the S-phase. The effect was maximal in the late S-phase and, because of the differential rates of progress of cells in various cycle positions, led to the development of a synchronous S-phase peak. This peak moved to the G2/M-phase position at 11 h after the removal of DACA or at 14 h after the removal of amsacrine or doxorubicin. The effects of the drugs on cells initially in the G2-phase was measured by scoring mitotic cells in the presence and absence of colchicine. DACA had an immediate inhibitory effect on the progression of cells from the G2-phase to mitosis. This effect was much greater for DACA than for the other two drugs, consistent with the greater effect of DACA on the G2/M-phase to G1-phase transition. The results suggest that DACA causes cell-cycle changes expected for a DNA-damaging drug but differs from doxorubicin and amsacrine mainly by its effect on the transition of G2-phase cells to mitosis and the G1-phase.

Relationship of cell cycle parameters to in vitro and in vivo chemosensitivity for a series of Lewis lung carcinoma lines

The sensitivity of three Lewis lung carcinoma sublines, which grow in culture and in vivo, and vary in in vivo drug sensitivity, have been compared using topoisomerase II poisons amsacrine, amsacrine analogue CI-921, doxorubicin and etoposide. D10 (drug concentration for 10% clonogenic survival) values were determined in vitro for low and high density cultures, and ex vivo for cells from subcutaneous tumours. The cytokinetic parameters of these populations were obtained by flow cytometric analysis of bromodeoxyuridine-labelled cells. Regression analysis showed that logarithmic D10 values were significantly correlated (r greater than 0.95) with G1- and S-phase proportions and highly correlated (r = 0.99) with calculated G1 transit times. The slopes of the regression lines were similar for all topoisomerase II poisons tested and it is suggested that this slope reflects the disappearance of topoisomerase II during G1 phase.

Novel carbamate analogues of amsacrine with activity against non-cycling murine and human tumour cells

The cytotoxicity of a class of compounds related to the topoisomerase-II poison amsacrine was investigated against plateau-phase murine Lewis lung carcinoma cells (LLTC), HCT-8 human colon carcinoma cells and other cell lines. MethylN-[4-(9-acridinylamino)-2-methoxy-phenyl]carbamate hydrochloride and the corresponding demethoxy compound, which contain a methylcarbamate instead of the methylsulphonylamino group, manifested relatively high cytotoxic activity against plateau-phase cells as measured by clonogenic survival. The concentration of drug required for a given cytotoxic effect on plateau-phase cells was about 2 times higher than that required for an equitoxic effect on actively proliferating cells. In contrast, at least 5 times more amsacrine, doxorubicin or etoposide was needed for an equitoxic effect on plateau-phase cells. Cells taken directly from subcutaneous LLTC tumours and exposed to drugs displayed the same differential drug sensitivity to the carbamate compounds, suggesting that the plateau-phase cells provide an appropriate model for cells growing in vivo. The greater cytotoxicity of the carbamate drugs was shown to depend critically on the provision of an energy source such as glucose, suggesting that nutrient starvation both in plateau-phase cells and in tumours induced a glucose-sensitive resistance mechanism. It is suggested that the carbamate analogues of amsacrine recognize a form of topoisomerase II, possibly topoisomerase IIβ, the activity of which increases relative to that of topoisomerase IIα in non-cycling cells, and might be used to devise new strategies for the treatment of solid tumours.

Cellular responses to methyl-N-[4-(9-acridinylamino)-2-methoxyphenyl]carbamate hydrochloride, an analogue of amsacrine active against non-proliferating cells

The acridine derivative m-AMCA (methyl-N-[4-(9-acridinylamino)-2-methoxyphenyl]carbamate hydrochloride), a carbamate analogue of the topoisomerase II poison amsacrine, is distinguished by its high cytotoxicity against non-cycling tumour cells. We compared the response of cultured Lewis lung carcinoma cells to m-AMCA, amsacrine and the topoisomerase I poison camptothecin. The DNA polymerase inhibitor aphidicolin reversed the cytotoxicity of camptothecin fully, that of amsacrine partially, and that of m-AMCA minimally. The ability of m-AMCA to induce the enzyme poly(ADP-ribose)polymerase (PARP) was markedly lower than that of camptothecin or amsacrine. Cell cycle responses to m-AMCA and amsacrine were similar, with slowing of progress through S-phase and arrest in G2-phase. These cell cycle changes were also observed when plateau phase cultures were exposed to drug for 1 h, washed free of drug and cultured in fresh medium, with m-AMCA having a more pronounced effect than amsacrine and camptothecin having no effect. We also examined the role of p53 protein in the response using cultured human H460 cells. Both m-AMCA and amsacrine induced p53 protein expression in proliferating but not in non-proliferating H460 cells, and induced p21WAF1 regardless of proliferation status. Both induced G1-phase cell cycle arrest. It is suggested that two cytotoxicity mechanisms can be distinguished using these drugs. The first is specific for S-phase cells, is reversed by aphidicolin and induces PARP activity. The second is cell cycle non-specific, does not induce PARP and is unaffected by aphidicolin. Camptothecin activates only the first, m-AMCA primarily the second and amsacrine activates both.

Modulation of Protein-DNA Interactions by Intercalating and Nonintercalating Agents

The cytotoxicity of a series of DNA-binding drugs was measured in a number of both mouse and human cell lines and some multidrug resistant sublines. The intercalating drugs used were derivatives of 9-anilinoacridine, including amsacrine. The nonintercalating drugs were phenylbisbenzimidazole derivatives, including the Hoechst dyes H33258 and H33342. Comparison of the ability of the intercalators to inhibit the growth of cultured L1210 cells showed that the presence of certain substituents on the anilino moiety, such as a methoxy group or a benzenesulphonamide group in the correct place, greatly enhances the degree of growth inhibition for a given amount of DNA binding. Divergence between activity and DNA binding was also found with two multidrug resistant Jurkat sublines but in this case the anilino substituents for optimal activity were quite different. The Hoechst dyes at low concentrations did not inhibit cell growth, but some did strongly potentiate the cytotoxicity of the amsacrine analogue CI-921. In this case, potentiation was strongly dependent on the substituents on the terminal phenyl ring. The results can be explained if it is assumed that both the anilino substituents of the intercalators and the phenyl substituents of the Hoechst dyes interact with an enzyme receptor while the remainder of the molecule binds to DNA. The most likely enzyme receptors are isozymes of the topoisomerase II isozymes.

Growth of Fetal Thymocytes in Organ Culture: Effect of Recombinant Lymphokines on Thymocyte Maturation

Interleukin 2 (IL2) has a dose-dependent inhibitory effect on the growth and phenotypic maturation of thymocyte populations grown in fetal thymus organ culture. Addition of IL2 (100 U/ml) to 14-day fetal thymus organ cultures induces the appearance of a population of lymphokine-activated killer (LAK) cells which lyse allogeneic, syngeneic, and syngeneic tumor cell targets. The addition of the monoclonal antibody, PC-61, blocks the IL2-dependent growth and activation of LAK cells but does not influence the maturation of CD4+ CD8+ fetal thymocytes. These data imply that IL2 is not a major regulator of normal fetal thymocyte maturation. The effects of a range of recombinant lymphokines (IL1 alpha, IL1 beta, IL3, IL4, GM-CSF, G-CSF, M-CSF) on the proliferation and phenotypic maturation of fetal 14-day thymocytes in organ culture has been analysed. Two significant changes were seen. First, IL1 alpha and IL1 beta inhibited growth and the expression of the CD4 and CD8 antigens in organ culture, and second, GM-CSF increased the expression of Mac-1+ cells. IL4, which has known T cell growth-promoting activity, IL3, G-CSF, and M-CSF did not alter either normal growth or surface antigen expression in fetal thymocytes. While some of these lymphokines may function as accessory molecules in fetal thymocyte development, our experiments suggest that they do not have a significant influence on thymocyte maturation when used alone.