A semi-automated neutral red based chemosensitivity assay for drug screening.

A semi-automated colorimetric chemosensitivity assay was developed. The assay utilizes the vital stain neutral red for the rapid screening of potential anticancer agents using solid tumor cell lines. The cell lines used in this assay and presented in this report are CX-1 colon adenocarcinoma and A549 lung carcinoma. The assay, performed in 96 well tissue culture plates, allows for short drug exposure times (3 hrs.) followed by quantitation of cell number (neutral red absorbance) following four cell doubling times. Cell number directly correlated with absorbance of eluted neutral red at 540 nm. However, optimal amounts of dye and staining times varied between cell lines. IC50 concentrations (for inhibition of cell growth) determined using this assay were in good agreement with results from clonogenic assays using similar drug treatment conditions. The assay technique was determined to be capable of detecting antineoplastic compounds operating by a wide variety of mechanisms.

Failure of human myeloma cells to heterotransplant into the brain of athymic rats.

Sixty-four separate intracranial inoculations of bone marrow cells obtained from 26 patients with human myeloma were performed and the animals were kept under observation for 9-10 months. Samples were obtained from a heterogenous group of patients with diverse types of paraprotein production, clinical status, and response to treatment. Inocula size ranged from 3.5 x 10(5) nucleated cells to about 2 x 10(7), while the percentage of plasma cells varied from nondetectable to 90%. Only one animal (of 2) injected with an aliquot of the bone marrow aspirate from a patient developed a small, clinically undetectable tumor, noticed at the end of the observation period. No other animal developed tumors. Thus, our studies indicate that the intracerebral inoculation of human myeloma cells may not be a profitable means of establishing additional human myeloma cell lines.

Expression of transferrin receptors is unrelated to proliferative status in cultured human colon cancer cells.

Transferrin is an iron-carrying compound that stimulates cell growth and division by binding to specific receptors (TR) which are preferentially expressed by actively growing cells or by the malignant counterpart of normal cells. However, quiescent cells may not necessarily cease expressing TR. We evaluated TR expression by flow cytometric analysis utilizing a monoclonal antibody (OKT-9) specific for TR on six established human colon cancer cell lines with distinct degrees of phenotypic differentiation and growth rates at sequential stages of in vitro growth (exponential and stationary phase). There were no significant differences in the proportion of cells expressing TR among the fast and slow growing cell lines at any time point of the study, nor did the cultures change the proportion of TR positive cells in their transit from exponential into stationary phase of growth. Hence, direct measurements of TR expression in malignant cell populations may not provide a useful clinical marker to distinguish highly proliferative tumors from those with a slower growth resulting from a larger proportion of quiescent cells.

Heterogeneous expression of cytokeratins in metastatic mammary adenocarcinoma cells in vitro and in vivo.

The expression of cytokeratins was investigated in rat 13762NF adenocarcinoma cell lines and clones growing in vitro and in vivo. The anti-cytokeratin monoclonal antibody PKK1 used in this study recognized one cytokeratin with a molecular weight of 54,000 in these cells. Immunofluorescence staining of cultured tumor cells with PKK1 detected cytokeratins in less than 1% of the tumor cells isolated from the locally growing parental tumors, but intense cytokeratin staining in the form of fibrils and networks was found in the cell cultures derived from spontaneous metastases. There were variable degrees of staining with the anti-cytokeratin antibody in cultured cells from metastases, indicating phenotypic diversity in the expression of cytokeratins. Staining with PKK1 of frozen sections from locally growing tumors obtained after injecting the cells into the mammary fat pads of syngeneic rats also demonstrated considerable heterogeneity in cytokeratin expression. The anti-cytokeratin reagent stained only a few, isolated cells in frozen sections of tumors established from parental tumor cells, while this reagent stained intensely cells in local tumors and their metastases established from tumor cells cultured from spontaneous lymph node or lung metastases.

Successful heterotransplantation of human colon cancer cells to athymic animals is related to tumor cell differentiation and growth kinetics and to host natural killer cell activity.

Six established human colon carcinoma cell lines with distinct degrees of cell differentiation were inoculated into infant (less than 4 weeks) and adult (greater than 8 weeks) nude rats. The most differentiated tumor cells (group I) had nearly a 100% rate of tumor takes whether inoculated subcutaneously, intraperitoneally, or intracerebrally into adult rats; subcutaneous growth continued unabated for a 120-day observation period. Cells with intermediate differentiation (group II) had nearly an 80% incidence in tumor takes when injected subcutaneously and 14-60% when injected intraperitoneally. Subcutaneous growth continued only for about 30 days, after which time growth declined, and tumors regressed completely. Intracerebral inoculations of group II cells resulted in 64-83% tumor takes. Subcutaneous injections of cells from groups I and II into 5- to 10-day-old rats resulted in 100% tumor takes; tumors induced by group II did not regress, and after about 60 days reached volumes comparable to those originated by cells from group I. No tumors developed when cells from group III (undifferentiated) were injected either subcutaneously or intraperitoneally (and even intravenously) into adult rats. Only when the intracerebral route was employed was there a 60-71% incidence of tumor takes. Also, for one of the cell lines in this group, subcutaneous injection into infant rats resulted in 100% tumor takes. NK cell activity of infant rats against all of the colon cells (measured by the 51Cr release assay) was negligible; in adult rats, the activity varied according to the cell type, being usually highest against the less differentiated tumors. Our data on the incidence of tumor takes, and on the dynamics of tumor growth and decline suggest that successful heterotransplantation of human colon carcinoma cells into nude rats depends on the activity of host NK cells. In turn, this activity seems related to the degree of cell differentiation and the growth kinetics of the xenografted tumor cells. These observations highlight important differences in biological characteristics of human colon carcinoma with important implications for their intrinsic ability to grow and metastasize, and, possibly, their response to biological response modifiers.

Mixed function oxidase activities of established human colon carcinoma cell lines in the activation of cyclophosphamide.

Three established human colon carcinoma cell lines (LoVo, SW620, and SW403) with different degrees of phenotype differentiation were investigated for their sensitivity to the cytotoxic effects of cyclophosphamide (CP) and to its active metabolite, 4-hydroxycyclophosphamide (4-OH-CP), and for their mixed function oxidase (MFO) activities. None of the cell lines showed sensitivity to CP as determined by the inhibition of colony formation assay, even after continuous drug treatment at high concentrations (200 microgram/ml) for up to 72 h. CP also had no effect on the cellular doubling time or on the incorporation of [3H]-thymidine. Pretreatment with phenobarbital (PB) plus hydrocortisone (HC) was unable to induce CP cytotoxicity. In contrast, 4-OH-CP, the major metabolite formed from CP by MFO, was highly toxic to the cells. About 90% cell kill was obtained at drug concentrations of 17.5 microgram/ml (LoVo), 15 microgram/ml (SW620), and 55 microgram/ml (SW403) after 1-h incubation at 37 degrees. MFO activities were determined by measuring p-nitroanisole demethylase (PNAD) and arylhydrocarbon hydroxylase (AHH) in microsomes prepared from noninduced cells or from cells treated with benzanthracene or PB plus HC. Intrinsic AHH activities were below the level of detection for all cell lines [less than 1 pmol of 3-hydroxybenzo(a)pyrene (3-OH-BP) formed per min per mg of protein]. Treatment with benzanthracene resulted in AHH activities of 12 to 15 pmol of 3-OH-BP per min per mg of protein, but treatment with PB plus HC failed to induce significant AHH activities. PNAD activities in noninduced cells as well as in cells treated with benzanthracene were 0.05 to 0.08 nmol of p-nitrophenol formed per min per mg of protein; treatment with PB plus HC increased PNAD activities by only 1.5-fold. Thus, in contrast to reports for rat colon and for a single human colon cancer cell line, CP is inactive when applied directly to several other human colon carcinoma cell lines. Because these cells have minimally detectable intrinsic and induced MFO activities, we conclude that CP cannot be successfully metabolized into 4-OH-CP to induce a significant degree of cell kill.