Stable quantitative differences of antigen expression in human melanoma cells isolated by flow cytometric cell sorting

Changes in the immunologic phenotype of human malignant glioma cells after passaging in vitro

Although immunotherapeutic strategies against glioblastomas have been promising both in vitro and in animal models, similar successes have not been realized in human clinical trials. One reason may be that immunotherapeutic strategies are based on prior studies that primarily have used human glioblastoma cell lines passaged in vitro, which may not accurately reflect the in vivo properties of glioblastoma cells. In this report, we used flow cytometry to quantify the expression of immunological cell surface molecules on human glioblastomas directly ex vivo (prior to any in vitro culturing) and after varying passages in vitro. Furthermore, we used ELISA to quantitate cytokine secretion after various passages in vitro. We demonstrate that in vitro culturing of established cell lines led to increases in the cell surface expression of MHC class I and ICAM-1 and secretion of IL-6 and TGF-beta(2). Furthermore, there were significant changes in the expression of MHC class I, MHC class II, B7-2, ICAM-1, and FasL when comparing ex vivo tumor cells to those after a single passage in vitro. After passaging once in vitro, there were also significant changes in the secretion of TGF-beta(2) and IL-10. This report indicates that in vitro culturing leads to significant changes in both cell surface molecules and secreted cytokines, which are known to affect the ability of immune cells to initiate an anti-tumor immune response. These changes in the immunological phenotype of glioblastomas after in vitro culturing may in part explain the limited success of immunotherapeutic strategies against glioblastomas in human clinical trials.

Intermediate steps in cellular iron uptake from transferrin. II. A cytoplasmic pool of iron is released from cultured cells via temperature-dependent mechanical wounding

A previous study described a cytoplasmic, transferrin (Tf)-free, iron (Fe) pool that was detected only when cells were mechanically detached from the culture substratum at 4 degrees C, after initial incubation with 59Fe-125I-Tf at 37 degrees C (Richardson and Baker, 1992a). The release of this internalized 59Fe could be markedly reduced if the cells were treated with proteases or incubated at 37 degrees C prior to detachment. The present study was designed to characterize this Fe pool and understand the mechanism of its release. The results show that cellular 59Fe release increased linearly as a function of preincubation time with 59Fe-Tf subsequent to mechanical detachment at 4 degrees C using a spatula. These data suggest that the 59Fe release was largely composed of end product(s) and was not an "intermediate Fe pool." When the Fe(II) chelator, dipyridyl (DP), was incubated with 59Fe-Tf and the cells, it prevented the accumulation of 59Fe that was released following mechanical detachment at 4 degrees C. Other chelators had much less effect on the proportion of 59Fe released. Examination of the 59Fe released showed that after a 4-h preincubation with 59Fe-Tf, approximately 50% of the 59Fe was present in ferritin. These data indicate that mechanical detachment of cells at 4 degrees C resulted in membrane disruptions that allow the release of high M(r), molecules. Moreover, electron microscopy studies showed that detachment of cells from the substratum at 4 degrees C resulted in pronounced membrane damage. In contrast, when cells were detached at 37 degrees C, or at 4 degrees C after treatment with pronase, membrane damage was minimal or not apparent. These results may imply that temperature-dependent processes prevent the release of intracellular contents on membrane wounding, or alternatively, prevent wounding at 37 degrees C. The evidence also indicates that caution is required when interpreting data from experiments where cells have been mechanically detached at 4 degrees C.

Expression of melanoma-associated antigen of thermotolerant human cells

The expression of the melanoma-associated antigen p250 on thermotolerant cells and the effect of a second heat dose on the antigen expression have been measured by flow cytometry. The human melanoma cell line FME was heated at 43.5 degrees C for 120 min after a priming heat dose at 43.5 degrees C for 20, 40 or 60 min. Cells preheated at 43.5 degrees C for 40 and 60 min followed the same kinetics of development and decay of thermotolerance, with maximum thermotolerance 16 h after the priming heating, and the thermotolerance had decayed by 48 h. Cells preheated at 43.5 degrees C for 20 min showed maximum thermotolerance after 7 h and decay by 24 h. Heat reduced the expression of the melanoma-associated antigen in a dose-dependent manner. Thermotolerant cells were given a second heat dose (43.5 degrees C for 120 min) and the antigen expression measured immediately after heating. Fractionated hyperthermia using the lower predose (43.5 degrees C for 20 min) might have an additive effect on the reduction of antigen expression, while the highest predose (43.5 degrees C for 60 min) protected against reduction in antigen expression. The development and decay of resistance against heat-induced reduction in expression of melanoma-associated antigen followed a similar time course as thermotolerance in terms of cell survival. Maximum resistance was observed 12 h after the priming heat treatment, and the resistance had decayed by 48 h.

Mycobacterial induction of activated killer cells: possible role of tyrosine kinase activity in interleukin-2 receptor alpha expression

Mycobacterium avium is an intracellular opportunistic pathogen commonly seen in AIDS patients. M. avium-infected monocytes have been recently shown to be lysed by interleukin-2 (IL-2)-activated killer cells. Since some bacterial products can directly augment natural killer activity, we examined the ability of these microorganisms to induce killer cell activity. Coculture of M. avium with large granular lymphocytes (LGL) was found to augment the ability of LGL to lyse both tumor cells and bacterially infected autologous monocytes. The induction of tumoricidal activity by M. avium was only partially neutralized by the presence of anti-IL-2 antibodies, indicating that both IL-2-dependent and IL-2-independent mechanisms are responsible for activation of killer cells. Furthermore, only the direct interaction between bacterium and LGL could induce the expression of both IL-2 receptor alpha protein and mRNA, an effect which was abrogated by the presence of genistein, a tyrosine kinase inhibitor. Thus, M. avium was seen to induce killer cells, an activity that is concomitant with the up-regulation of IL-2 receptor alpha, or Tac antigen, expression and which involves signal transduction mechanisms mediated by tyrosine kinase activity.

The uptake of inorganic iron complexes by human melanoma cells

The human melanoma cell line, SK-MEL-28, expresses high levels of melanotransferrin. The uptake of inorganic iron (Fe) complexes compared to transferrin-bound Fe by these cells has been investigated to determine whether melanotransferrin has a role in Fe uptake. The mechanisms of Fe uptake have been characterised using 59Fe complexes of citrate, nitrilotriacetate, desferrioxamine, and 59Fe added to Eagle's minimum essential medium (MEM) and compared with human transferrin (Tf) labelled with 59Fe and iodine-125. Iron uptake from the Fe complexes of citrate, nitrilotriacetate and MEM were similar, and far greater than that from Tf at the same Fe concentration (2.5 microM). Ammonium chloride and a monoclonal antibody to the transferrin receptor (42/6), had no effect on the uptake of Fe from inorganic Fe complexes, suggesting that receptor-mediated endocytosis of Tf was not involved. The monoclonal antibody, 96.5, specific for melanotransferrin did not alter total Fe uptake but slightly increased the proportion of Fe internalised, possibly due to the modulation of the antigen by the antibody. However, from the time required for modulation to occur (approximately 2 h), the small increase in internalisation observed and the fact that no increase in total cell Fe occurred, it is suggested that melanotransferrin has little role in Fe uptake.

The release of iron and transferrin from the human melanoma cell

The role of the transferrin homologue, melanotransferrin (p97), in iron metabolism has been studied using the human melanoma cell line, SK-MEL-28, which expresses this antigen in high concentrations. The release of iron and transferrin were studied after prelabelling cells with human transferrin doubly labelled with iron-59 and iodine-125. Approx. 45% of internalised iron was in ferritin with little redistribution during reincubation. Iron release was linear with time, while transferrin release was biphasic, suggesting that iron was leaving the cell independently of transferrin. Unlabelled diferric transferrin increased transferrin release, implying a degree of coupling between cell surface binding, internalisation and release of transferrin. Increasing the preincubation time increased the amount of transferrin which remained internalised within the cell. A membrane-bound, iron-binding component with properties consistent with melanotransferrin was observed. Desferrioxamine or pyridoxal isonicotinoyl hydrazone could not remove iron from this compartment, suggesting a high affinity for iron. The number of membrane iron-binding molecules per cell was estimated to be 387,000 +/- 7000 . The non-transferrin-bound membrane Fe did not decrease during reincubation periods up to 5 h, suggesting that the cell was not utilising it. Hence, melanotransferrin may not have a role in internalising iron in melanoma cells.

Hyperthermia-induced shedding and masking of melanoma-associated antigen

Hyperthermia has been reported to induce a dose-dependent reduction in the expression of melanoma-associated surface antigens. The objective of the present work was to study the mechanisms for the reduction in the expression of the p250 antigen recognized by the monoclonal antibody 9.2.27. Measurements at 37 degrees C showed that antibody binding induced a certain degree of modulation (internalization) of the melanoma-associated antigen. Masking of the antigen due to internalization and/or damage in situ, as well as shedding of the antigen, were measured after hyperthermia, and found to increase in a heat-dose-dependent manner, although for antigen masked this increase was not significant compared with control cells at 37 degrees C. The sum of antigen shed and masked after hyperthermia correlated with the overall reduction in antigen expression measured independently. During hyperthermia, antigen was shed and masked in approximately equal amounts. After the treatment, hyperthermia-induced shedding continued as a function of time and caused a further reduction in antigen expression, but masking did not differ from 37 degrees C controls.

Changes in antigen expression on human FME melanoma cells after exposure to hypoxia and acidic pH, alone or in combination

Quantitative changes in the expression of two tumour-associated surface antigens on human FME melanoma cells were studied by flow cytometry after exposure to hypoxia and acidic pH, either alone or in combination. The expression of the p250 antigen recognized by the monoclonal antibody (MAb) 9.2.27 was reduced immediately after exposure to hypoxia. The magnitude and duration of the reduction increased with increasing exposure time. Twelve to 16 hr after the end of a 6-hr exposure to hypoxia the antigen expression reached the normal level, followed by a temporary increase above this level. The p97a antigen recognized by the 4.1 MAb underwent similar changes after exposure to hypoxia for 6 hr. After exposure to hypoxia in acidic environment, the magnitude and duration of the reduction in the expression of the p250 antigen increased with increasing acidity. The enhancement in antigen expression above the normal level was less after hypoxia at acidic pH than after hypoxia at physiological pH. The combined treatment had an additive effect on the expression of the melanoma-associated antigen but did not enhance hypoxia-induced cell killing. The observed changes in antigen expression might be of importance if hypoxic tumour cells are subjected to MAbs conjugated to radioisotopes or cytotoxic agents for diagnostic or therapeutic purposes.

Relationship between changes in antigen expression and protein synthesis in human melanoma cells after hyperthermia and photodynamic treatment

Hyperthermia and photoactivated hematoporphyrin derivative induce a dose-dependent reduction in the expression of the p250 surface melanoma-associated antigen on the human FME cell line. Expression of this glycoprotein antigen was quantitated by immunofluorescence flow cytometry based on the monoclonal antibody 9.2.27. Decrease in antigen expression was followed by a transient increase above the level for untreated cells, before normalization occurred about one week after treatment. These treatment-induced changes in antigen expression could partly be explained by changes in protein synthesis. This conclusion was based on the following observations: Hyperthermia and photoactivated hematoporphyrin derivative both inhibited protein synthesis. The latter increased again rapidly to rates above normal until antigen expression reached normal level, whereupon the protein synthesis rate decreased to normal. Inhibition of protein synthesis by cycloheximide 1 day after heating, prevented the recovery of antigen expression, demonstrating that protein synthesis is necessary for resumption of normal antigen expression. The changes in both antigen expression and protein synthesis were dose-dependent, and the magnitude and duration of the changes increased with increasing dose. The time courses of the changes in protein synthesis after two different treatments which both inactivated two logs of cells were almost identical, as were the time courses after two lower heat doses inactivating one log of cells. These similarities were reflected in the changes in antigen expression. At the same time as protein synthesis reached its maximum and antigen expression resumed normal level, an increase in the Golgi apparatus was observed ultrastructurally, indicating an increased synthesis rate and transportation of glycoproteins to the cell surface.

Cellular heterogeneity in human epithelial neoplasms

Tumor cell heterogeneity has in recent years been the subject of numerous excellent review articles, but comprehensive reviews may not always distinguish between that which is known about tumors from direct observation and that which is inferred from the study of analagous systems. The purpose of this review is to describe what is known about cellular heterogeneity in human tumors and to discuss current models of the pathogenesis of cellular heterogeneity in light of the evidence available from the study of human cancer.

Human melanoma-associated antigens: analysis of antigenic heterogeneity by molecular, serologic and flow-cytometric approaches

The relationship of antigenic heterogeneity to the epitope recognized by an antibody was examined with monoclonal antibodies to human melanoma-associated antigens. Expression of the human melanoma-associated antigens, 250-Kd glycoprotein/proteoglycan and p97, was examined quantitatively by flow cytometry on fresh cell suspensions of human melanoma. Percent positive cells and mean fluorescence intensity were consistently higher with antibody 9.2.27 to the 250-Kd glycoprotein/proteoglycan than with antibody to p97. In addition, assessment of percent positive cells in multiple skin lesions biopsied from individual patients indicated that in 26 of 30 lesions, greater than 90% of the cells stained positively with 9.2.27. This relative lack of antigenic heterogeneity with antibody 9.2.27 contrasted with previous reports which showed considerable antigenic heterogeneity with other antibodies to the 250-Kd glycoprotein/proteoglycan. The explanation for this distinction was sought by quantitative flow cytometric and sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) techniques. Comparison by flow cytometry and immunoperoxidase of three antibodies, which recognized distinct epitopes of the 250-Kd glycoprotein/proteoglycan, indicated that 9.2.27 reacted more intensely with cultured cells and tissue sections than other antibodies to the same antigen. Examination by SDS-PAGE indicated that 9.2.27 could immunoprecipitate a larger proportion of 250-Kd glycoprotein molecules than other antibodies. In addition, immunodepletion experiments in gels indicated that the 9.2.27 determinant was present on a higher proportion of 250-Kd glycoprotein molecules than PG-2 antibody to a separate determinant. It is likely that 9.2.27 antibody displays less antigenic heterogeneity because its epitope is represented on a higher proportion of the antigen molecules. Thus, not only the nature of the antigen but also the epitope recognized by an antibody influences the degree of antigenic heterogeneity.

Cell cycle analysis using flow cytometry

This manuscript reviews the utility of flow cytometry for the study of cell proliferation. The applications of univariate DNA distribution analysis to cytokinetic studies of asynchronous and perturbed cell populations are discussed briefly. The newly developed technique for simultaneous flow cytometric measurement of cellular DNA content and amount of incorporated bromodeoxyuridine is discussed in more detail. The cytochemistry required for this analysis is reviewed as are its applications to: determination of the fractions of cells in the G1-, S- and G2 + M-phases of the cell cycle; determination of the G1-, S- and G2 + M phase durations and dispersions and growth fraction for asynchronous cells; detection of ara-C resistant cells present at low frequency in an otherwise sensitive population; and analysis of the cytokinetic response of a solid murine tumour to treatment in vivo with a cell cycle specific agent.