Inhibition of carcinogen-altered rat tracheal epithelial cells by normal epithelial cell-conditioned medium.

There is mounting evidence that normal cells can either inhibit the growth of carcinogen-altered cells and/or affect progression to a neoplastic phenotype. This effect(s) has been observed both in vivo in intact rat tracheal tissues and in rat tracheal epithelial cell cultures. The inhibition of carcinogen-altered cells in culture appears to be associated with the production of an acid and heat stable, dithiothreitol sensitive, nondialyzable protein produced by normal tracheal epithelial cells or esophageal epithelial cells in primary culture. It was found to be optimally produced by 3-4-week-old cultures of normal epithelial cells. In the presence of a 1:4 dilution of normal cell conditioned medium, the colony forming efficiency of a sensitive cell line is decreased 5-fold. Biochemical properties of the inhibitor are similar to those associated with type beta transforming growth factor.

Multiple changes in gene expression are associated with normal cell-induced modulation of the neoplastic phenotype.

Specific regulatory pathways in neoplastic cells seem to be responsive to control signals provided by the normal cell/tissue environment. The present experiments were designed to define, at the molecular level, the growth-regulatory signals in neoplastic cells that are associated with the modulation of expression of the neoplastic phenotype by normal cell populations. When cultured in the presence of normal cell-conditioned medium, a highly malignant rat tracheal carcinoma-derived cell population (IC-12) undergoes dramatic changes in morphology, and the anchorage-independent growth of these cells is inhibited. This phenomenon is termed normalization. The strategy adopted for elucidating the cellular/molecular changes associated with the induction of these phenotypic alterations was to define the differences in mRNA expression patterns between IC-12 populations exhibiting the neoplastic phenotype (wild-type cells) and those exhibiting the normalized phenotype. For this purpose, the differential display technique and subsequent Northern blot analyses were used. Once specific, differentially expressed genes were identified, the temporal sequence of altered gene expression was determined by monitoring the levels of mRNA expression after the addition of normal cell-conditioned medium. Some of the identified known genes are grouped into three general categories: (a) group I genes are those involved in cellular adhesion processes; (b) group II genes are those involved in signal transduction pathways; and (c) group III genes are those involved in transcriptional and translational processes. Genes that are differentially expressed during the normalization process seemed to exhibit characteristic temporal expression patterns after the addition of normal cell-conditioned medium. Identification of these differentially expressed genes and their associated cellular functions provide insight into some of those regulatory pathways in neoplastic cells that are amenable to regulation by normal cells. An analysis of the temporal sequence of altered gene expression provides further information that allows the identification of those genes that are likely to be critical upstream effectors regulating transcriptional regulatory events that result in the moderation of neoplastic behavior.

Interactions between cell populations influence expression of the transformed phenotype in irradiated rat tracheal epithelial cells.

A combined in vivo-in vitro model has been utilized to evaluate the influence of cell-cell interactions on expression of radiation-induced transformation in irradiated rat tracheal epithelial cells. Two types of cell interactions are evaluated. One type of interaction appears to involve direct cell-cell contact as occurs within the intact tissue. The other type of communication appears to be mediated via a diffusible factor, most likely transforming growth factor type-beta. The data suggest the need to be cautious when extrapolating from transformation data obtained following irradiation of low-density rapidly proliferating cell cultures to the in vivo model.

Induction of preneoplastic alterations by X rays and neutrons in exposed rat tracheas and isolated tracheal epithelial cells.

The relative potential of high- and low-LET radiation to induce preneoplastic alterations in rat tracheal epithelial cells was evaluated using a combined in vivo-cell culture model. The capacity of X rays and high- and low-dose-rate neutrons to induce preneoplastic changes in isolated rat tracheal epithelial cells and in the intact tissue was compared. The presence of altered populations was determined in culture in terms of the frequency of tracheal epithelial cell populations which exhibit enhanced growth capacity in culture and in terms of the induction of persistent morphological alterations in exposed transplanted tracheas. Prior to assaying for altered cells, tracheal epithelial cells were irradiated as part of the intact tissue or as single cells. Low- and high-LET radiation induced similar maximum frequencies of altered cells when cultures were exposed as single cells, although high-LET radiation was more radiobiologically effective (RBE = 20) than low-LET radiation. The most marked difference between high- and low-LET radiation was observed after irradiation of the intact tissue. Damage induced by low-LET radiation, giving rise to altered populations, was modified in the intact tissue, whereas similar damage induced by high-LET radiation was apparently not.

Effects of 210Po alpha particles on survival and preneoplastic transformation of primary rat tracheal epithelial cells irradiated while in suspension or in the intact tissue.

Rat tracheal epithelial cells exhibited exponential cell killing when exposed to 210Po alpha particles as single cell suspensions or in the intact tissue. Survival of cells in the intact tissue was not significantly different from that observed with cell suspensions. Comparison of survival of cells exposed in suspension to 300 kVpX rays yielded an RBE of 6.3. Measurements of basal cell nuclei were used to determine that a single traversal of a cell nucleus had a high probability of causing cell inactivation. This was also observed in mink lung cells and CHO cells exposed in an identical manner. There were no significant increases in frequencies of preneoplastic transformation observed for a range of exposures (0.0007 to 0.05 alpha particles/micron2). Examination of intact tracheal transplants which were irradiated with alpha particles also failed to reveal any preneoplastic or neoplastic changes.

Influence of cell position relative to planar alpha-particle sources on survival and preneoplastic transformation of primary rat tracheal epithelial cells.

Rat tracheal epithelial cells exposed directly on planar 210Po sources exhibited exponential cell killing; however, no significant increase in induction of preneoplastic transformation was observed over a range of alpha-particle fluences (0.017-0.050 micron-2). In contrast, up to 10-fold increases in frequencies of preneoplastic transformants, above control levels, were observed after exposure of rat tracheal epithelial cells to similar alpha-particle fluences on 238Pu and 241Am sources. Two alternative hypotheses are evaluated as an explanation for this apparent difference in the biological effect of alpha particles emitted from different sources: (a) possible interactions between effects produced by alpha particles and by low-energy photons, which occur with 238Pu and 241Am but not with 210Po; and (b) the influence of spatial relationships between exposed cells and the surface of the planar source. The data suggest that cell-to-source spatial relationships affect both survival and transformation markedly.

Radiotoxicity of bismuth-213 bound to membranes of monolayer and spheroid cultures of tumor cells.

Monoclonal antibody 13A to murine CD44 was used to bind the alpha-particle emitter 213Bi to cell surfaces of cultured EMT-6 or Line 1 tumor cells. Data on kinetics and saturation of binding, cell shape and nuclear size were used to calculate the absorbed dose to the nuclei. Treatment of monolayer cells with [213Bi]MAb 13A produced a classical exponential survival curve with no apparent shoulder. Microdosimetry analyses indicated that 1.4-1.7 Gy produced a 37% surviving fraction (D0). Multicellular spheroids were shown to bind [213Bi]MAb 13A mainly on the outer cell layer. Relatively small amounts of activity added to the spheroids resulted in relatively large absorbed doses. The result was that 3-6-fold less added radioisotope was necessary to kill similar fractions of cells in spheroids than in monolayer cells. These data are consistent with the interpretation that the alpha particles from a single 213Bi atom bound to one cell can penetrate and kill adjacent cells. Flow cytometry was used to sort cells originating from the periphery or from the interior of spheroids. Cells from the outside of the [213Bi]MAb 13A exposed spheroids had a lower surviving fraction per administered activity than cells from the interior. Cells were killed efficiently in spheroids up to 20-30 cells in diameter. The data support the hypothesis that alpha-particle emitters should be very efficient at killing cells in micrometastases of solid tumors.

Effects of radiation on rat respiratory epithelial cells: critical target cell populations and the importance of cell-cell interactions.

The oncongenic effects of radiation on rat respiratory tissues are modulated in vivo within the intact tissue. The degree of modulation as well as the mechanism whereby modulation occurs appears to be different for different types of ionizing radiations. A combined cell culture -in vivo model is described. This model has been developed to evaluate the influence of the host and tissue environment on development and expression of the neoplastic phenotype in irradiated rat trachea. Our data indicates that the potentially oncogenic effects of neutrons, X Rays, and alpha-particles are different depending on the exposure conditions employed and the conditions under which exposed cells are maintained following exposure.

Inhibitor production by normal rat tracheal epithelial cells influences the frequency of spontaneous and X-ray-induced enhanced growth variants.

A cell culture model was used to assay for the induction of cell populations with enhanced growth capacity in culture in irradiated normal rat tracheal epithelial cells (NTEC). When cultures were maintained in minimally enriched Ham's F-12 plus 5% fetal bovine serum, it was noted that increasing the seeding density from 1 to 100 colony forming units per 60 mm dish decreased the frequency of proliferating epithelial foci (PEF) scored 5-6 weeks after seeding, from 2.5 to 0.07% in controls and from 5 to 0.9% in irradiated cultures. When low density (50 colonies per dish) control cultures were fed medium conditioned by higher density control cultures (300 colonies per dish) the observed frequency of PEF was 10-fold lower than that observed in cultures fed fresh medium or cultures fed medium conditioned by a carcinogen-induced tracheal cell line. These data suggest that some factor(s) present in conditioned medium derived from higher density normal cell cultures inhibits the emergence of many PEF. If cultures are maintained in serum-free enriched Ham's, many (17-21%) PEF were observed in both control and irradiated cultures. Medium harvested from high density cultures maintained in minimally enriched Ham's F-12 (5% FBS) was found to maximally inhibit normal cell thymidine uptake and growth as well as induce cornified envelope formation (terminal differentiation) in proliferating normal cell cultures. Medium harvested from normal cultures maintained in serum-free enriched Ham's was not inhibitory to normal cells and did not induce cornified envelope formation. Subcultured PEF-derived populations and PEF in 4 to 5-week primary cultures also did not produce inhibitor(s). In addition, subcultured PEF-derived cells were not induced to form cornified envelopes in the presence of inhibitory conditioned medium. The data suggest that the elaboration of an inhibitor(s) by normal epithelial cells affects the emergence of PEF. Irradiated cells appear to have some growth advantage over spontaneous PEF in that they are more likely to survive and form PEF under growth conditions associated with higher levels of inhibitor(s) in the medium.

Inhibition of carcinogen-altered rat tracheal epithelial cell proliferation by normal epithelial cells in vivo.

The experiments described investigate the potential influence of surrounding normal tracheal epithelial cells on the survival and growth of carcinogen-exposed epithelial cells in tracheal mucosa reconstructed from known cell mixtures. Cell mixtures containing preneoplastic or neoplastic rat tracheal epithelial cells and a small fraction of normal tracheal or esophageal epithelial cells were inoculated into the lumen of previously frozen-thawed tracheas which were then transplanted s.c. into syngeneic hosts. Within 2-3 weeks an intact tracheal mucosa was regenerated from the inoculated cells. At various times after cell inoculation and transplantation, cells were harvested from the repopulated trachea and the number of diploid normal and/or aneuploid carcinogen-exposed tracheal epithelial cells determined by flow cytometry (DNA content) and the frequency of cells with altered in vitro growth capacity determined. The data suggest that normal tracheal epithelial cells have an enhanced 'survival' capacity relative to carcinogen-exposed cell lines in the regenerated tracheal mucosa. When greater than 10(4) normal cells were inoculated with a 2- to 100-fold excess of carcinogen-exposed cell lines the regenerated epithelium was comprised almost entirely of normal-diploid-epithelial cells. Independent of the ratio of normal to altered cells, when less than 10(4) normal epithelial cells were inoculated in the mixture some carcinogen-altered cells and some normal cells were present in the regenerated epithelium. When established repopulated tracheas containing mixtures of normal and neoplastic epithelial cells were left in the animal no tumors developed. In contrast tracheas containing neoplastic cells alone or partially scraped tracheas containing comparable numbers of neoplastic cells covering a contiguous area of the submucosa, tumors developed within 2-5 weeks of cell inoculation and transplantation. These data suggest that normal tracheal epithelial cells have the capacity to modulate expression of the neoplastic phenotype (tumor development) in neoplastic populations. This effect appears to require close contact with the normal cells as inhibition is not observed when the neoplastic cells occupy a single contiguous area of the trachea.

Factors regulating the emergence of spontaneous and X-ray-induced variants in primary rat tracheal epithelial cell cultures.

A series of experiments have been carried out to identify those factors that affect the number of altered populations detected in control, nonexposed, and radiation-exposed primary cultures of rat tracheal epithelial cells. The number of colony forming cells per milliliter of culture medium and the frequency with which the culture medium is changed seemed to be the most critical factors regulating the emergence of induced and spontaneous variants. Increasing the number of cells plated so that of colony forming cells increase from 25 to 200 per ml, regardless of the dish size used, was associated with a 200-fold decline in the frequency of spontaneous variants and a 40-fold decline in X-ray-induced variants. Increasing the interval between medium changes from 3 to 7 days after the first week of culture was associated with a 10-fold decrease in the frequency of spontaneous variants. The frequency of spontaneous and induced variants is markedly less dependent on culture density at densities between 150 and 600 colony forming cells per ml. The type of medium used to establish primary cultures had little effect on the frequency of variants detected. Similarly, when assays were performed at densities in excess of 150 colony forming cells per ml the frequency of spontaneous and x-ray-induced variants was not affected by the absence of epidermal growth factor, increased levels of calcium (final concentration, 0.8 mM), or by removal of pyruvate from the selection medium.

Changes in response to, and production of, transforming growth factor type beta during neoplastic progression in cultured rat tracheal epithelial cells.

As rat tracheal epithelial cells progress from a normal to a neoplastic phenotype there are systematic changes in their ability to produce and activate latent transforming growth factor type beta (TGF-beta) as well as systematic changes in their response to this growth factor. Using a TGF-beta radioreceptor binding competition assay it was found that normal proliferating rat tracheal cells in early primary culture produced latent TGF-beta. With the emergence of terminally differentiated cell populations active TGF-beta was also detected in the conditioned medium. When normal cells were cultured under conditions allowing for continued proliferation, no active TGF-beta was detected in the conditioned medium. Colonies of proliferating epithelial cells in 4-6 week primary cultures or subculturable tracheal cell lines did not produce detectable levels of active or latent growth factor. With neoplastic progression there was likewise a change in response to active TGF-beta. Normal tracheal cells in primary culture were highly sensitive to growth-factor-induced decreases in thymidine uptake as well as to the induction of terminal differentiation. Proliferating epithelial cells in late (4-6 week) primary cultures and preneoplastic, subculturable cell lines were often as sensitive as normal cells to the growth factor-induced decline in thymidine uptake. None of these altered populations, however, was induced to differentiate (to form cornified, cross-linked envelopes) in the presence of TGF-beta.

Characteristics of magnetically separated rat tracheal epithelial cell populations.

A simple magnetic separation technique has been developed using lectins specific for two of the cell types found in the tracheal mucosa. The resulting populations of basal and secretory cells were examined for proliferative capacity in culture and in vivo. The basal cell fraction contains the cells that proliferate in culture and respond to 12-O-tetradecanoylphorbol-13-acetate. In addition, the basal cell fraction exhibited the highest proliferative capacity in vivo during the first few days after transplantation. Repopulation of inverted intestinal segments showed that only with suspensions containing a significant proportion of basal cells could a mucociliary lining be established. Segments receiving the same number of unsorted or predominantly mucous secreting cells did not repopulate in vivo. These data support the hypothesis that the basal cell is most likely the stem cell of the tracheal epithelium.

Basal cells are the progenitors of primary tracheal epithelial cell cultures.

The goal of this study was to identify the cells from the rat tracheal epithelium which attach and proliferate in primary culture. When cells isolated from tracheas by enzymatic digestion were held in suspension at 37 degrees C for several hours most of the differentiated cells died. The kinetics of this selective cell death were not dependent on the constituents of the holding medium. With time in suspension, the colony forming efficiency of the surviving cells increased two- to threefold. Comparison of the growth curves of cells held or plated directly showed no difference in the number of cells in the proliferating populations. Using two lectins, it was possible to monitor the loss of specific populations in suspension. BS1-B4 is a marker for basal cells and UEA-1 is a secretory cell marker. Only those cells that were BS1-B4 positive survived in suspension. Further, the colonies that formed in primary culture were positive for this marker. Single cell suspensions of cells were sorted by flow cytometry and a fivefold increase in the colony forming efficiency of BS1-B4 positive cells compared to that of the negative cells was observed. These findings suggest that the cells that survived in suspension and proliferated in culture originated from the basal cells of the trachea.

Analysis of differentiation antigens on normal and carcinogen-altered rat epithelial cells in vivo and in culture.

Two murine monoclonal antibodies, 3BG8 and 9BG8, which were raised against a rat tracheal squamous-cell-carcinoma cell line, recognize cell-surface antigens on normal rat squamous epithelium (skin, esophagus, vagina, and cornea) as well as on carcinogen-exposed, immortalized, rat tracheal epithelial cells. Monoclonal antibody 3BG8 binds to a 115-kilodalton cell-surface protein on undifferentiated basal cells of the epithelium, while the binding of the other antibody, 9BG8, occurs in both differentiated and undifferentiated populations of normal squamous epithelium and squamous cell carcinomas. Undifferentiated tracheal carcinomas bound only the 3BG8 antibody. No binding of either antibody was detected on normal tracheal mucociliary epithelium. Only under conditions that induce squamous differentiation of rat tracheal epithelium was binding of 3BG8 and 9BG8 detected. For reasons which are not clear at present, 9BG8 dramatically inhibits the growth of normal tracheal and esophageal cells in primary culture, whereas only 3BG8 affects the growth of carcinogen-altered tracheal cell lines. Based on antigen characterization and distribution, it is concluded that the 3BG8 and 9BG8 epitopes are localized on differentiation antigens which differ from others that have been previously described.

Emergence of undifferentiated rat tracheal cell carcinomas, but not squamous cell carcinomas, is associated with a loss of expression of E-cadherin and of gap junction communication.

A series of cells representing normal, non-tumorigenic cell lines, as well as differentiating neoplastic and undifferentiated neoplastic rat tracheal epithelial cell populations were evaluated for their ability to establish homologous and/or heterologous cell-cell gap junction communication in culture. Gap junction communication was evaluated by flow cytometric quantitation of the transfer of the fluorescent dye calcein from a donor to a recipient cell population via gap junctions. The data indicate that normal primary cultures of rat tracheal epithelial cells, as well as non-tumorigenic cell lines and squamous cell carcinomas cell populations, retain the ability to establish both homologous and heterologous gap junction communication. In all cases an average of >48% of recipient cells had acquired calcein label during a 5-h interval of co-culture of donor and recipient cells at confluent densities. Cells harvested directly from squamous cell carcinoma tumors exhibited similar levels of cell-cell communication. In contrast, cells giving rise to undifferentiated carcinomas, as well as cells harvested from undifferentiated carcinomas, exhibited very low levels or no homologous or heterologous cell-cell communication. Cell populations exhibiting distinctly different communication phenotypes were evaluated by Northern blot analysis for expression of connexins (Cx 26, 32 and 43) and E-cadherin. Neither communicating nor non-communicating cells expressed connexin 32. Those cell populations, which established functional gap junctions, expressed E-cadherin as well as connexin 26 and/or 43. In contrast, those cell populations that lacked the ability to communicate universally lacked expression of E-cadherin, and a quarter also lacked expression of detectable levels of connexin.

Receptor recognition by histidine 16 of human epidermal growth factor via hydrogen-bond donor/acceptor interactions.

Human epidermal growth factor (hEGF) and human transforming growth factor alpha (hTGFalpha) are prototypical of structurally related polypeptide mitogens which interact with the epidermal growth factor receptor (EGFR). Several determinants of receptor recognition that specify function have been proposed on the basis of structural criteria. This study evaluates the role of one such candidate, H16 of hEGF, by site-specific mutagenesis. When assayed for receptor tyrosine kinase stimulation using (Glu4,Tyr1)n as the exogenous substrate in vitro, the relative agonist activities of position 16 mutants range from 14-263% of wild-type hEGF. The rank order of potency was found to correlate with the relative receptor binding affinities of the mutants, which range from 7-272% of wild-type, as determined by radioreceptor competition assays. The mitogenic activity of the H16 mutants is similar to that of wild-type hEGF as determined by clonogenic assays using rat tracheal epithelial cells. While the colony forming efficiencies do not reflect significant differences in growth rate or survival characteristics in the presence of the hEGF variants, it is reduced to 1.6% in control cultures which lack EGF in the medium. The results show that H16 of hEGF, although not essential for mitogenic activity, optimizes receptor recognition by hydrogen-bond donor/acceptor interactions and may share this feature with H18 of hTGFalpha.