The actions of retinoids on cellular growth correlate with their actions on gap junctional communication.

Retinoic acid (a possible morphogen), its biological precursor retinol, and certain synthetic derivatives of retinol profoundly change junctional intercellular communication and growth (saturation density) in 10T 1/2 and 3T3 cells and in their transformed counterparts. The changes correlate: growth decreases as the steady-state junctional permeability rises, and growth increases as that permeability falls. Retinoic acid and retinol exert quite different steady-state actions on communication at noncytotoxic concentrations in the normal cells: retinoic acid inhibits communication at 10(-10)-10(-9) M and enhances at 10(-9)-10(-7) M, whereas retinol only enhances (10(-8)-10(-6) M). In v-mos-transformed cells the enhancement is altogether lacking. But regardless of the retinoid or cell type, all growth responses show essentially the same dependence on junctional permeability. This is the expected behavior if the cell-to-cell channels of gap junctions disseminate growth-regulating signals through cell populations.

Modulation of lung tumor colony formation by a subcutaneously growing tumor.

When monodisperse tumor cell suspensions of the mouse sarcoma L1 were injected iv into inbred BALB/c mice bearing subcutaneous grafts of the same tumor, the number of tumor nodules developing in the lungs was highest in animals bearing the oldest primary tumor (28 days) and lowest in animals bearing the oldest primary tumor (28 days) and lowest in animals bearing 16-day-old primary tumors. The number of metastases developing in mice bearing 9-day-old subcutaneous tumors was the same as that in control animals without a subcutaneous tumor. This state of low and high susceptibility to lung metastases can be transferred from tumor-bearing mice to whole-body irradiated mice by means of spleen cells, and this transfer of susceptibility is stable for at least 3 weeks. The results indicate a) that a growing subcutaneous tumor can modulate the number of artificial metastases occurring in the lungs as a result of the iv injection of tumor cells, and b) that this modulation occurs by immune mechanisms. In the model presented here, the host is first protected and then made more susceptible to artificial lung metastases by the presence of a progressively growing subcutaneous tumor.

Structure-activity relationships among various retinoids and their ability to inhibit neoplastic transformation and to increase cell adhesion in the C3H/10T1/2 CL8 cell line.

Various natural and synthetic retinoids have been studied for their activity in two biological systems: (a) their activity as inhibitors of methylcholanthrene-induced neoplastic transformation in the C3H/10T1/2 clone 8 mouse fibroblast line (System 1); and (b) their ability to increase the degree of adhesion of C3H/10T1/2 clone 8 cells to a plastic substrate (System 2). These activities were then compared with their known activity in maintaining epithelial differentiation (System 3). With the notable exception of retinoic acid and 13-cis-retinoic acid, which were inactive in Systems 1 and 2, an excellent correlation was observed between activities in Systems 1 and 3 for retinyl acetate, N-(4-hydroxyphenyl)retinamide, retinylidene dimedone, N-ethylretinamide, and N-benzoylretinylamine. Compounds shown to be inactive in System 1 had little or no activity in System 2. However, the ability of retinoids to cause increased adhesion could not be correlated with Systems 1 or 3 in all cases. For instance, retinyl acetate was highly active in Systems 1, 2, and 3, whereas retinylidene dimedone was highly active in Systems 1 and 3 but weakly active in System 2. Conversely, N-(4-hydroxyphenyl)retinylamide was highly active in Systems 1 and 3 but caused a decrease in System 2. The lack of activity of 3 but caused a decrease in System 2. The lack of activity of retinoic acid isomers in the C3H/10T1/2 clone 8 system is paradoxical and may provide important information on requirements for their activation and/or transport.

Modulation of cellular interactions between C3H/10T1/2 cells and their transformed counterparts by phosphodiesterase inhibitors.

It has been demonstrated previously that nontransformed C3H/10T1/2CL8 mouse embryo fibroblasts (10T1/2) can induce a state of reversible growth inhibition in cocultured malignantly transformed mouse fibroblasts and that this inhibition is modulated by serum concentration. The present study suggests that cyclic nucleotides may be implicated in this intercellular communication. The phosphodiesterase inhibitors theophylline, caffeine, and 3-isobutyl-1-methylxanthine (IBX) at concentrations of 10(-3) M, maintained continuously, were all found to inhibit the expression of 3-methylcholanthrene-induced malignant transformation when added 7 days after removal of carcinogen. IBX was the most potent, causing 100% inhibition at 10(-4) M and 70% inhibition at 10(-5) M. This inhibition was partially reversible in the former case and completely reversible in the latter case by removal of drug. Complete inhibition by 10(-4) M IBX was still observed when treatment was delayed 21 days postcarcinogen. In reconstruction experiments, utilizing confluent monolayers of 10T1/2 cells overlaid with transformed cells, IBX caused a dose-dependent inhibition of colony size of the transformed cells. Adenosine cyclic 2':3'-monophosphoric acid (cAMP) and N6,O2'-dibutyryladenosine cyclic 3':5'-monophophoric acid potentiated this response. The presence of non-transformed 10T1/2 cells was required for this effect, since a concentration of IBX (10(-4) M) inhibitory for the growth of transformed cells in mixed cultures was without effect on the growth rate, plating efficiency, or saturation density of pure cultures of 10T1/2 cells or of their transformed counterparts. Conditioned medium removed from IBX-treated 10T1/2 cells was not growth inhibitory for transformed cells, indicating a requirement for cell-cell contact. IBX caused a dose-dependent increase in intracellular cAMP in confluent 10T1/2 cells and a more pronounced increase in cAMP concentration in the culture medium of these cells. The dose-response effects of IBX on growth inhibition of malignant cells in mixed cultures appear to correlate well with its ability to elevate cAMP levels. Thus, IBX increased the capacity of 10T1/2 cells to cause reversible growth arrest of transformed cells and appears to act in a manner analogous to the previously reported effects of serum.

Effects of serum concentration on the expression of carcinogen-induced transformation in the C3H/10T1/2 CL8 cell line.

The C3H/10T1/2 CL8 cell line (10T1/2) is being widely used as a quantitative assay system for chemical and physical carcinogens. 10T1/2 cells, but not their transformed counterparts, exhibit a decreased final saturation density with decreasing serum concentration. Exposure of carcinogen-treated cultures to 5% serum 8 days posttreatment led to a 2- to 6-fold enhancement in transformation frequency in comparison with cultures maintained in 10% serum throughout. Exposure 14,21, or 28 days posttreatment also caused enhancement of transformation frequency, provided a sufficient time for expression of the malignant phenotype was allowed. Exposure to 5% serum 1 or 2 days posttreatment did not lead to significant enhancement of transformation frequency. In contrast, exposure to 15 or 20% serum after 8 days virtually abolished the expression of malignancy; however, this inhibition could be reversed by 5% serum. Morphologically transformed foci isolated from cultures exposed to 5% serum produced clones in agarose with the same frequency as did foci isolated from cultures exposed to 10% serum. Reconstruction experiments, utilizing confluent monolayers of 10T1/2 cells overlaid with transformed cells, demonstrated that the growth of transformed cells decreased proportionally with the log of serum concentration. This effect was not caused by depletion of medium and was dependent upon the presence of 10T1/2 cells. It is concluded that the expression of malignancy in this system is governed by the serum-modulated cell density of the mass of non-transformed cells in the culture.

Restoration of growth control in malignantly transformed mouse fibroblasts grown in a chemically defined medium.

The expression of growth control and morphological transformation was studied in methylcholanthrene-transformed C3H/10T 1/2 CL8 cells serially propagated in CDM by first exposing cells to albumin (0.1%) before dispersing them with trypsin (50 microgram/ml). In serum-supplemented media, methylcholanthrene-transformed C3H/10T 1/2 CL8 cells exhibit various aspects of the transformed phenotype such as irregular morphology, extensive cell overlap, lack of density-dependent inhibition of division, a saturation density of 1.1 X 10(5) cells/sq cm and tumorigenicity in vivo. Cell phenotype in CDM was dramatically altered. Methylcholanthrene-transformed C3H/10T 1/2 1/2 CL8 cells adapted to CDM exhibited a regular epithelioid morphology with no cell overlap and formed confluent monolayers of nonproliferating cells at a saturation density of 5 X 10(4) cells/sq cm. The mean generation time of logarithmic-phase cells was 25 to 27 hr. Reversion to the transformed phenotype followed addition of albumin (0.1%) or serum (2%) to logarithmic-phase cultures or exposure (30 to 60 sec) to trypsin (10 microgram/ml). Cultures in CDM reexposed to serum remained highly tumorigenic in vivo. The data suggest that absorbed proteins may block transformation-sensitive cell surface sites responsible for growth control and that these sites are inactivated by trypsin.

Requirements for and kinetics of growth arrest of neoplastic cells by confluent 10T1/2 fibroblasts induced by a specific inhibitor of cyclic adenosine 3':5'-phosphodiesterase.

This study was conducted to further characterize the previously described phenomenon of growth inhibition of neoplastically transformed C3H/10T1/2 cells (T10T1/2) by nontransformed C3H/10T1/2 clone 8 mouse embryo fibroblast (10T1/2) cells in the presence of inhibitors of cyclic adenosine 3':5'-monophosphate (cAMP) phosphodiesterase. The cAMP phosphodiesterase inhibitor dl-4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone (RO20-1724) was shown to be completely nontoxic to T10T1/2 cells at 10(-4) M, yet when added to mixed cultures of T10T1/2 cells and postconfluence growth-arrested 10T1/2 cells, colony formation and [3H]thymidine incorporation into T10T1/2 cells were virtually eliminated. This effect was dose dependent and was reversible upon drug withdrawal. In 10T1/2 cells, RO20-1724 caused a dose-dependent increase in cAMP levels from about 5 to 150 pmol/10(6) cells; in T10T1/2 cells, 10(-4) M drug treatment caused a 5-fold elevation in cAMP without a clear dose dependency. Cyclic guanosine 3':5'-monophosphate levels in 10T1/2 cells fell by 50% with drug treatment but were unmeasurable in T10T1/2 cells. When intracellular cyclic AMP levels were elevated by the adenyl cyclase stimulator forskolin, growth inhibition of T10T1/2 cells was again induced in mixed cultures but was not observed when added to T10T1/2 cells alone. Addition of RO20-1724 to low concentrations of forskolin produced a greater than additive effect on growth inhibition. Growth inhibition of T10T1/2 cells by RO20-1724 was shown to (a) require contact with, or extremely close proximity to, a confluent monolayer of 10T1/2 cells, (b) be maximum when seeded upon a growth-inhibited monolayer and not an actively growing 10T1/2 culture, and (c) not be decreased by continuous agitation of the culture medium, indicating that readily diffusible inhibitory factors are not involved. A model is presented whereby transformed cells can respond to but cannot themselves generate growth-inhibitory signals produced by post-confluence growth-inhibited nontransformed cells. The existence of these cellular interactions may well explain problems in the quantitation of transformed foci encountered in the use of this cell line as an assay system for chemical and physical carcinogens.

Retinoid effects on cell-cell interactions and growth characteristics of normal and carcinogen-treated C3H/1OT1/2 cells.

The effects of retinyl acetate and all-trans-retinoic acid on the growth rate, saturation density, and cytoplasmic underlapping of normal and carcinogen-initiated C3H/1OT1/2 cells were examined. Retinyl acetate (a) decreased the saturation density by as much as 45%, (b) had no effect on the growth rate, (c) reduced cytoplasmic underlapping of adjacent cells by 55 to 85%, and (d) inhibited neoplastic transformation by methylcholanthrene. The effects were dose dependent and not significantly affected by the serum concentrations over the range of 2.5 to 10%. In contrast, all-trans-retinoic acid (a) decreased the saturation density as effectively as retinyl acetate, but only in medium containing 10% serum; (b) significantly reduced the growth rate of cells at low density, especially at low serum concentrations; (c) had no effect upon cytoplasmic underlapping; and (d) enhanced transformation at nontoxic concentrations in medium containing 5% serum but had no effect in 10% serum. We conclude that the effects of retinoids on the growth rate and saturation density of cells in culture may not be relevant to their inhibition of neoplastic transformation. Rather, we interpret the results of our experiments on cytoplasmic underlapping as an indication that retinoids inhibit transformation by stabilizing and/or enhancing cell surface receptors involved in cell-cell contact-dependent formation of a stable monolayer.

Augmentation of postconfluence growth arrest of 10T1/2 fibroblasts by endogenous cyclic adenosine 3':5'-monophosphate.

We have previously demonstrated that, by elevating intracellular adenosine 3':5'-cyclic monophosphate (cAMP) levels by inhibition of cAMP phosphodiesterase with Ro 20-1724 or by forskolin stimulation of adenylcyclase, the growth of neoplastically transformed 10T1/2 fibroblasts could be inhibited when these cells were in contact with growth-inhibited nontransformed 10T1/2 cells (J. S. Bertram and M. B. Faletto, Cancer Res., 45: 1946-1952, 1985) and furthermore that the extent of this growth inhibition correlated strongly with the degree of junctional communication between the two cell types (P.P. Mehta et al., Cell, 44: 187-196, 1986). To determine if these treatments enhance the degree of growth control of the nontransformed 10T1/2 cells, cultures were exposed to varying concentrations of Ro 20-1724 and/or forskolin. Drug treatment caused no significant effects on growth rate or cell spreading when cells were treated during logarithmic growth phase; however, major reductions of up to 70% in confluent saturation density and concomitant increases in cell spreading occurred in cultures making extensive cell/cell contacts. Decreases in saturation density correlated strongly with induced elevations of both intra- and extracellular cAMP concentrations. These effects could not be duplicated by the addition of exogenous cAMP agonists 8-bromo-cAMP and/or dibutyryl-cAMP. Two-dimensional electrophoresis of phosphate-labeled proteins revealed that forskolin treatment induced a quantitatively and qualitatively different phosphorylation profile than did 8-bromo-cAMP. Both basal and drug-induced intracellular cAMP levels fell as cells progressed from logarithmic to confluent growth state, implying that cells become sensitized to cAMP by the attainment of extensive cell/cell contacts. It is suggested that the drug-induced elevations of endogenously synthesized cAMP are accentuating a physiological role of cAMP on the postconfluent growth arrest of murine fibroblasts. The requirements for cell/cell contact and the known increased junctional communication induced by cAMP furthermore suggest that cAMP is enhancing the junctional transfer of a growth-inhibiting regulatory molecule. A likely candidate is cAMP itself.

Carotenoids up-regulate connexin43 gene expression independent of their provitamin A or antioxidant properties.

Epidemiological evidence and studies in whole animals and cell culture have indicated that carotenoids have cancer chemopreventive action. In mouse C3H10T1/2 cells, this activity is highly correlated with the ability of carotenoids to up-regulate gap junctional intercellular communication. Here, we report that in mouse cells, carotenoids increase the expression of connexin43, a gene that encodes a major gap junction protein. This effect appears unrelated to their provitamin A or antioxidant properties, since carotenoids with and without provitamin A activity increased levels of connexin43 mRNA and protein, whereas the antioxidants methyl-bixin and alpha-tocopherol were inactive. Moreover, the active carotenoid canthaxanthin did not induce the vitamin A-inducible gene retinoic acid receptor-beta. Connexin43 is the first carotenoid-inducible gene described in mammals. By indicating an additional pathway through which carotenoids function, these data provide a mechanistic basis for cancer chemoprevention by carotenoids and may lead to a re-evaluation of carotenoid physiology.

Quantitative neoplastic transformation of C3H/10T1/2 fibroblasts: dependence upon the size of the initiated cell colony at confluence.

The efficiency of transformation of C3H/10T1/2 fibroblasts by chemical or physical carcinogens varies inversely with the seeding density of the cells. Colony size at confluence, accumulated cell generations, and epigenetic events independent of seeding density have been proposed to explain this variability. By controlling the total number and colony size (i.e., cells/colony) of initiated cells at confluence and their accumulated cell generations, we have determined that (a) the colony size of the initiated cells at confluence is directly related to the expression of the transformed phenotype, (b) the probability that an initiated cell will form a colony of transforming cells does not vary with the accumulation of large numbers of cell generations, and (c) the total number of initiated cells in a confluent culture does not determine the number of transformed foci formed. We have developed a mathematical representation which, when applied to our results and the published results from other laboratories, clearly describes the relationship between the size of the initiated cell colony at confluence and the expression of the transformed phenotype. These results have significant implications in the quantification of the transformation of C3H/10T1/2 cells by chemical and physical carcinogens.

Rationale and strategies for chemoprevention of cancer in humans.

The potential for chemical intervention (chemoprevention) as a means of halting or delaying the process of carcinogenesis is assessed as a strategy for reducing the incidence of human cancer. The process of carcinogenesis is dissected into its constituent steps, thereby exposing sites for intervention. These sites are then critically discussed with regard to the existence of chemicals active at these sites using data gained from the laboratory and from epidemiological studies, intrinsic problems or advantages associated with intervention at specific sites in the carcinogenic process, and practical aspects of intervention in humans. The design and potential long-term positive and negative consequences of chemoprevention clinical trials are critically discussed, with the objective of exposing the major differences that exist between clinical trials in cancer chemoprevention and those in cancer chemotherapy. Results of completed prevention trials and details of ongoing trials are presented and discussed. Based on the laboratory, epidemiological, and clinical evidence presented, it is concluded that chemoprevention offers excellent prospects as a means of reducing cancer incidence. Among currently available agents, the retinoids possess the best combination of properties. However, much more research is needed to optimize drugs and protocols and to develop interim end points for assessing response. The authors finally caution that overambitious claims for the prospects for chemoprevention may lead to reduced emphasis on the need for changes in life-style (principally in smoking and diet) that are viewed as having the greatest potential for reducing cancer incidence.

The isolation and characterization of polycyclic hydrocarbon-binding proteins from mouse liver and skin cytosols.

The major protein to which metabolites of methylcholanthrene are covalently bound has been purified from C3H mouse liver cytosol. Its properties are identical to the mouse skin h-protein, which may be primary arget of carcinogenic hydrocarbon metabolites during transformation to caner. It has a molecular wight of 44,000, consists of 2 subunits o- M.W. 20,000, has an isoelectric point (pI) of 8.05 to 8.6, and a sedimentation coefficient of 3.6 S. These physical properties are rather similar to those of ligandin, a hepatic protein that binds carcinogen metabolites, steroid anionic metabolites, bilirubin, and exogenous organic anions, but not to those of the rat liver azo dye carcinogen binding 'slow h-2-5S' protein. The h-protein and ligandin consistently give different pl values. Two minor basic proteins (molecular weights around 44,000 each), to whcih methylcholanthrene metabolites are convalently bound, have been separated from the h-protein by carboxymethyl-cellulose chromatography. Prelininary results indicate that these 2 minor proteins are related to ligandin. A protein to which methylcholanthrene is noncovalently bound was also identified in the acidic fraction of the mouse liver and skin sytosols and has been partially purified and characterized. It has a molecular weight of 60,000, a pl of 5.0, and a sedimentation coefficient of 4.5S.

Cell cycle-specific oncogenic transformation of C3H/10T1/2 clone 8 mouse embryo cells by 1-beta-D-arabinofuranosylcytosine.

1-beta-Darabinofuranosylcytosine at concentrations ranging from 10(-3) to 10(-6) M induces oncogenic transformation in the C3H/10T1/2 clone 8 mouse embryo cell line. Cell lines derived from type III transformed foci grew in soft agarose and produced tumors in immunosuppressed syngeneic mice. With cells synchronized by postconfluent inhibition of growth or isoleucine deprivation, transformation was cell cycle dependent. Maximal transformation was seen in cells treated when in S phase, although some transformation was seen in cells treated in G1 phase of the cell cycle.

Differential uptake, binding, and metabolism of retinol and retinoic acid by 10T1/2 cells.

Previous studies have shown that all-trans-retinoic acid fails to inhibit chemically induced transformation in 10T1/2 cells except at toxic levels, whereas retinol and many synthetic retinoids are potent inhibitors. In contrast, in many systems retinoic acid is a more effective modulator of differentiation and carcinogenesis than is retinol. In any attempt to explain this anomaly, we have studied the differential metabolism of retinoic acid and retinol by 10T1/2 cells and by their initiated and transformed derivatives, and have also reexamined these cells for the presence of retinoid-binding proteins. Whereas retinoic acid was depleted from the medium bathing 10T1/2 and initiated 10T1/2 cells within 48 h of treatment, retinol was concentrated 500-fold by these cells, and disappeared from the culture medium no faster than from cell-free medium. Retinoic acid metabolism by a number of transformed cell lines varied widely. There was no apparent correlation between metabolizing ability and transforming agent (methylcholanthrene, X-rays, fission spectrum neutrons, and plasmid oncogene transfection). Uptake of retinoic acid was seen in all cell lines and was not correlated with its metabolism. Retinol was metabolized minimally by all cell lines tested; metabolism of retinol was not correlated with retinoic acid metabolizing ability. Retinoic acid-induced growth inhibition and cytotoxicity were not correlated with metabolizing ability, suggesting that the rate of metabolism of retinoic acid is not a major determinant of its acute biological effects. Using sensitive radioimmunoassays, cellular retinoic acid- (CRABP) and retino-binding proteins (CRBP) were both detected in 10T1/2 and initiated 10T1/2 cells. CRABP levels of about 16 ng/10(6) cells were about 4-fold higher than CRBP levels. Therefore, lack of CRABP does not explain the failure of retinoic acid to inhibit carcinogen-induced transformation in these cells. These studies suggest that the inability of retinoic acid to inhibit transformation in the 10T1/2 cell system may be due to its rapid metabolism and clearance from the medium. On the other hand, the high cellular uptake and stability of retinol in these cells could be an important factor in the inhibition of neoplastic transformation by this retinoid.

Changes in ribo- and deoxyribonucleoside triphosphate pools within the cell cycle of a synchronized mouse fibroblast cell line.

Intracellular pool levels of ribo- and deoxyribonucleoside triphosphates were monitored throughout the cell cycle of C3H10T1/2 mouse embryo fibroblast cells synchronized by isoleucine deprivation. Absolute pool sizes of ribonucleoside triphosphates were approximately 30 fold greater than those of the corresponding deoxyribonucleoside triphosphates. Of the ribonucleoside triphosphates, pool sizes of ATP exhibited the greatest change, increasing from a low of 32.7 nmol/10(7) cells during G1 to a high of 81.6 nmol/10(7) cells 2 h prior to mid S-phase. Levels of ATP subsequently declined to 40.2 nmol/10(7) cells during late S-phase, followed by a second peak of 65.8 nmol/10(7) with the onset of cell division. No significant changes in the pool sizes of UTP and GTP were found throughout the cell cycle. Of the deoxyribonucleoside triphosphates, pool sizes of pyrimidine deoxyribonucleoside triphosphates were approx. 5-10 fold greater than those of purine deoxyribonucleoside triphosphates. Low levels of deoxyribonucldoside triphosphates during G1 (0.3-1.3 pmol/10(7) cells) increased coordinately with the initiation of DNA synthesis to an initial peak during mid S-phase (0.5-6.4 pmol/10(7) cells). Declining levels of deoxyribonucleoside triphosphates during late S-phase were followed by a subsequent larger second peak (1.7-10.7 pmol/10(7) cells) during G2-M.

Contractile proteins. Major components of nuclear and chromosome non-histone proteins.

Two of the major non-histone proteins from Physarum polycephalum have been isolated under nondenaturing conditions and identified as actin and myosin. A third protein has been purified from crude nuclear actomyosin and from residual nonhistone fractions and found to bind actomyosin in the presence of Mg2+. In Physarum these proteins are not components of the nuclear membrane. Based on sodium dodecylsulfate-polyacrylamide gel electrophoresis, similar proteins are also present in nuclei of HeLa cells and mouse embryo fibroblasts. Isolated metaphase chromosomes from Physarum show a several-fold enrichment in myosin and an altered ratio of actin to the Mg2+-dependent actomyosin binding protein as compared to interphase nuclei. When non-proliferative states are induced in any of these cells, the Mg2+-dependent actomyosin binding protein decreases while actin increases several fold in intranuclear concentration; concomitantly, there is a generalized condensation and inactivation of chromatin. Experiments with added purified radioactive nuclear actomyosin; comparative studies on nuclear protein during stepwise nuclear purification; and studies on isolated metaphase chromosomes indicate that these proteins exist in nuclei in vivo. These observations suggest that contractile proteins may function in the structural interconversions of chromatin and in the regulation of cell proliferation;

Gap-junctional protein connexin 43 is expressed in dermis and epidermis of human skin: differential modulation by retinoids.

Retinoids are effective modulators of proliferation and differentiation of keratinocytes in vivo and in vitro. In mouse 10T1/2 cells, retinoid action on proliferation and neoplastic transformation is correlated with the upregulation of gap-junctional communication and expression of connexin 43 (Cx43). In the present study we have determined if retinoids induce similar effects on gene expression in human skin. Studies were conducted in intact skin and on cultured keratinocytes and dermal fibroblasts. In a clinical study, 2 weeks of treatment with 0.05% all-trans retinoic acid resulted in increased expression of Cx43 mRNA and protein in epidermis. Expression occurred predominantly in the suprabasal layer. Cultured cells exhibited a differential response to retinoic acid. In keratinocytes, increased expression of Cx43 occurred at low (10(-11) M) concentrations, whereas inhibition occurred at high (10(-7) M) concentrations; however, junctional communication, measured by dye transfer, was not altered over this concentration range. Dermal fibroblasts, in contrast, exhibited a dose-dependent increased expression of Cx43 at concentrations up to 10(-7) M retinoic acid and proportionately increased their junctional communication over this dose range. These data indicate that control of Cx43 gene expression by retinoids in human skin cells is complex. The production of gradients of junctional channels could play a role in the control of growth and differentiation in epidermis.

The molecular biology of cancer.

The process by which normal cells become progressively transformed to malignancy is now known to require the sequential acquisition of mutations which arise as a consequence of damage to the genome. This damage can be the result of endogenous processes such as errors in replication of DNA, the intrinsic chemical instability of certain DNA bases or from attack by free radicals generated during metabolism. DNA damage can also result from interactions with exogenous agents such as ionizing radiation, UV radiation and chemical carcinogens. Cells have evolved means to repair such damage, but for various reasons errors occur and permanent changes in the genome, mutations, are introduced. Some inactivating mutations occur in genes responsible for maintaining genomic integrity facilitating the acquisition of additional mutations. This review seeks first to identify sources of mutational damage so as to identify the basic causes of human cancer. Through an understanding of cause, prevention may be possible. The evolution of the normal cell to a malignant one involves processes by which genes involved in normal homeostatic mechanisms that control proliferation and cell death suffer mutational damage which results in the activation of genes stimulating proliferation or protection against cell death, the oncogenes, and the inactivation of genes which would normally inhibit proliferation, the tumor suppressor genes. Finally, having overcome normal controls on cell birth and cell death, an aspiring cancer cell faces two new challenges: it must overcome replicative senescence and become immortal and it must obtain adequate supplies of nutrients and oxygen to maintain this high rate of proliferation. This review examines the process of the sequential acquisition of mutations from the prospective of Darwinian evolution. Here, the fittest cell is one that survives to form a new population of genetically distinct cells, the tumor. This review does not attempt to be comprehensive but identifies key genes directly involved in carcinogenesis and demonstrates how mutations in these genes allow cells to circumvent cellular controls. This detailed understanding of the process of carcinogenesis at the molecular level has only been possible because of the advent of modern molecular biology. This new discipline, by precisely identifying the molecular basis of the differences between normal and malignant cells, has created novel opportunities and provided the means to specifically target these modified genes. Whenever possible this review highlights these opportunities and the attempts being made to generate novel, molecular based therapies against cancer. Successful use of these new therapies will rely upon a detailed knowledge of the genetic defects in individual tumors. The review concludes with a discussion of how the use of high throughput molecular arrays will allow the molecular pathologist/therapist to identify these defects and direct specific therapies to specific mutations.

Dietary carotenoids inhibit neoplastic transformation and modulate gene expression in mouse and human cells.

Many epidemiologic studies have associated the consumption of diets rich in fruit and green and yellow vegetables with a decreased risk of cancer. Of the many components of such a diet, the content of carotenoids, particularly beta-carotene, has been most consistently linked to decreased risk. The biological mechanism for such protection is currently unclear. Multiple possibilities exist: carotenoids are potent antioxidants and oxidative stress is known to contribute to carcinogenesis; many carotenoids can be converted to retinoids, these are known cancer preventive agents at several anatomic sites; and carotenoids may possess additional actions in mammalian cells. In a model in vitro system we showed that carotenoids both with and without provitamin A activity inhibit carcinogen-induced neoplastic transformation, inhibit plasma membrane lipid oxidation, and cause up-regulated expression of connexin 43, a gene coding for the structural unit of a gap junction. This last activity was statistically correlated with the ability to inhibit neoplastic transformation. Activity has also been shown in human cells: in fibroblasts CONNEXIN 43 expression is also up-regulated whereas in human keratinocytes grown in organotypic culture beta-carotene and canthaxanthin modulate differentiation in a manner qualitatively similar to that of retinoids. These results strongly suggest that carotenoids have intrinsic cancer chemopreventive action in humans.