The hedgehog processing pathway is required for NSCLC growth and survival.

Considerable interest has been generated from the results of recent clinical trials using smoothened (SMO) antagonists to inhibit the growth of hedgehog (HH) signaling-dependent tumors. This interest is tempered by the discovery of SMO mutations mediating resistance, underscoring the rationale for developing therapeutic strategies that interrupt HH signaling at levels distinct from those inhibiting SMO function. Here, we demonstrate that HH-dependent non-small cell lung carcinoma (NSCLC) growth is sensitive to blockade of the HH pathway upstream of SMO, at the level of HH ligand processing. Individually, the use of different lentivirally delivered shRNA constructs targeting two functionally distinct HH-processing proteins, skinny hedgehog (SKN) or dispatched-1 (DISP-1), in NSCLC cell lines produced similar decreases in cell proliferation and increased cell death. Further, providing either an exogenous source of processed HH or a SMO agonist reverses these effects. The attenuation of HH processing, by knocking down either of these gene products, also abrogated tumor growth in mouse xenografts. Finally, we extended these findings to primary clinical specimens, showing that SKN is frequently overexpressed in NSCLC and that higher DISP-1 expression is associated with an unfavorable clinical outcome. Our results show a critical role for HH processing in HH-dependent tumors, identifies two potential druggable targets in the HH pathway, and suggest that similar therapeutic strategies could be explored to treat patients harboring HH ligand-dependent cancers.

Uncovering residues that regulate cyclin D1 proteasomal degradation.

Cyclin D1 regulates G1 cell-cycle progression and is aberrantly expressed in carcinogenesis. Proteasomal degradation of cyclin D1 was highlighted as a cancer chemopreventive mechanism. To understand this mechanism better, residues responsible for degradation and ubiquitination of cyclin D1 were investigated. Eighteen lysines in cyclin D1 had single, double or multiple mutations engineered before transfection into BEAS-2B human bronchial epithelial (HBE) cells to evaluate stabilities after all-trans-retinoic acid (RA) or cycloheximide treatments. Specific mutations stabilized cyclin D1, including substitutions of lysines surrounding the cyclin box domain that inhibited RA-mediated degradation and extended the cyclin D1 half-life. Mutation of all cyclin D1 lysines blocked polyubiquitination. N-terminus (but not C-terminus) modification stabilized cyclin D1. Ubiquitination-resistant mutants preferentially localized cyclin D1 to the nucleus, directly implicating subcellular localization in regulating cyclin D1 degradation. Taken together, these findings uncover specific residues conferring ubiquitination of cyclin D1. These provide a mechanistic basis for proteasomal degradation of cyclin D1.

Frequent requirement of hedgehog signaling in non-small cell lung carcinoma.

Although it had previously been suggested that the hedgehog (HH) pathway might be activated in some lung tumors, the dependence of non-small cell lung carcinomas (NSCLC) for HH activity had not been comprehensively studied. During a screen of a panel of 60 human tumor cell lines with an HH antagonist, we observed that the proliferation of a subset of NSCLC cell lines was inhibited. These NSCLC cell lines express HH, as well as key HH target genes, consistent with them being activated through an autocrine mechanism. Interestingly, we also identified a number of NSCLC cell lines that express high levels of the downstream transcription factor GLI1 and harbor enhanced levels of HH activity, but appear insensitive to known HH antagonists. We hypothesized that the high levels of GLI1 in these cells would function downstream of the HH antagonist target, allowing them to bypass the antagonist-mediated block in proliferation. Consistent with this hypothesis, when the levels of GLI1 are knocked down in such cells, they become sensitive to these inhibitors. We go on to show that a large percentage of primary NSCLC samples express GLI1, consistent with constitutive activation of the HH pathway in these samples. Taken together, these results establish the involvement of the HH signaling pathway in a subset of NSCLCs.

Retinoid target genes in acute promyelocytic leukemia.

All-trans-retinoic acid (RA)-based differentiation therapy induces clinical remissions in acute promyelocytic leukemia (APL). This has propelled interest in elucidating the molecular mechanisms responsible for these remissions. The t(15;17) rearrangement results in the expression of the PML/RARalpha fusion transcript that is paradoxically linked to the etiology and clinical retinoid response in APL. PML/RARalpha expression blocks terminal myeloid differentiation in APL. Treatment with pharmacological RA dosages overcomes the dominant-negative effects of PML/RARalpha to activate transcription of retinoid target genes. This regulation is linked directly to RA effects in APL, including PML/RARalpha degradation and induction of differentiation. Identifying retinoid target genes is an important step in developing a mechanistic understanding of RA effects in APL. RA target genes have been uncovered through the use of molecular genetic approaches as well as unique cellular and transgenic APL models. Recent developments in the proteomic and functional genomic fields are providing useful tools for elucidating mechanisms of RA response or resistance in APL. These target genes represent potential therapeutic targets in APL and other retinoid-responsive diseases. Previous spotlights in Leukemia have highlighted the importance of cytokine effects and signal transduction crosstalk in retinoid response in APL and in normal hematopoiesis. This review builds on prior work by addressing the role of retinoid target genes in mediating retinoid response or resistance in APL.

Retinoic acid activates p53 in human embryonal carcinoma through retinoid receptor-dependent stimulation of p53 transactivation function.

Although retinoids are known to regulate gene transcription by activating retinoid receptors, the targets of retinoid receptors are largely unknown. This study indicates effective all-trans retinoic acid (RA)-induced differentiation of human embryonal carcinoma cells engages p53. Unexpectedly, RA has been found to activate the transactivation function of p53 in the human embryonal carcinoma cell line, NT2/D1, in a retinoid receptor-dependent manner. A derived RA-resistant line, NT2/D1-R1, is deficient in this activity and is co-resistant to cisplatin. This indicates that RA and cisplatin responses may share a common pathway involving p53 in embryonal carcinomas. RA has no effect on p53 steady-state protein levels in either line. RA enhances endogenous p53 transactivation activity in NT2/D1 but not NT2/D1-R1 cells. In addition, RA induces transactivation activity of a gal4-p53 fusion protein, suggesting that RA activates p53 independent of increasing p53 levels or sequence-specific DNA binding. This activity is absent in retinoic acid receptor gamma (RARgamma)-deficient NT2/D1-R1 cells but can be restored upon co-transfection with specific RARs. Transient transfection of a dominant-negative p53 construct in NT2/D1 cells blocks the RA-mediated transcriptional decline of a differentiation-sensitive reporter plasmid and enhances survival of NT2/D1 cells following cisplatin treatment. Taken together, these findings indicate that RA activates the intrinsic activation function of p53 by a novel mechanism independent of effects on p53 stability or DNA binding and that this activation may be a general mechanism that contributes to RA-mediated G1 arrest.

Multiple high-grade bronchial dysplasia and squamous cell carcinoma: concordant and discordant mutations.

Loss of heterozygosity (LOH) involving chromosomes 3p, 5q, 9p, or 17p and aberrant expression or mutation of p53 are reported previously in selected bronchial dysplasias and squamous cell cancers (SCCs). Yet, comprehensive analyses of LOH patterns at these chromosomal sites and of p53 alterations are not reported for histologically normal bronchial epithelium, high-grade bronchial dysplasia, and SCC present in the same pulmonary resections. Whether concordant or discordant genetic changes are detected in these bronchial tissues, especially when multiple high-grade dysplastic bronchial lesions are present, was studied. Genomic DNA was microdissected from eight pulmonary SCCs and high-grade dysplastic lesions that were associated with SCC. In four cases, two independent high-grade dysplastic bronchial lesions were identified. When available, histologically normal bronchial epithelium was microdissected. Germ-line genomic DNA was isolated from normal lymph nodes. LOH was assessed for 15 microsatellite markers on chromosomes 3p, 5q, 9p, or 17p, sites frequently deleted in lung cancers. Immunohistochemical p53 expression was studied and correlated with p53 DNA sequence analyses. Progressive LOH for these markers was found when SCCs were compared with high-grade dysplasia and histologically normal bronchial epithelium present in the same resections. Histologically normal bronchial specimens had LOH in up to 27% of informative markers. High-grade dysplastic lesions exhibited LOH for 18-45% and SCC had LOH for 18-73% of the markers. Common regions of LOH were found in some dysplasias compared with SCCs. In other dysplasias, discordance was found relative to SCCs, especially for p53 mutations. In cases with a single or second high-grade dysplasia associated with SCC, heterogeneity in LOH markers was detected. These concordant and discordant changes were consistent with convergent and divergent clonal selection pathways in pulmonary squamous cell carcinogenesis. Some histologically normal bronchial epithelial tissues had genetic changes more similar to those in the SCCs than in dysplastic lesions. DNA loss or mutations accumulate in SCC, but discordant genetic changes can exist in the same carcinogen-exposed bronchial tissues. These findings have implications for lung cancer prevention trials.

Autofluorescence characteristics of immortalized and carcinogen-transformed human bronchial epithelial cells.

Tissue autofluorescence has been explored as a potential method of noninvasive pre-neoplasia (pre-malignancy) detection in the lung. Here, we report the first studies of intrinsic cellular autofluorescence from SV40 immortalized and distinct tobacco-carcinogen-transformed (malignant) human bronchial epithelial cells. These cell lines are useful models for studies seeking to distinguish between normal and pre-neoplastic human bronchial epithelial cells. The cells were characterized via spectrofluorimetry and confocal fluorescence microscopy. Spectrofluorimetry revealed that tryptophan was the dominant fluorophore. No change in tryptophan emission intensity was observed between immortalized and carcinogen-transformed cells. Confocal autofluorescence microscopy was performed using a highly sensitive, spectrometer-coupled instrument capable of limiting emission detection to specific wavelength ranges. These studies revealed two additional endogenous fluorophores, whose excitation and emission characteristics were consistent with nicotinamide adenine dinucleotide (NADH) and flavins. In immortalized human bronchial epithelial cells, the fluorescence of these species was localized to cytoplasmic granules. In contrast, the carcinogen-transformed cells showed an appreciable decrease in the fluorescence intensity of both NADH and flavins and the punctate, spatial localization of the autofluorescence was lost. The observed autofluorescence decrease was potentially the result of changes in the redox state of the fluorophores. The random cytoplasmic fluorescence pattern found in carcinogen-transformed cells may be attributed to changes in the mitochondrial morphology. The implications of these results to pre-neoplasia detection in the lung are discussed.

Cyclin proteolysis as a retinoid cancer prevention mechanism.

The retinoids, natural and synthetic derivatives of vitamin A, are active in cancer therapy and prevention. Their biological effects are mediated through ligand-dependent interactions with retinoid receptors that associate with specific co-regulators. A better understanding of retinoid chemopreventive mechanisms is needed. Our prior work revealed that all-trans-retinoic acid (RA) prevented tobacco-specific carcinogenic transformation of cultured human bronchial epithelial cells. RA signaled G1 arrest that permitted repair of genomic DNA damage caused by these carcinogens. RA triggered G1 arrest at least partly through proteasome-dependent degradation of cyclin D1. Proteasomal inhibitors blocked RA-mediated cyclin D1 degradation. To confirm that a specific proteolysis pathway was induced by RA-treatment, a degradation assay was established using in vitro translated cyclin D1 and cellular extracts from RA-treated or untreated human bronchial epithelial cells. Incubation of RA-treated but not the control cellular extracts with in vitro translated cyclin D1 led to cyclin degradation. This degradation depended on the PEST domain of cyclin D1, implicating ubiquitination in this retinoid degradation. Retinoid receptor selective agonists demonstrated that retinoic acid receptor (RAR)beta and retinoid X receptor (RXR) but not RARalpha- or RARgamma-dependent pathways signaled this cyclin degradation. Findings were extended to the NT2/D1 human embryonal carcinoma differentiation model where a similar pathway was activated by RA-treatment. To determine whether G1 cyclins were involved directly in bronchial preneoplasia, immunohistochemical expression profiles for cyclins D1 and E were examined. Aberrant expression of these cyclins was frequent in bronchial preneoplasia. Taken together, these findings indicate that ubiquitin-dependent proteolysis of G1 cyclins is a retinoid chemoprevention mechanism. Whether the retinoids represent the optimal agents to activate this pathway is the subject of ongoing work. These findings provide a rationale for combining the retinoids in chemoprevention trials with other agents that do not activate this proteolysis pathway. What is now known about the retinoids as cancer prevention agents will be reviewed. Emphasis is placed on retinoid effects on cell cycle progression at G1.

The retinoids and cancer prevention mechanisms.

Carcinogenesis is a multistep process that converts normal cells into malignant cells. Once transformed, malignant cells acquire the ability to invade and metastasize, leading to clinically evident disease. During this continuum from normal to metastatic cells, carcinogenic steps can be arrested or reversed through pharmacological treatments, known as cancer chemoprevention. Chemoprevention strategies represent therapeutic interventions at early stages of carcinogenesis, before the onset of invasive cancer. Effective chemoprevention should reduce or avoid the clinical consequences of overt malignancies by treating early neoplastic lesions before development of clinically apparent signs or symptoms. Preclinical, clinical, and epidemiological data provide considerable support for cancer chemoprevention as an attractive therapeutic strategy. This clinical approach was validated in the recent tamoxifen randomized trial, demonstrating that a selective estrogen receptor modulator reduces the risk of breast cancer in women at high risk for this malignancy. Derivatives of vitamin A, the retinoids, have reported activity in treating specific premalignant lesions and reducing incidence of second primary tumors in patients with prior head and neck, lung or liver cancers. Whether the retinoids will prevent primary cancers at these sites is not yet known. Notably, a carotenoid (beta-carotene) was shown as inactive in primary prevention of lung cancers in high-risk individuals. This underscores the need for relevant in vitro models to identify pathways signaling chemopreventive effects. These models should assess the activity of candidate chemoprevention agents before the conduct of large and costly prevention trials. An improved understanding of cancer prevention mechanisms should aid in the discovery of new therapeutic targets and chemoprevention agents. Ideally, these agents should have tolerable clinical toxicities suitable for chronic administration to individuals at high risk for developing primary or second cancers. This article reviews what is now known from clinical and preclinical studies about the retinoids as cancer prevention agents.

Transcriptional activation of the cyclin-dependent kinase inhibitor p21 by PML/RARalpha.

Acute promyelocytic leukemia (APL) is a result of clonal expansion of hematopoietic precursors blocked at the promyelocytic stage and is associated with a t(15;17) chromosomal translocation and the expression of the PML/RARalpha fusion protein. Treatment of APL cells with retinoic acid (RA) leads to complete remission by inducing growth arrest and differentiation of these cells into granulocytes. The cyclin-dependent kinase inhibitor p21WAF1/CIP1 may be involved in terminal differentiation associated growth arrest. We showed in this study that PML/RARalpha increased the transcription of p21WAF1/CIP1 gene and the activation was further induced by RA treatment. Deletion analysis revealed a region upstream of the p21WAF1/CIP1 promoter that is required for transactivation by PML/RARalpha. Transient transfection of PML/RARalpha in cells increased the endogenous p21WAF1/CIP1 protein levels. These results suggest that the induction of APL cells differentiation by RA may be a result of the activation of p21WAF1/CIP1 by PML/RARalpha.

4HPR triggers apoptosis but not differentiation in retinoid sensitive and resistant human embryonal carcinoma cells through an RARgamma independent pathway.

Retinoids signal biological effects through retinoic acid receptors (RAR) and retinoid X receptors (RXR) and their co-regulators. We previously reported that all-trans retinoic acid (RA) triggers terminal differentiation in the human embryonal carcinoma cell line NTERA-2 clone D1 (NT2/D1), through an RARgamma dependent pathway. RARgamma repression in NT2/D1-R1 cells accounts for RA resistance in this line. This report finds RARgamma repression is due to selective repression of RARgamma but not RARbeta transcription in NT2/D1-R1 cells. The repression is neither due to mutations in RARgamma nor its promoter containing the RA response element. Prior work was confirmed and extended by demonstrating that an RARgamma selective agonist preferentially signals differentiation of NT2/D1 cells, while RARalpha/beta, RARbeta, RXR agonists and an RAR pan-antagonist do not even when NT2/D1 cells are treated with these retinoids at 10 microM dosages. None of these examined retinoids induced differentiation of the RA resistant NT2/D1-R1 cells. In contrast, N-(4-hydroxyphenyl)retinamide (4HPR), a reported transcriptional activator of RARgamma was shown to potently induce growth inhibition and apoptosis in both NT2/D1 and NT2/D1-R1 cells. 4HPR-induced apoptosis was unaffected by co-treatment of both cell lines with equimolar RAR antagonist. Semi-quantitative reverse transcription-polymerase chain reaction (RT - PCR) assays of total RNA from 4HPR-treated NT2/D1 and NT2/D1-R1 cells did not reveal RARgamma induction. Since 4HPR signals in RA-resistant NT2/D1-R1 cells having an RARgamma transcriptional block, these results indicate that 4HPR triggers apoptosis but not differentiation through an RARgamma independent pathway. Taken together, these findings implicate a therapeutic role for 4HPR mediated apoptosis in germ cell tumors even when a maturation block is present.

Telomerase activity in germ cell cancers and mature teratomas.

BACKGROUND: An inverse relationship has been reported between the presence of telomerase enzymatic activity and the induction of differentiation in human tumor cell lines. Male germ cell tumors represent an attractive clinical model to assess this relationship further because high telomerase activity is present in normal germ cell progenitors and in embryonal carcinomas that can differentiate into mature teratomas. To investigate how telomerase activity and the differentiation state of germ cell tumors are related, telomerase activities and telomere lengths were measured in benign testicular tissues, germ cell cancers, and mature or immature teratomas. METHODS: By use of a modified telomeric repeat amplification protocol (TRAP) assay, telomerase activity was measured in four specimens of benign testicular tissue, in 27 germ cell cancers, in seven mature teratomas, and in one immature teratoma. Telomere lengths were measured in all specimens by restriction digestion of genomic DNA and Southern blot hybridization analysis. Associations between telomerase activity and tissue histopathology were assessed with two-sided Fisher's exact tests. RESULTS: Telomerase activity was detected in all examined germ cell cancers and in the benign testicular tissue specimens. In marked contrast, telomerase activity was not detected in any mature teratoma (P<.0001). Very long telomeres were detected in some mature teratomas, consistent with telomerase repression as a late event in teratoma formation. The immature teratoma, with malignant transformation, had high telomerase activity. CONCLUSION: Telomerase is active in germ cell cancers and repressed in mature teratomas. The absence of telomerase activity may contribute to the limited proliferative capacity of mature teratomas. These findings support the existence of an inverse relationship between telomerase activity and the differentiation state of clinical germ cell tumors.

Genesis, a Winged Helix transcriptional repressor, has embryonic expression limited to the neural crest, and stimulates proliferation in vitro in a neural development model.

A novel repressor of the Winged Helix (formerly HNF-3/Forkhead) transcriptional regulatory family, termed Genesis (also called HFH2), was previously found to be exclusively expressed in primitive embryonic cell lines. In this study in situ cRNA hybridization experiments revealed that Genesis was expressed during embryogenesis only in developing neural crest cells. Its expression diminished upon their terminal differentiation into sympathetic and parasympathetic neurons. Based on that finding, Genesis was retrovirally transduced into pluripotent N-Tera-2 clone D1 (NT2/D1) teratocarcinoma cells, which are a well-described in vitro model of neural development. Retinoic acid (RA) treatment will drive these cells to differentiation toward the neuronal lineage and cause an increase in expression of the cyclin-dependent kinase inhibitor p21 protein, which leads to an inhibition in cellular proliferation. Although RA-induced expression of neuronal differentiation markers was not influenced by forced overexpression of Genesis in NT2-D1 cells, proliferation of Genesis-transduced cells continued following RA treatment. RA was unable to induce the expression of the cyclin-dependent kinase inhibitor p21 in the Genesis-transduced cells, but Go/G1 tumor suppressor p53 expression was induced normally. Therefore, Genesis may play a role in the regulation of primitive neural crest development by preventing terminal quiescence through inhibition of p21 protein expression. These data also lend evidence for the hypothesis that proliferation and differentiation pathways are not irrevocably linked, but can function independently.

Retinoic acid promotes ubiquitination and proteolysis of cyclin D1 during induced tumor cell differentiation.

Mechanisms by which differentiation programs engage the cell cycle are poorly understood. This study demonstrates that retinoids promote ubiquitination and degradation of cyclin D1 during retinoid-induced differentiation of human embryonal carcinoma cells. In response to all-trans-retinoic acid (RA) treatment, the human embryonal carcinoma cell line NT2/D1 exhibits a progressive decline in cyclin D1 expression beginning when the cells are committed to differentiate, but before onset of terminal neuronal differentiation. The decrease in cyclin D1 protein is tightly associated with the accumulation of hypophosphorylated forms of the retinoblastoma protein and G(1) arrest. In contrast, retinoic acid receptor gamma-deficient NT2/D1-R1 cells do not growth-arrest or accumulate in G(1) and have persistent cyclin D1 overexpression despite RA treatment. Notably, stable transfection of retinoic acid receptor gamma restores RA-mediated growth suppression and differentiation to NT2/D1-R1 cells and restores the decline of cyclin D1. The proteasome inhibitor LLnL blocks this RA-mediated decline in cyclin D1. RA treatment markedly accelerates ubiquitination of wild-type cyclin D1, but not a cyclin D1 (T286A) mutant. Transient expression of cyclin D1 (T286A) in NT2/D1 cells blocks RA-mediated transcriptional decline of a differentiation-sensitive reporter plasmid and represses induction of immunophenotypic neuronal markers. Taken together, these findings strongly implicate RA-mediated degradation of cyclin D1 as a means of coupling induced differentiation and cell cycle control of human embryonal carcinoma cells.

Overexpression of cyclins D1 and E is frequent in bronchial preneoplasia and precedes squamous cell carcinoma development.

Increased protein expression of the G1 cyclins D1 and E is reported in invasive non-small cell lung carcinoma. However, during transformation of the bronchial epithelium, overexpression of these species occurs, and their relationship to aberrant expression of p53 and retinoblastoma (Rb) has not been described previously. To determine the expression of these cell cycle regulators during the development of invasive squamous cell carcinoma (SCC) of the lung, the immunohistochemical expression patterns in normal bronchial epithelium (n = 36), squamous metaplasia (SM; n = 28), and epithelial atypia (n = 34) were compared with that in low-grade dysplasia (LGD; n = 17), high-grade bronchial dysplasia (HGD; n = 30), and SCC (n = 36). Monoclonal anti-p53 Pab1801, polyclonal anti-cyclin D1 DCS6, monoclonal anti-cyclin E HE12, and monoclonal anti-Rb OP-66 antibodies were used. Cyclin D1 was not expressed in normal bronchial epithelium but was detected in 7% of SMs, 15% of atypias; 18% of LGDs, 47% of HGDs, and 42% of SCCs. Cyclin E was not detected in normal epithelium (n = 24), SM (n = 16), or LGD (n = 12), but it was found in 9% of atypias (2 of 22), 33% of HGDs (7 of 21), and 54% of SCCs (13 of 24). p53 was not expressed in normal epithelium, SM, and LGD, but it was overexpressed in 6% of atypias, 53% of HGDs, and 61% of SCCs. Abnormal Rb expression was found only in 2 of 36 cases of SCC. A total of 91% of HGDs and 92% of SCCs exhibited overexpression of at least one of the p53, cyclin D1, or cyclin E species. However, no link was observed between overexpression of p53 and the overexpressed G1 cyclins in preneoplastic lesions. Overexpression of cyclin D1, cyclin E, and p53 occurs frequently and independently in pulmonary SCC and is detected in lesions before the development of invasive carcinoma. In contrast, altered Rb expression is a late and infrequent event in squamous cell carcinogenesis.

Interpreting patterns of gene expression with self-organizing maps: methods and application to hematopoietic differentiation.

Array technologies have made it straightforward to monitor simultaneously the expression pattern of thousands of genes. The challenge now is to interpret such massive data sets. The first step is to extract the fundamental patterns of gene expression inherent in the data. This paper describes the application of self-organizing maps, a type of mathematical cluster analysis that is particularly well suited for recognizing and classifying features in complex, multidimensional data. The method has been implemented in a publicly available computer package, GENECLUSTER, that performs the analytical calculations and provides easy data visualization. To illustrate the value of such analysis, the approach is applied to hematopoietic differentiation in four well studied models (HL-60, U937, Jurkat, and NB4 cells). Expression patterns of some 6,000 human genes were assayed, and an online database was created. GENECLUSTER was used to organize the genes into biologically relevant clusters that suggest novel hypotheses about hematopoietic differentiation-for example, highlighting certain genes and pathways involved in "differentiation therapy" used in the treatment of acute promyelocytic leukemia.

Cyclin D1 proteolysis: a retinoid chemoprevention signal in normal, immortalized, and transformed human bronchial epithelial cells.

BACKGROUND: Retinoids (derivatives of vitamin A) are reported to reduce the occurrence of some second primary cancers, including aerodigestive tract tumors. In contrast, beta-carotene does not reduce the occurrence of primary aerodigestive tract cancers. Mechanisms explaining these effective retinoid and ineffective carotenoid chemoprevention results are poorly defined. Recently, the all-trans-retinoic acid (RA)-induced proteolysis of cyclin D1 that leads to the arrest of cells in G1 phase of the cell cycle was described in human bronchial epithelial cells and is a promising candidate for such a mechanism. In this study, we have investigated this proteolysis as a common signal used by carotenoids or receptor-selective and receptor-nonselective retinoids. METHODS: We treated cultured normal human bronchial epithelial cells, immortalized human bronchial epithelial cells (BEAS-2B), and transformed human bronchial epithelial cells (BEAS-2BNNK) with receptor-selective or receptor-nonselective retinoids or with carotenoids and studied the effects on cell proliferation by means of tritiated thymidine incorporation and on cyclin D1 expression by means of immunoblot analysis. We also examined whether calpain inhibitor I, an inhibitor of the 26S proteasome degradation pathway, affected the decline (i.e., proteolysis) of cyclin D1. RESULTS: Receptor-nonselective retinoids were superior to the carotenoids studied in mediating the decline in cyclin D1 expression and in suppressing the growth of bronchial epithelial cells. Retinoids that activated retinoic acid receptor beta or retinoid X receptor pathways preferentially led to a decrease in the amount of cyclin D1 protein and a corresponding decline in growth. The retinoid-mediated degradation of cyclin D1 was blocked by cotreatment with calpain inhibitor I. CONCLUSIONS: Retinoid-dependent cyclin D1 proteolysis is a common chemoprevention signal in normal and neoplastic human bronchial epithelial cells. In contrast, carotenoids did not affect cyclin D1 expression. Thus, the degradation of cyclin D1 is a candidate intermediate marker for effective retinoid-mediated cancer chemoprevention in the aerodigestive tract.