Identification of Methylated Genes Associated with Aggressive Clinicopathological Features in Mantle Cell Lymphoma

Background: Mantle cell lymphoma (MCL) is genetically characterized by the t(11;14)(q13;q32) translocation and a high number of secondary chromosomal alterations. The contribution of DNA methylation to MCL lymphomagenesis is not well known. We sought to identify epigenetically silenced genes in these tumours that might have clinical relevance. Methodology/Principal Findings: To identify potential methylated genes in MCL we initially investigated seven MCL cell lines treated with epigenetic drugs and gene expression microarray profiling. The methylation status of selected candidate genes was validated by a quantitative assay and subsequently analyzed in a series of primary MCL (n=38). After pharmacological reversion we identified 252 potentially methylated genes. The methylation analysis of a subset of these genes (n=25) in the MCL cell lines and normal B lymphocytes confirmed that 80% of them were methylated in the cell lines but not in normal lymphocytes. The subsequent analysis in primary MCL identified five genes (SOX9,HOXA9,AHR,NR2F2 ,and ROBO1) frequently methylated in these tumours. The gene methylation events tended to occur in the same primary neoplasms and correlated with higher proliferation, increased number of chromosomal abnormalities, and shorter survival of the patients. Conclusions: We have identified a set of genes whose methylation degree and gene expression levels correlate with aggressive clinicopathological features of MCL. Our findings also suggest that a subset of MCL might show a CpG island methylator phenotype (CIMP) that may influence the behaviour of the tumours.

Methylation status of ANAPC1, CDKN2A and TP53 promoter genes in individuals with gastric cancer

Gastric cancer is the forth most frequent malignancy and the second most common cause of cancer death worldwide. DNA methylation is the most studied epigenetic alteration, occurring through a methyl radical addition to the cytosine base adjacent to guanine. Many tumor genes are inactivated by DNA methylation in gastric cancer. We evaluated the DNA methylation status of ANAPC1, CDKN2A and TP53 by methylation-specific PCR in 20 diffuse- and 26 intestinal-type gastric cancer samples and 20 normal gastric mucosa in individuals from Northern Brazil. All gastric cancer samples were advanced stage adenocarcinomas. Gastric samples were surgically obtained at the João de Barros Barreto University Hospital, State of Pará, and were stored at -80°C before DNA extraction. Patients had never been submitted to chemotherapy or radiotherapy, nor did they have any other diagnosed cancer. None of the gastric cancer samples presented methylated DNA sequences for ANAPC1 and TP53. CDKN2A methylation was not detected in any normal gastric mucosa; however, the CDKN2A promoter was methylated in 30.4% of gastric cancer samples, with 35% methylation in diffuse-type and 26.9% in intestinal-type cancers. CDKN2A methylation was associated with the carcinogenesis process for ~30% diffuse-type and intestinal-type compared to non-neoplastic samples. Thus, ANAPC1 and TP53 methylation was probably not implicated in gastric carcinogenesis in our samples. CDKN2A can be implicated in the carcinogenesis process of only a subset of gastric neoplasias.

Gene duplication and an accelerated evolutionary rate in 11S globulin genes are associated with higher protein synthesis in dicots as compared to monocots

Background: Seed storage proteins are a major source of dietary protein, and the content of such proteins determines both the quantity and quality of crop yield. Significantly, examination of the protein content in the seeds of crop plants shows a distinct difference between monocots and dicots. Thus, it is expected that there are different evolutionary patterns in the genes underlying protein synthesis in the seeds of these two groups of plants. Results: Gene duplication, evolutionary rate and positive selection of a major gene family of seed storage proteins (the 11S globulin genes), were compared in dicots and monocots. The results, obtained from five species in each group, show more gene duplications, a higher evolutionary rate and positive selections of this gene family in dicots, which are rich in 11S globulins, but not in the monocots. Conclusion: Our findings provide evidence to support the suggestion that gene duplication and an accelerated evolutionary rate may be associated with higher protein synthesis in dicots as compared to monocots.

Adaptive evolution of four microcephaly genes and the evolution of brain size in anthropoid primates

The anatomical basis and adaptive function of the expansion in primate brain size have long been studied; however, we are only beginning to understand the genetic basis of these evolutionary changes. Genes linked to human primary microcephaly have received much attention as they have accelerated evolutionary rates along lineages leading to humans. However, these studies focus narrowly on apes, and the link between microcephaly gene evolution and brain evolution is disputed. We analyzed the molecular evolution of four genes associated with microcephaly (ASPM, CDK5RAP2, CENPJ, MCPH1) across 21 species representing all major clades of anthropoid primates. Contrary to prevailing assumptions, positive selection was not limited to or intensified along the lineage leading to humans. In fact we show that all four loci were subject to positive selection across the anthropoid primate phylogeny. We developed clearly defined hypotheses to explicitly test if selection on these loci was associated with the evolution of brain size. We found positive relationships between both CDK5RAP2 and ASPM and neonatal brain mass and somewhat weaker relationships between these genes and adult brain size. In contrast, there is no evidence linking CENPJ and MCPH1 to brain size evolution. The stronger association of ASPM and CDK5RAP2 evolution with neonatal brain size than with adult brain size is consistent with these loci having a direct effect on prenatal neuronal proliferation. These results suggest that primate brain size may have at least a partially conserved genetic basis. Our results contradict a previous study that linked adaptive evolution of ASPM to changes in relative cortex size; however, our analysis indicates that this conclusion is not robust. Our finding that the coding regions of two widely expressed loci has experienced pervasive positive selection in relation to a complex, quantitative developmental phenotype provides a notable counterexample to the commonly asserted hypothesis that cisregulatory regions play a dominant role in phenotypic evolution. Key words: ASPM, MCPH1, CDK5RAP2, CENPJ, brain, neurogenesis, primates.

No mutations found in exons of TP53, H-RAS and K-RAS genes in liver of male Wistar rats submitted to a medium-term chemical carcinogenesis assay

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Methylation status of ANAPC1, CDKN2A and TP53 promoter genes in individuals with gastric cancer

Gastric cancer is the forth most frequent malignancy and the second most common cause of cancer death worldwide. DNA methylation is the most studied epigenetic alteration, occurring through a methyl radical addition to the cytosine base adjacent to guanine. Many tumor genes are inactivated by DNA methylation in gastric cancer. We evaluated the DNA methylation status of ANAPC1, CDKN2A and TP53 by methylation-specific PCR in 20 diffuse- and 26 intestinal-type gastric cancer samples and 20 normal gastric mucosa in individuals from Northern Brazil. All gastric cancer samples were advanced stage adenocarcinomas. Gastric samples were surgically obtained at the João de Barros Barreto University Hospital, State of Pará, and were stored at -80°C before DNA extraction. Patients had never been submitted to chemotherapy or radiotherapy, nor did they have any other diagnosed cancer. None of the gastric cancer samples presented methylated DNA sequences for ANAPC1 and TP53. CDKN2A methylation was not detected in any normal gastric mucosa; however, the CDKN2A promoter was methylated in 30.4% of gastric cancer samples, with 35% methylation in diffuse-type and 26.9% in intestinal-type cancers. CDKN2A methylation was associated with the carcinogenesis process for ~30% diffuse-type and intestinal-type compared to non-neoplastic samples. Thus, ANAPC1 and TP53 methylation was probably not implicated in gastric carcinogenesis in our samples. CDKN2A can be implicated in the carcinogenesis process of only a subset of gastric neoplasias.

Genetic variability of inflammatory genes in the brazilian population

Inflammatory gene variants have been associated with several diseases, including cancer, diabetes, vascular diseases, neurodegenerative diseases, arthritis, and others. Therefore, determining the population genetic composition of inflammation-related genes can be useful for the determination of general risk, prognostic and therapeutic strategies to prevent or cure specific diseases. We have aimed to identify polymorphism genotype frequencies in genes related to the inflammatory response in the Brazilian population, namely, IjBL - 62AT, IjBL - 262CT, tumor necrosis factors alpha (TNFa) - 238GA, TNFa - 308GA, lymphotoxin-alpha (LTa) + 80AC, LTa + 252AG, FAS - 670AG, and FASL - 844TC, considering the white, black, and Pardo ethnicities of the Sa˜o Paulo State. Our results suggest that the Brazilian population is under a miscegenation process at the current time, since some genotypes are not in the Hardy–Weinberg equilibrium. In addition, we conclude that the Pardo ethnicity is derived from a complex mixture of ethnicities, including the native Indian population.

Absence of inactivating mutations and deletions in the DMRT1 and FGF9 genes in a large cohort of 46,XY patients with gonadal dysgenesis

Despite advances in our understanding of the mechanisms involved in sex determination and differentiation, the specific roles of many genes in these processes are not completely understood in humans. Both DMRT1 and FGF9 are among this group of genes. Dmrt1 controls germ cell differentiation, proliferation, migration and pluripotency and Sertoli cell proliferation and differentiation. Fgf9 has been considered a critical factor in early testicular development and germ cell survival in mice. We screened for the presence of DMRT1 and FGF9 mutations in 33 patients with 46,XY gonadal dysgenesis. No deletions in either DMRT1 or FGF9 were identified using the MLPA technique. Eight allelic variants of DMRT1 were identified, and in silico analysis suggested that the novel c.968-15insTTCTCTCT variant and the c.774G>C (rs146975077) variant could have potentially deleterious effects on the DMRT1 protein. Nine previously described FGF9 allelic variants and six different alleles of the 3' UTR microsatellite were identified. However, none of these DMRT1 or FGF9 variants was associated with increased 46,XY gonadal dysgenesis. In conclusion, our study suggests that neither DMRT1 nor FGF9 abnormalities are frequently involved in dysgenetic male gonad development in patients with non-syndromic 46,XY disorder of sex development. (C) 2012 Published by Elsevier Masson SAS.

Genome analysis of DNA repair genes in the alpha proteobacterium Caulobacter crescentus

Abstract Background The integrity of DNA molecules is fundamental for maintaining life. The DNA repair proteins protect organisms against genetic damage, by removal of DNA lesions or helping to tolerate them. DNA repair genes are best known from the gamma-proteobacterium Escherichia coli, which is the most understood bacterial model. However, genome sequencing raises questions regarding uniformity and ubiquity of these DNA repair genes and pathways, reinforcing the need for identifying genes and proteins, which may respond to DNA damage in other bacteria. Results In this study, we employed a bioinformatic approach, to analyse and describe the open reading frames potentially related to DNA repair from the genome of the alpha-proteobacterium Caulobacter crescentus. This was performed by comparison with known DNA repair related genes found in public databases. As expected, although C. crescentus and E. coli bacteria belong to separate phylogenetic groups, many of their DNA repair genes are very similar. However, some important DNA repair genes are absent in the C. crescentus genome and other interesting functionally related gene duplications are present, which do not occur in E. coli. These include DNA ligases, exonuclease III (xthA), endonuclease III (nth), O6-methylguanine-DNA methyltransferase (ada gene), photolyase-like genes, and uracil-DNA-glycosylases. On the other hand, the genes imuA and imuB, which are involved in DNA damage induced mutagenesis, have recently been described in C. crescentus, but are absent in E. coli. Particularly interesting are the potential atypical phylogeny of one of the photolyase genes in alpha-proteobacteria, indicating an origin by horizontal transfer, and the duplication of the Ada orthologs, which have diverse structural configurations, including one that is still unique for C. crescentus. Conclusion The absence and the presence of certain genes are discussed and predictions are made considering the particular aspects of the C. crescentus among other known DNA repair pathways. The observed differences enlarge what is known for DNA repair in the Bacterial world, and provide a useful framework for further experimental studies in this organism.

Comparative DNA sequence and methylation analyses of orthologous genes in humans and non-human primates: Genetic and epigenetic footprints of evolution

Welche genetische Unterschiede machen uns verschieden von unseren nächsten Verwandten, den Schimpansen, und andererseits so ähnlich zu den Schimpansen? Was wir untersuchen und auch verstehen wollen, ist die komplexe Beziehung zwischen den multiplen genetischen und epigenetischen Unterschieden, deren Interaktion mit diversen Umwelt- und Kulturfaktoren in den beobachteten phänotypischen Unterschieden resultieren. Um aufzuklären, ob chromosomale Rearrangements zur Divergenz zwischen Mensch und Schimpanse beigetragen haben und welche selektiven Kräfte ihre Evolution geprägt haben, habe ich die kodierenden Sequenzen von 2 Mb umfassenden, die perizentrischen Inversionsbruchpunkte flankierenden Regionen auf den Chromosomen 1, 4, 5, 9, 12, 17 und 18 untersucht. Als Kontrolle dienten dabei 4 Mb umfassende kollineare Regionen auf den rearrangierten Chromosomen, welche mindestens 10 Mb von den Bruchpunktregionen entfernt lagen. Dabei konnte ich in den Bruchpunkten flankierenden Regionen im Vergleich zu den Kontrollregionen keine höhere Proteinevolutionsrate feststellen. Meine Ergebnisse unterstützen nicht die chromosomale Speziationshypothese für Mensch und Schimpanse, da der Anteil der positiv selektierten Gene (5,1% in den Bruchpunkten flankierenden Regionen und 7% in den Kontrollregionen) in beiden Regionen ähnlich war. Durch den Vergleich der Anzahl der positiv und negativ selektierten Gene per Chromosom konnte ich feststellen, dass Chromosom 9 die meisten und Chromosom 5 die wenigsten positiv selektierten Gene in den Bruchpunkt flankierenden Regionen und Kontrollregionen enthalten. Die Anzahl der negativ selektierten Gene (68) war dabei viel höher als die Anzahl der positiv selektierten Gene (17). Eine bioinformatische Analyse von publizierten Microarray-Expressionsdaten (Affymetrix Chip U95 und U133v2) ergab 31 Gene, die zwischen Mensch und Schimpanse differentiell exprimiert sind. Durch Untersuchung des dN/dS-Verhältnisses dieser 31 Gene konnte ich 7 Gene als negativ selektiert und nur 1 Gen als positiv selektiert identifizieren. Dieser Befund steht im Einklang mit dem Konzept, dass Genexpressionslevel unter stabilisierender Selektion evolvieren. Die meisten positiv selektierten Gene spielen überdies eine Rolle bei der Fortpflanzung. Viele dieser Speziesunterschiede resultieren eher aus Änderungen in der Genregulation als aus strukturellen Änderungen der Genprodukte. Man nimmt an, dass die meisten Unterschiede in der Genregulation sich auf transkriptioneller Ebene manifestieren. Im Rahmen dieser Arbeit wurden die Unterschiede in der DNA-Methylierung zwischen Mensch und Schimpanse untersucht. Dazu wurden die Methylierungsmuster der Promotor-CpG-Inseln von 12 Genen im Cortex von Menschen und Schimpansen mittels klassischer Bisulfit-Sequenzierung und Bisulfit-Pyrosequenzierung analysiert. Die Kandidatengene wurden wegen ihrer differentiellen Expressionsmuster zwischen Mensch und Schimpanse sowie wegen Ihrer Assoziation mit menschlichen Krankheiten oder dem genomischen Imprinting ausgewählt. Mit Ausnahme einiger individueller Positionen zeigte die Mehrzahl der analysierten Gene keine hohe intra- oder interspezifische Variation der DNA-Methylierung zwischen den beiden Spezies. Nur bei einem Gen, CCRK, waren deutliche intraspezifische und interspezifische Unterschiede im Grad der DNA-Methylierung festzustellen. Die differentiell methylierten CpG-Positionen lagen innerhalb eines repetitiven Alu-Sg1-Elements. Die Untersuchung des CCRK-Gens liefert eine umfassende Analyse der intra- und interspezifischen Variabilität der DNA-Methylierung einer Alu-Insertion in eine regulatorische Region. Die beobachteten Speziesunterschiede deuten darauf hin, dass die Methylierungsmuster des CCRK-Gens wahrscheinlich in Adaption an spezifische Anforderungen zur Feinabstimmung der CCRK-Regulation unter positiver Selektion evolvieren. Der Promotor des CCRK-Gens ist anfällig für epigenetische Modifikationen durch DNA-Methylierung, welche zu komplexen Transkriptionsmustern führen können. Durch ihre genomische Mobilität, ihren hohen CpG-Anteil und ihren Einfluss auf die Genexpression sind Alu-Insertionen exzellente Kandidaten für die Förderung von Veränderungen während der Entwicklungsregulation von Primatengenen. Der Vergleich der intra- und interspezifischen Methylierung von spezifischen Alu-Insertionen in anderen Genen und Geweben stellt eine erfolgversprechende Strategie dar.

Identification of target genes of the T-box proteins in Drosophila melanogaster

In a prior bioinformatic analysis by Hüyseyin Binbas, potential Tbx targets sequences in wing-related genes have been identified. Guided by this information, enhancer trap/reporter lacZ insertions were characterized by X-gal staining first in wildtype and then in l(1)omb imaginal discs.rnIn several lines I observed an increase in reporter expression in a l(1)omb mutant background. Since Omb is assumed to function predominantly as a transcriptional repressor, this may indicate direct regulation. Repression by Omb was observed e.g. for brk and tkv. These genes are negatively regulated by Dpp, while omb is induced by Dpp. Omb which mediates the effects of Dpp on proliferation could, thus, also mediate the Dpp effect on patterning of the wing disc. However, brk and tkv were not completely derepressed in l(1)omb indicating that Dpp represses these genes also by an Omb-independent mechanism.rnMore frequently I observed loss of reporter expression in an l(1)omb mutant background. In these cases, regulation by Omb presumably is indirect. For example, STAT92E-lacZ expression in the wildtype eye was symmetrically expressed at the dorsal and ventral margins. In l(1)omb, ventral expression was selectively lost. Loss of omb is known to cause ventral overproliferation of the eye by activation of the Jak/STAT pathway. STAT92E expression is negatively regulated by Jak/STAT signaling suggesting that loss of omb activates Jak/STAT further upstream in the pathway.rnRegional overproliferation of eye and wing in the l(1)omb mutant background proved a complicating issue in the search for Omb targets. This effect made it difficult to decide whether an expanded reporter expression pattern was due to tissue expansion or reporter gene derepression. For instance hth-lacZ appeared to expand along the ventral eye disc margin in l(1)omb. Without addtional experiments it cannot be concluded whether this is due to de-repression or to activation in association with the proliferative state. Parallel to my experiments, evidence accumulated in our laboratory that loss of omb may attenuate Wg and Hegehog signaling. Since these diffusible proteins are the main patterning molecules in the wing imaginal disc, with dpp being downstream of Hh, many of the observed effects could be secondary to reduced Wg and Hh activity. Examples are ab-lacZ, Dll-lacZ and vgBE-lacZ (reduced expression on the dorso-ventral boundary) and inv-lacZ (late larval expression in the anterior wing disc compartment is lost) or sal-lacZ. Epistasis experiment will be required to clarifiy these issues.rnFurthermore, loss of omb appeared to induce cell fate changes. It was reported previously that in an omb null mutant, the dorsal determinant apterous (ap) is ectopically expressed in the ventral compartment (an effect I did not observe with the strongly hypomorphic l(1)omb15, indicating strong dose dependence). Ventral repression of ap is maintained by epigenetic mechanisms. The patchy and variable nature of ectopic expression of ap or grn-1.1-lacZ points to an effect of omb on epigenetic stability.rnIn the second part of my thesis, an analysis of Omb expression in the Drosophila embryonic ventral nervous system was performed. Omb was found co-expressed with Eve in the medial aCC and RP2 motorneurons as well as the fpCC interneuron and the mediolateral CQ neurons. Additionally, Omb was detected in the Eg positive NB7-3 GW serotonergic motoneuron and the N2-4 neurons. Omb was not found in Repo positive glial cells. During embryonic stage 14, Omb showed some coepression with Dpn or Pros. At the embryonic stage 16, Omb was expressed in minor subset of Mid and Wg positive cells.

Expression profiling of stem cell-related genes in neoadjuvant-treated gastric cancer: a NOTCH2, GSK3B and β-catenin gene signature predicts survival

Cancer stem cell (CSC) based gene expression signatures are associated with prognosis in various tumour types and CSCs are suggested to be particularly drug resistant. The aim of our study was first, to determine the prognostic significance of CSC-related gene expression in residual tumour cells of neoadjuvant-treated gastric cancer (GC) patients. Second, we wished to examine, whether expression alterations between pre- and post-therapeutic tumour samples exist, consistent with an enrichment of drug resistant tumour cells. The expression of 44 genes was analysed in 63 formalin-fixed, paraffin embedded tumour specimens with partial tumour regression (10-50% residual tumour) after neoadjuvant chemotherapy by quantitative real time PCR low-density arrays. A signature of combined GSK3B(high), β-catenin (CTNNB1)(high) and NOTCH2(low) expression was strongly correlated with better patient survival (p<0.001). A prognostic relevance of these genes was also found analysing publically available gene expression data. The expression of 9 genes was compared between pre-therapeutic biopsies and post-therapeutic resected specimens. A significant post-therapeutic increase in NOTCH2, LGR5 and POU5F1 expression was found in tumours with different tumour regression grades. No significant alterations were observed for GSK3B and CTNNB1. Immunohistochemical analysis demonstrated a chemotherapy-associated increase in the intensity of NOTCH2 staining, but not in the percentage of NOTCH2. Taken together, the GSK3B, CTNNB1 and NOTCH2 expression signature is a novel, promising prognostic parameter for GC. The results of the differential expression analysis indicate a prominent role for NOTCH2 and chemotherapy resistance in GC, which seems to be related to an effect of the drugs on NOTCH2 expression rather than to an enrichment of NOTCH2 expressing tumour cells.

COCAINE DEPENDENCE: THE ROLE OF SEROTONIN GENES IN

Animal studies have shown that behavioral responses to cocaine-related cues are altered by serotonergic medications. The effects of pharmacological agents on serotonin receptors 2a (5-HT2A) and 2c (5-HT2C), have yielded results suggesting that selective 5-HT2A antagonists and 5-HT2C agonists promote the disruption of cocaine-associated memories. One measure of cocaine related cues in humans is attentional bias, in which cocaine dependent individuals show greater response latency for cocaine related words than neutral words. Data from our laboratory shows that cocaine dependent subjects have altered attentional bias compared to controls. The purpose of this thesis was to investigate the role of the serotonin system in attentional bias and impulsivity in cocaine dependent individuals. We focused on the serotonin transporter, serotonin receptors 2A and 2C and tryptophan hydroxylase 1 and 2 (TPH1 and TPH2). We predicted that attentional bias and impulsivity would be higher in cocaine dependent individuals who had lower serotonin function. In the current study, we found a significant association between TPH2 genotype and attentional bias for the second block of the cocaine Stroop task. There was also a significant association between average attentional bias and HTTLPR genotype in the cocaine dependent individuals. The HT2C receptor genotype and attentional bias in our study sample also showed a significant difference. We did not find a significant difference between the serotonin 2A receptor variants or the TPH1 variants and attentional bias in the cocaine dependent group. In conclusion, the current study suggests that serotonergic medications should be utilized as pharmacotherapeutic treatment for cocaine addiction. Our results indicate that TPH2, the serotonin transporter and 2C receptor should be targeted in such a way as to modulate both, leading to increased synaptic serotonin function.

Computational identification and functional validation of regulatory motifs in cartilage-expressed genes.

Chondrocyte gene regulation is important for the generation and maintenance of cartilage tissues. Several regulatory factors have been identified that play a role in chondrogenesis, including the positive transacting factors of the SOX family such as SOX9, SOX5, and SOX6, as well as negative transacting factors such as C/EBP and delta EF1. However, a complete understanding of the intricate regulatory network that governs the tissue-specific expression of cartilage genes is not yet available. We have taken a computational approach to identify cis-regulatory, transcription factor (TF) binding motifs in a set of cartilage characteristic genes to better define the transcriptional regulatory networks that regulate chondrogenesis. Our computational methods have identified several TFs, whose binding profiles are available in the TRANSFAC database, as important to chondrogenesis. In addition, a cartilage-specific SOX-binding profile was constructed and used to identify both known, and novel, functional paired SOX-binding motifs in chondrocyte genes. Using DNA pattern-recognition algorithms, we have also identified cis-regulatory elements for unknown TFs. We have validated our computational predictions through mutational analyses in cell transfection experiments. One novel regulatory motif, N1, found at high frequency in the COL2A1 promoter, was found to bind to chondrocyte nuclear proteins. Mutational analyses suggest that this motif binds a repressive factor that regulates basal levels of the COL2A1 promoter.

Activation of RARα induces autophagy in SKBR3 breast cancer cells and depletion of key autophagy genes enhances ATRA toxicity

All-trans retinoic acid (ATRA), a pan-retinoic acid receptor (RAR) agonist, is, along with other retinoids, a promising therapeutic agent for the treatment of a variety of solid tumors. On the one hand, preclinical studies have shown promising anticancer effects of ATRA in breast cancer; on the other hand, resistances occurred. Autophagy is a cellular recycling process that allows the degradation of bulk cellular contents. Tumor cells may take advantage of autophagy to cope with stress caused by anticancer drugs. We therefore wondered if autophagy is activated by ATRA in mammary tumor cells and if modulation of autophagy might be a potential novel treatment strategy. Indeed, ATRA induces autophagic flux in ATRA-sensitive but not in ATRA-resistant human breast cancer cells. Moreover, using different RAR agonists as well as RARα-knockdown breast cancer cells, we demonstrate that autophagy is dependent on RARα activation. Interestingly, inhibition of autophagy in breast cancer cells by either genetic or pharmacological approaches resulted in significantly increased apoptosis under ATRA treatment and attenuated epithelial differentiation. In summary, our findings demonstrate that ATRA-induced autophagy is mediated by RARα in breast cancer cells. Furthermore, inhibition of autophagy results in enhanced apoptosis. This points to a potential novel treatment strategy for a selected group of breast cancer patients where ATRA and autophagy inhibitors are applied simultaneously.