Differential proteomic comparison of breast cancer secretome using a quantitative paired analysis workflow.

BACKGROUND: Worldwide, breast cancer is the main cause of cancer mortality in women. Most cases originate in mammary ductal cells that produce the nipple aspirate fluid (NAF). In cancer patients, this secretome contains proteins associated with the tumor microenvironment. NAF studies are challenging because of inter-individual variability. We introduced a paired-proteomic shotgun strategy that relies on NAF analysis from both breasts of patients with unilateral breast cancer and extended PatternLab for Proteomics software to take advantage of this setup. METHODS: The software is based on a peptide-centric approach and uses the binomial distribution to attribute a probability for each peptide as being linked to the disease; these probabilities are propagated to a final protein p-value according to the Stouffer's Z-score method. RESULTS: A total of 1227 proteins were identified and quantified, of which 87 were differentially abundant, being mainly involved in glycolysis (Warburg effect) and immune system activation (activated stroma). Additionally, in the estrogen receptor-positive subgroup, proteins related to the regulation of insulin-like growth factor transport and platelet degranulation displayed higher abundance, confirming the presence of a proliferative microenvironment. CONCLUSIONS: We debuted a differential bioinformatics workflow for the proteomic analysis of NAF, validating this secretome as a treasure-trove for studying a paired-organ cancer type.

Proteomic profiling of nipple aspirate fluid (NAF): Exploring the complementarity of different peptide fractionation strategies.

NAF is a breast fluid that is closely related to the tumor microenvironment and a valuable sample for studying breast cancer. To perform an in-depth proteomic analysis of this sample, aliquots of a single NAF digest were analyzed by the following peptide-centric fractionation strategies: a) 30-cm reversed-phase (RP) column on-line with an LTQ-Orbitrap XL; b) off-line strong cation-exchange (SCX) column; and c) pI-based OFFGEL fractionation. All fractions from approaches (b) and (c) were further analyzed on a 10-cm RP column hyphenated to the same mass spectrometer. The RP-30cm, SCX/RP-10cm, and OFFGEL/RP-10cm approaches identified 1676, 2930, and 3240 peptides, which corresponded to 193, 390 and 528 proteins, respectively. In our cumulative dataset, 4466 distinct NAF peptides corresponded to a total of 557 proteins, of which only 34% were identified by all three approaches. No exclusive protein identification was associated to the RP-30cm approach, while SCX/RP-10cm and OFFGEL/RP-10cm contributed to 28 and 166 exclusive identifications, respectively. Each approach provided additional information related to energy metabolism (fermentation process/carbohydrate biosynthesis). In conclusion, the pre-fractionation platforms used were complementary for the comprehensive characterization of NAF and our work provides methodological information for future quantitative cancer-related NAF sample studies. BIOLOGICAL SIGNIFICANCE: High-resolution peptide separation is a sine qua non condition for achieving extensive proteome coverage. Various techniques have been employed to improve peptide fractionation prior to LC-MS/MS, thus allowing a comprehensive characterization of complex biological samples. Although fractionation efficiency is very sample-dependent, this issue is commonly overlooked, and a "cookbook" approach is routinely used during this critical step. The present study provides a systematic comparison of analytical information needed for the successful large-scale differential proteomic analysis of NAF samples from breast cancer patients, a precious and volume-limited biological sample. It reinforces the importance of optimizing sample-specific fractionation protocols for information retrieval from mass spectrometric analysis.

Epigenetic modifications, chromatin distribution and TP53 transcription in a model of breast cancer progression.

In the present paper we aimed to characterize epigenetic aspects and analyze TP53 transcription in the 21 T series, composed of breast cell lines: non-cancerous H16N2; Atypical Ductal Hyperplasia 21PT; Ductal Carcinoma in situ 21NT and Invasive Metastatic Carcinoma 21MT1. We detected a global genomic hypomethylation in 21NT and 21MT1. The histone modification markers analysis showed an important global decrease of the active chromatin mark H4Ac in 21MT1 relative to the other cell lines while the repressive mark H3K9Me3 were not significantly altered. The mRNA levels of DNA methylation and histone modification key enzymes are consistent with the observed genomic hypomethylation and histone hypoacetylation. The expression of DNMT3A/B increased at the initial stages of oncogenesis and the expression of DNMT1 and HAT1 decreased at the advanced stages of breast cancer. Using a confocal immunofluorescent assay, we observed that H4Ac was mostly located at the periphery and the repressive mark H3K9Me3, at the center of 21NT and 21MT1 cells nuclei. TP53 P1 promoter was found to be in an open chromatin state, with a relatively high enrichment of H4Ac and similar TP53 transcription levels in all 21 T cell lines. In conclusion, we observed epigenetic alterations (global genome hypomethylation, global hypoacetylation and accumulation of pericentric heterochromatin) in metastatic breast cancer cells of the 21 T series. These alterations may act at later stages of breast cancer progression and may not affect TP53 transcription at the P1 promoter.

Prion-like aggregation of mutant p53 in cancer.

p53 is a master regulatory protein that participates in cellular processes such as apoptosis, DNA repair, and cell cycle control. p53 functions as a homotetrameric tumor suppressor, and is lost in more than 50% of human cancers. Recent studies have suggested that the formation of mutant p53 aggregates is associated with loss-of-function (LoF), dominant-negative (DN), and gain-of-function (GoF) effects. We propose that these phenomena can be explained by a prion-like behavior of mutant p53. We discuss the shared properties of cancer and neurodegenerative diseases and how the prion-like properties of p53 aggregates offer potential targets for drug development.

MiR-34a is up-regulated in response to low dose, low energy X-ray induced DNA damage in breast cells.

BACKGROUND: MicroRNAs are non-coding RNAs involved in the regulation of gene expression including DNA damage responses. Low doses of low energy X-ray radiation, similar to those used in mammographic exams, has been described to be genotoxic. In the present work we investigated the expression of miR-34a; a well described p53-regulated miRNA implicated in cell responses to X-ray irradiation at low doses. METHODS: Non-cancerous breast cell line MCF-10A and cancerous T-47D and MCF-7 cell lines were submitted to a low-energy X-ray irradiation (ranging from 28-30 Kv) using a dose of 5 Gy. The expression level of miR-34a, let-7a and miR-21 was assessed by qRT-PCR at 4 and 24 hours post-irradiation. DNA damage was then measured by comet assay and micronuclei estimation in MCF-10A and MCF-7 cell lines, where an increase of miR-34a levels could be observed after irradiation. The rate of apoptotic cells was estimated by nuclear staining and fluorescence microscopy. These experiments were also performed at low doses (3; 12 and 48 mGy) in MCF-10A and MCF-7 cell lines. RESULTS: We have observed an increase in miR-34a expression 4 hours post-irradiation at 5 Gy in MCF-10A and MCF-7 cell lines while its level did not change in T-47D, a breast cancer cell line bearing non-functional p53. At low doses, miR-34a was up-regulated in non-tumoral MCF-10A to a higher extent as compared to MCF-7. MiR-34a levels decreased 24 hours post-irradiation. We have also observed DNA damage and apoptosis at low-energy X-ray irradiation at low doses and the high dose in MCF-10A and MCF-7 4 and 24 hours post-irradiation relative to the mock control. CONCLUSION: Low energy X-ray is able to promote DNA strand breaks and miR-34a might be involved in cell responses to low energy X-ray DNA damage. MiR-34a expression correlates with X-ray dose, time after irradiation and cell type. The present study reinforces the need of investigating consequences of low dose X-ray irradiation of breast cells.

Expanding the prion concept to cancer biology: dominant-negative effect of aggregates of mutant p53 tumour suppressor.

p53 is a key protein that participates in cell-cycle control, and its malfunction can lead to cancer. This tumour suppressor protein has three main domains; the N-terminal transactivation domain, the CTD (C-terminal domain) and the core domain (p53C) that constitutes the sequence-specific DBD (DNA-binding region). Most p53 mutations related to cancer development are found in the DBD. Aggregation of p53 into amyloid oligomers and fibrils has been shown. Moreover, amyloid aggregates of both the mutant and WT (wild-type) forms of p53 were detected in tumour tissues. We propose that if p53 aggregation occurred, it would be a crucial aspect of cancer development, as p53 would lose its WT functions in an aggregated state. Mutant p53 can also exert a dominant-negative regulatory effect on WT p53. Herein, we discuss the dominant-negative effect in light of p53 aggregation and the fact that amyloid-like mutant p53 can convert WT p53 into more aggregated species, leading into gain of function in addition to the loss of tumour suppressor function. In summary, the results obtained in the last decade indicate that cancer may have characteristics in common with amyloidogenic and prion diseases.

Mutant p53 aggregates into prion-like amyloid oligomers and fibrils: implications for cancer.

Over 50% of all human cancers lose p53 function. To evaluate the role of aggregation in cancer, we asked whether wild-type (WT) p53 and the hot-spot mutant R248Q could aggregate as amyloids under physiological conditions and whether the mutant could seed aggregation of the wild-type form. The central domains (p53C) of both constructs aggregated into a mixture of oligomers and fibrils. R248Q had a greater tendency to aggregate than WT p53. Full-length p53 aggregated into amyloid-like species that bound thioflavin T. The amyloid nature of the aggregates was demonstrated using x-ray diffraction, electron microscopy, FTIR, dynamic light scattering, cell viabilility assay, and anti-amyloid immunoassay. The x-ray diffraction pattern of the fibrillar aggregates was consistent with the typical conformation of cross ß-sheet amyloid fibers with reflexions of 4.7 Å and 10 Å. A seed of R248Q p53C amyloid oligomers and fibrils accelerated the aggregation of WT p53C, a behavior typical of a prion. The R248Q mutant co-localized with amyloid-like species in a breast cancer sample, which further supported its prion-like effect. A tumor cell line containing mutant p53 also revealed massive aggregation of p53 in the nucleus. We conclude that aggregation of p53 into a mixture of oligomers and fibrils sequestrates the native protein into an inactive conformation that is typical of a prionoid. This prion-like behavior of oncogenic p53 mutants provides an explanation for the negative dominance effect and may serve as a potential target for cancer therapy.

Loop domain organization of the p53 locus in normal and breast cancer cells correlates with the transcriptional status of the TP53 and the neighboring genes.

P53 is a tumor suppressor protein critical for genome integrity. Although its control at the protein level is well known, the transcriptional regulation of the TP53 gene is still unclear. We have analyzed the organization of the TP53 gene domain using DNA arrays in several breast cancer and control cell lines. We have found that in the control breast epithelial cell line, HB2, the TP53 gene is positioned within a relatively small DNA domain, encompassing 50 kb, delimited by two nuclear matrix attachment sites. Interestingly, this domain structure was found to be radically different in the studied breast cancer cell lines, MCF7, T47D, MDA-MB-231, and BT474, in which the domain size was increased and TP53 transcription was decreased. We propose a model in which the organization of the TP53 gene domain correlates with the transcriptional status of TP53 and neighboring genes.

Co-localization of mutant p53 and amyloid-like protein aggregates in breast tumors.

P53 is one of the most important tumor suppressor proteins in human cancers. Mutations in the TP53 gene are common features of malignant tumors and normally correlate to a more aggressive disease. In breast cancer, these gene alterations are present in approximately 20% of cases and are characteristically of missense type. In the present work we describe TP53 mutations in breast cancer biopsies and investigate whether wild and mutant p53 participate in protein aggregates formation in these breast cancer cases. We analyzed 88 biopsies from patients residing in the metropolitan area of Rio de Janeiro, and performed TP53 mutation screening using direct sequencing of exons 5-10. Seventeen mutations were detected, 12 of them were of missense type, 2 nonsenses, 2 deletions and 1 insertion. The presence of TP53 mutation was highly statistically associated to tumor aggressiveness of IDC cases, indicated here by Elston Grade III (p<0.0001). Paraffin embedded breast cancer tissues were analyzed for the presence of p53 aggregates through immunofluorescence co-localization assay, using anti-aggregate primary antibody A11, and anti-p53. Our results show that mutant p53 co-localizes with amyloid-like protein aggregates, depending on mutation type, suggesting that mutant p53 may form aggregates in breast cancer cells, in vivo.

Cognate DNA stabilizes the tumor suppressor p53 and prevents misfolding and aggregation.

The tumor suppressor protein p53 is a nuclear protein that serves as an important transcription factor. The region responsible for sequence-specific DNA interaction is located in its core domain (p53C). Although full-length p53 binds to DNA as a tetramer, p53C binds as a monomer since it lacks the oligomerization domain. It has been previously demonstrated that two core domains have a dimerization interface and undergo conformational change when bound to DNA. Here we demonstrate that the interaction with a consensus DNA sequence provides the core domain of p53 with enhanced conformational stability at physiological salt concentrations (0.15 M). This stability could be either increased or abolished at low (0.01 M) or high (0.3 M) salt concentrations, respectively. In addition, interaction with the cognate sequence prevents aggregation of p53C into an amyloid-like structure, whereas binding to a nonconsensus DNA sequence has no effect on p53C stability, even at low ionic strength. Strikingly, sequence-specific DNA binding also resulted in a large stabilization of full-length p53, whereas nonspecific sequence binding led to no stabilization. The effects of cognate DNA could be mimicked by high concentrations of osmolytes such as glycerol, which implies that the stabilization is caused by the exclusion of water. Taken together, our results show an enhancement in protein stability driven by specific DNA recognition. When cognate DNA was added to misfolded protein obtained after a pressurization cycle, the original conformation was mostly recovered. Our results may aid the development of therapeutic approaches to prevent misfolded species of p53.

Interferon-inducible guanylate binding protein (GBP)-2: a novel p53-regulated tumor marker in esophageal squamous cell carcinomas.

TP53 mutations are common in esophageal squamous cell carcinomas (SCC). To identify biological markers of possible relevance in esophageal SCC, we (i) searched for genes expressed in a p53-dependent manner in TE-1, an esophageal SCC cell line expressing the temperature-sensitive TP53 mutant V272M, and (ii) investigated the expression of one of those genes, the interferon-inducible Guanylate Binding Protein 2 (GBP-2), in esophageal SCC tissues. Clontech Human Cancer 1.2 arrays containing 1,176 human cancer gene-related sequences were used to identify differentially expressed genes in TE-1 cells at permissive (32 degrees C) and nonpermissive (37 degrees C) temperatures. The expression of GBP-2 and IRF-1, its main transcriptional regulator, was analyzed by immunohistochemistry in a retrospective series of 41 esophageal SCC cases with a clear transition zone from noncancer, apparently normal epithelium to invasive cancer. The expression of the GBP-2 gene is consistently increased in TE-1 at 32 degrees C in a p53-dependent manner, as confirmed by inhibition of p53 expression by RNA interference. Increase in GBP-2 is accompanied by an increase in protein levels of IRF-1, the main transcriptional regulator of GBP-2, and in the formation of complexes between p53 and IRF-1. GBP-2 expression is significantly higher in esophageal SCC than in adjacent normal epithelium (p<0.01), in which GBP-2 staining is limited to the basal layer. Our results suggest that p53 up-regulates GBP-2 by cooperating with IRF-1. The association of GBP-2 expression with proliferative squamous cells suggests that GBP-2 may represent a marker of interest in esophageal SCC.

Cross-talks between cyclooxygenase-2 and tumor suppressor protein p53: Balancing life and death during inflammatory stress and carcinogenesis.

Overexpression of Cyclooxygenase-2 (COX-2) is observed in most tumor types. Increased COX-2 activity and synthesis of prostaglandins stimulates proliferation, angiogenesis, invasiveness and inhibits apoptosis. Many stress and proinflammatory signals induce COX-2 expression, including oxyradicals or DNA-damaging agents. The latter also induces p53, a transcription factor often inactivated by mutation in cancer. Several studies have identified complex cross-talks between p53 and COX-2, whereby p53 can either up- or down-regulate COX-2, which in turn controls p53 transcriptional activity. However, the molecular basis of these effects are open to debate, in particular since no p53 binding sequences have been identified in COX-2 regulatory regions. In this review, we summarize the molecular mechanisms by which COX-2 contributes to carcinogenesis and discuss the experimental set-up, results and conclusions of studies analyzing cross-talks between p53 and COX-2. We propose 2 scenarios accounting for overexpression of COX-2 in precursor and cancer lesions. In the "inflammatory" scenario, p53, activated by DNA damage induced by oxygen and nitrogen species, recruits NF-kappaB to activate COX-2, resulting in antiapoptotic effects that contribute to cell expansion in inflammatory precursor lesions. In the "constitutive proliferation" scenario, oncogenic stress due to activation of growth signaling cascades may upregulate COX-2 promoter independently of NF-kappaB and p53, synergizing with TP53 mutation to promote cancer progression. These 2 scenarios, although not mutually exclusive, may account for the diversity of the correlations between COX-2 expression and TP53 mutation, which vary according to cancer types and biological contexts, and have implications for the use of COX-2 inhibitors in cancer prevention and therapy.

CYP19 (TTTA)n polymorphism and breast cancer risk in Brazilian women.

A prolonged or increased exposure to endogenous estrogens associated with genetic factors are considered to be the main risk factors for breast cancer. The CYP19 gene encodes the enzyme aromatase, which catalyzes the conversion of androgens into estrogens. CYP19 alleles containing different numbers of tetranucleotide TTTA repeats in intron 4 have been associated with increased breast cancer risk. In this study we investigated, for the first time, the frequency of CYP19 (TTTA)n alleles in a South American population (n = 475) and analyzed the risk for developing breast cancer in a case-control study comprising 135 cases and 270 age-matched controls. It is shown that Brazilians possess not only the alleles identified in all the other populations studied so far (alleles containing from 7 to 13 TTTA repeats), but also the (TTTA)6 allele, that had never been described before. The (TTTA)10 allele was three times more frequent in cases when compared to controls and presented a significant positive association (p = 0.048) with breast cancer development in Brazilians.

TP53 mutations as biomarkers for cancer epidemiology in Latin America: current knowledge and perspectives.

Due to particular social and economical development, and to the impact of globalization of lifestyles, Latin America shows a superposition of cancers that are frequent in low resource countries (gastric, oesophageal squamous cell and cervical cancers) and high resource countries (cancers of breast, colon and rectum, lung and prostate). Latin America thus offers opportunities for investigating the impact on changing lifestyle patterns on the occurrence of cancer. At the molecular level, mutations in the tumor suppressor gene TP53 are common in many cancers and their distribution can be informative of the nature of the mutagenic mechanisms, thus giving clues to cancer etiology and molecular pathogenesis. However most of the data available are derived from studies in industrialized countries. In this review, we discuss current trends on cancer occurrence in Latin American countries, and we review the literature available on TP53 mutations and polymorphisms in patients from Latin America. Overall, a total of 285 mutations have been described in 1213 patients in 20 publications, representing 1.5% of the total number of mutations reported world-wide. Except for hematological cancers, TP53 mutation frequencies are similar to those reported in other regions of the world. The only tumor site presenting significant differences in mutation pattern as compared to other parts of the world is colon and rectum. However, this difference is based on a single study with 35 patients. Recently, a characteristic TP53 mutation at codon 337 (R337H) has been identified in the germline of children with adrenocortical carcinoma in Southern Brazil. Further and better focused analyses of TP53 mutation patterns in the context of epidemiological studies, should help to improve our understanding of cancer etiology in order to develop appropriate health policies and public health programs in Latin America.

Mechanisms of esophageal cancer development in Brazilians.

Esophageal cancer represents one of the most common and lethal cancers around the World. Some areas of South America, including parts of Brazil, present the highest incidence of the disease in the West. The main etiological factors that have been associated with the disease in Brazil are alcohol consumption, tobacco smoking and, in the South, consumption of hot maté. Nitrosamines are the only carcinogens capable of inducing tumors in the esophagus of experimental animals, with the rat being the most susceptible species, mainly due to tissue specific metabolic activation by CYP enzymes. Studies of CYP2A expression in the esophagus of rodents suggest an association between CYP2A expression and esophageal susceptibility to tumor induction. CYP2A6 and CYP2E1, the main enzymes to activate nitrosamines in humans, are the only carcinogen activating CYP enzymes to be expressed in the esophagus of Brazilians. Patients who presented high levels of CYP2A6 expression could activate nitrosamines at rates comparable to the rat. This expression profile is different from those present in French patients. We investigated 34 Brazilian patients regarding the risk associated with polymorphisms in drug metabolizing enzymes and TP53 mutations. A GSTP1 polymorphism presented a clear risk to white and non-white patients to develop esophageal cancer. GSTM1 null polymorphism also seemed to be associated with an increased risk. CYP2A6, CYP2E1, SOD2, and GSTT1 polymorphisms were not associated with an increased risk of esophageal cancer. TP53 mutations occurred mostly in exon 7, differing from the mutation profile found in the IARC database. The preliminary results obtained with polymorphisms of drug metabolizing enzymes and TP53 mutations need to be confirmed in a larger number of samples in order to compare the mechanisms of esophageal cancer development in Brazilians with that of other populations.

TP53 mutations in breast cancer tumors of patients from Rio de Janeiro, Brazil: association with risk factors and tumor characteristics.

Somatic mutations in the TP53 gene are the most frequently observed genetic alterations in human malignancies, including breast cancer, which is one of the leading causes of death among women in Brazil. In our study, we determined the frequency and the pattern of TP53 mutations in malignant breast tumors from 120 patients living in Rio de Janeiro, Brazil. TP53 mutations were found in 20% of the tumors, which contained a diversity of mutation types: missense (62.5%), nonsense (8.3%), silent (4.2%), intronic (12.5%), insertion (4.2%) and deletion (8.3%). Of a total of 15 missense mutations, 4 were observed at Arg248 and 2 at Cys242, which are directly involved in DNA binding and in zinc binding, respectively. A subgroup of 51 patients was analyzed with respect to the relation between the presence of TP53 mutations and classical risk factors and with tumor and patient characteristics. For this analysis, we used logistic regression and, in order to obtain more precise confidence intervals, they were recalculated using a bootstrap resampling technique. Our results demonstrate that these mutations are not statistically associated with the risk factors or the patients' characteristics. However, the presence of TP53 mutations is strongly associated with the aggressiveness of the tumors, measured by Elston classification (OR = 11.97 and 95% CI of 2.24-307.05). This finding is in agreement with previous studies, which report the presence of TP53 mutations in tumors with poor prognosis. This correlation between tumor aggressiveness and TP53 mutations could be a crucial variable for the treatment and prognosis of breast cancer.

CYP1A1, GSTM1, and GSTT1 polymorphisms and breast cancer risk in Brazilian women.

The frequency of CYP1A1 (CYP1A1*2A), GSTM1, and GSTT1 polymorphisms, as well as the main risk factors associated with breast cancer were studied in Brazilian women, with malignant breast cancer (n=128), or age-matched controls (n=256). Only a family history of breast cancer presented a significant risk (OR=3.00, CI=1.27-7.06). Among non-whites, the CYP1A1*2A allele was underrepresented among patients. Statistical analysis indicated that this polymorphism may decrease the risk of breast cancer among these individuals, particularly after adjusting for the risk presented by selected risk factors (OR=0.30, 95% CI=0.12-0.76).