Compositional Data in Biomedical Research

Modern methods of compositional data analysis are not well known in biomedical research. Moreover, there appear to be few mathematical and statistical researchers working on compositional biomedical problems. Like the earth and environmental sciences, biomedicine has many problems in which the relevant scienti c information is encoded in the relative abundance of key species or categories. I introduce three problems in cancer research in which analysis of compositions plays an important role. The problems involve 1) the classi cation of serum proteomic pro les for early detection of lung cancer, 2) inference of the relative amounts of di erent tissue types in a diagnostic tumor biopsy, and 3) the subcellular localization of the BRCA1 protein, and it's role in breast cancer patient prognosis. For each of these problems I outline a partial solution. However, none of these problems is \solved". I attempt to identify areas in which additional statistical development is needed with the hope of encouraging more compositional data analysts to become involved in biomedical research

Underarm cosmetics and breast cancer

Although risk factors are known to include the loss of function of the susceptibility genes BRCA1/BRCA2 and lifetime exposure to oestrogen, the main causative agents in breast cancer remain unaccounted for. It has been suggested recently that underarm cosmetics might be a cause of breast cancer, because these cosmetics contain a variety of chemicals that are applied frequently to an area directly adjacent to the breast. The strongest supporting evidence comes from unexplained clinical observations showing a disproportionately high incidence of breast cancer in the upper outer quadrant of the breast, just the local area to which these cosmetics are applied. A biological basis for breast carcinogenesis could result from the ability of the various constituent chemicals to bind to DNA and to promote growth of the damaged cells. Multidisciplinary research is now needed to study the effect of long-term use of the constituent chemicals of underarm cosmetics, because if there proves to be any link between these cosmetics and breast cancer then there might be options for the prevention of breast cancer. Copyright D 2003 John Wiley Sons, Ltd.

Environmental oestrogens and breast cancer: long-term low-dose effects of mixtures of various chemical combinations

The incidence of breast cancer has risen worldwide to unprecedented levels in recent decades, making it now the major cancer of women in many parts of the world.1 Although diet, alcohol, radiation and inherited loss of BRCA1/2 genes have all been associated with increased incidence, the main identified risk factors are life exposure to hormones including physiological variations associated with puberty/pregnancy/menopause,1 personal choice of use of hormonal contraceptives2 and/or hormone replacement therapy.3–6 On this basis, exposure of the human breast to the many environmental pollutant chemicals capable of mimicking or interfering with oestrogen action7 should also be of concern.8 Hundreds of such environmental chemicals have now been measured in human breast tissue from a range of dietary and domestic exposure sources7 ,9 including persistent organochlorine pollutants (POPs),10 polybrominated diphenylethers and polybromobiphenyls,11 polychlorinated biphenyls,12 dioxins,13 alkyl phenols,14 bisphenol-A and chlorinated derivatives,15 as well as other less lipophilic compounds such as parabens (alkyl esters of p-hydroxybenzoic acid),16 but studies investigating any association between raised levels of such compounds and the development of breast cancer remain inconclusive.7–16 However, the functionality of these chemicals has continued to be assessed on the basis of individual chemicals rather than the environmental reality of long-term low-dose exposure to complex mixtures. This misses the potential for individuals to have high concentrations of different compounds but with a common mechanism of action. It also misses the complex interactions between chemicals and physiological hormones which together may act to alter the internal homeostasis of the oestrogenic environment of mammary tissue.

Effects of aluminium chloride and aluminium chlorohydrate on DNA repair in MCF10A immortalised non-transformed human breast epithelial cells

Use of underarm aluminium (Al)-based antiperspirant salts may be a contributory factor in breast cancer development. At the 10th Keele meeting, Al was reported to cause anchorage-independent growth and double strand DNA breaks in MCF10A immortalised non-transformed human breast epithelial cells. We now report that exposure of MCF10A cells to Al chloride or Al chlorohydrate also compromised DNA repair systems. Longterm (19–21 weeks) exposure to Al chloride or Al chlorohydrate at a 10−4 M concentration resulted in reduced levels of BRCA1 mRNA as determined by real-time RT-PCR and BRCA1 protein as determined by Western immunoblotting. Reduced levels of mRNA for other DNA repair genes (BRCA2, CHK1, CHK2, Rad51, ATR) were also observed using real-time RT-PCR. Loss of BRCA1 or BRCA2 gene function has long been associated with inherited susceptibility to breast cancer but these results suggest that exposure to aluminium-based antiperspirant salts may also reduce levels of these key components of DNA repair in breast epithelial cells. If Al can not only damage DNA but also compromise DNA repair systems, then there is the potential for Al to impact on breast carcinogenesis.

Toxicogenomic activity of gemcitabine in two TP53-mutated bladder cancer cell lines: special focus on cell cycle-related genes

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

An Integrative Genomic and Transcriptomic Analysis Reveals Potential Targets Associated with Cell Proliferation in Uterine Leiomyomas

Background: Uterine Leiomyomas (ULs) are the most common benign tumours affecting women of reproductive age. ULs represent a major problem in public health, as they are the main indication for hysterectomy. Approximately 40-50% of ULs have non-random cytogenetic abnormalities, and half of ULs may have copy number alterations (CNAs). Gene expression microarrays studies have demonstrated that cell proliferation genes act in response to growth factors and steroids. However, only a few genes mapping to CNAs regions were found to be associated with ULs. Methodology: We applied an integrative analysis using genomic and transcriptomic data to identify the pathways and molecular markers associated with ULs. Fifty-one fresh frozen specimens were evaluated by array CGH (JISTIC) and gene expression microarrays (SAM). The CONEXIC algorithm was applied to integrate the data. Principal Findings: The integrated analysis identified the top 30 significant genes (P<0.01), which comprised genes associated with cancer, whereas the protein-protein interaction analysis indicated a strong association between FANCA and BRCA1. Functional in silico analysis revealed target molecules for drugs involved in cell proliferation, including FGFR1 and IGFBP5. Transcriptional and protein analyses showed that FGFR1 (P = 0.006 and P<0.01, respectively) and IGFBP5 (P = 0.0002 and P = 0.006, respectively) were up-regulated in the tumours when compared with the adjacent normal myometrium. Conclusions: The integrative genomic and transcriptomic approach indicated that FGFR1 and IGFBP5 amplification, as well as the consequent up-regulation of the protein products, plays an important role in the aetiology of ULs and thus provides data for potential drug therapies development to target genes associated with cellular proliferation in ULs. © 2013 Cirilo et al.

Alterações genéticas e epigenéticas em meningiomas na população paraense

Os meningiomas são os tumores intracranianos mais frequentes, originando-se das meninges que revestem o cérebro e cordão espinhal. Apesar de terem sido um dos primeiros neoplasmas sólidos estudados citogeneticamente, ainda são escassos os estudos genéticos e epigenéticos nesses tumores. O presente trabalho teve como objetivo investigar alterações genéticas e epigenéticas que pudessem contribuir na iniciação e progressão tumoral em meningiomas na população paraense. Essa tese está subdivida em três capítulos. No Capítulo I foi investigada a associação entre o polimorfismo MTHFR C677T e meningioma em 23 pacientes da população paraense, utilizando 96 indivíduos sem histórico de lesões pré-neoplásicas como grupo controle. Essa associação não foi encontrada, apesar de sugerir um aumento não estatisticamente significante no risco de desenvolver meningioma em portadores do genótipo TT quando comparados ao genótipo CC. No Capítulo II foi avaliado o padrão de metilação em duas famílias do microRNA124 em meningiomas na população paraense. Hipermetilação na região promotora de miRN124a2 e miRNA124a3 parece ser um evento frequente, uma vez que foi encontrada em 73,9% e 69,56% das amostras analisadas, respectivamente. No Capítulo III, foi analisado o padrão de metilação dos genes APC, BRCA1, CDH1, CDH13, CDKN2A, DAPK1, ESR1, FHIT, GSTP1, MGMT, MLH1, NEUROG1, PDLIM4, PTEN, RARB, RASSF1, RUNX3, SOCS1, TIMP3, TP73, VHL e WIF1 em um meningioma de grau I e um de grau II através de uma placa comercial desenvolvida através da tecnologia MethylScreen. O padrão de metilação do gene CDKN2B também foi analisado na amostra coletada em 25 pacientes com meningioma através da conversão por bissulfito, PCR e sequenciamento direto. O gene RASSF1A apresentou-se metilado em 16,73% e 63,66% dos sítios CpGs analisados na amostra de meningioma de grau I e grau II, respectivamente. O gene RUNX3 se apresentou metilado apenas na amostra de grau II em 52,88% dos sítios CpG analisados. Nossos resultados apontam a importância das alterações epigenéticas na tumorigênese e progressão tumoral em meningiomas.

Genomic alterations in patients showing multiple primary tumors and family history of cancer

Multiple primary tumors (MPT) are a major cause of mortality and morbidity among patients that have survived after the treatment of a first cancer. It has been proposed that after the first primary tumor, high risk of a subsequent tumor could be associated with radiotherapy used as treatment for the first cancer. Other potential risk factors include unhealthy lifestyle, genetic predisposition, aging, environmental determinants or an interaction between these factors. However, an association between the presence of MPT and family history of cancer in cases without clinical and molecular evidence of a known hereditary cancer syndrome is rarely described. Genomic DNA from 12 patients with at least two primary tumors and without mutations on TP53 was evaluated by CytoScan HD Array (Affymetrix). Chromosome Analysis Suite (ChAS) software v.2.0.1 was used considering at least 50 markers for gains; 25 for losses and a minimum of 5Mb for cnLOHs. Data from 1038 phenotypically healthy individuals (Affymetrix) and from Database of Genomic Variants were used as reference. Only alterations found in <1% (rare) or never described (new rare) in the reference population were considered. All cases, except one, presented a family history of cancer. Five cases developed MTP after radiotherapy and only one was located in the same treated area. It was detected 67 rare and 15 new rare genomic alterations encompassing 5.906 genes: 17 losses, 29 gains, and 36 cnLOH. X chromosome presented the higher number of alterations. Two patients with breast cancer presented a large deletion/cnLOH on 7q21. Enrichment analysis revealed 1275 genes associated with breast cancer (p= 0.001), which was diagnosed in 6 patients and their family members (all negative for BRCA1/2 or TP53 mutations). cnLOHs accounted for 44% of all the alterations. A significant proportion of cases (11/12) presented family history of cancer and the patients were not submitted to radiotherapy (7/12). We demonstrated the presence of rare genomic alterations in patients with MPT suggesting their involvement in the MPT development. cnLOH may arise as a new mechanism associated with the risk to develop MPT. All authors have declared no conflicts of interest.

ROBO1 deletion as a novel germline alteration in breast and colorectal cancer patients

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

Germline DNA copy number variation in familial and early-onset breast cancer

Introduction: Genetic factors predisposing individuals to cancer remain elusive in the majority of patients with a familial or clinical history suggestive of hereditary breast cancer. Germline DNA copy number variation (CNV) has recently been implicated in predisposition to cancers such as neuroblastomas as well as prostate and colorectal cancer. We evaluated the role of germline CNVs in breast cancer susceptibility, in particular those with low population frequencies (rare CNVs), which are more likely to cause disease." Methods: Using whole-genome comparative genomic hybridization on microarrays, we screened a cohort of women fulfilling criteria for hereditary breast cancer who did not carry BRCA1/BRCA2 mutations. Results: The median numbers of total and rare CNVs per genome were not different between controls and patients. A total of 26 rare germline CNVs were identified in 68 cancer patients, however, a proportion that was significantly different (P = 0.0311) from the control group (23 rare CNVs in 100 individuals). Several of the genes affected by CNV in patients and controls had already been implicated in cancer. Conclusions: This study is the first to explore the contribution of germline CNVs to BRCA1/2-negative familial and early-onset breast cancer. The data suggest that rare CNVs may contribute to cancer predisposition in this small cohort of patients, and this trend needs to be confirmed in larger population samples.

Relationship between B-Cell-specific moloney murine leukemia virus integration site 1 (BMI-1) and homologous recombination regulatory genes in invasive ductal breast carcinomas

B-cell-specific Moloney murine leukemia virus integration site 1 (Bmi-1) is a Polycomb group protein that is able to induce telomerase activity, enabling the immortalization of epithelial cells. Immortalized cells are more susceptible to double-strand breaks (DSB), which are subsequently repaired by homologous recombination (HR). BRCA1 is among the HR regulatory genes involved in the response to DNA damage associated with the RAD51 protein, which accumulates in DNA damage foci after signaling H2AX, another important marker of DNA damage. Topoisomerase III beta (topoIII beta) removes HR intermediates before chromosomal segregation, preventing damage to cellular DNA structure. In breast carcinomas positive for BMI-1 the role of proteins involved in HR remains to be investigated. The aim of this study was to evaluate the association between BMI-1 and homologous recombination proteins. Using tissue microarrays containing 239 cases of primary breast tumors, the expression of Bmi-1, BRCA-1, H2AX, Rad51, p53, Ki-67, topoIII beta, estrogen receptors (ER), progesterone receptors (PR), and HER-2 was analyzed by immunohistochemistry. We observed high Bmi-1 expression in 66 cases (27.6%). Immunohistochemical overexpression of BMI-1 was related to ER (p=0.004), PR (p<0.001), Ki-67 (p<0.001), p53 (p=0.003), BRCA1 (p=0.003), H2AX (p=0.024) and topoIII beta (p<0,001). Our results show a relationship between the expression of BMI-1 and HR regulatory genes, suggesting that Bmi-1 overexpression might be an important event in HR regulation. However, further studies are necessary to understand the mechanisms in which Bmi-1 could regulate HR pathways in invasive ductal breast carcinomas.

SET overexpression decreases cell detoxification efficiency: ALDH2 and GSTP1 are downregulated, DDR is impaired and DNA damage accumulates

Alcohol and tobacco consumption are risk factors for head and neck squamous cell carcinoma (HNSCC). Aldehyde dehydrogenase 2 (ALDH2) and glutathione Stransferase pi 1 (GSTP1) are important enzymes for cellular detoxification and low efficiencies are implicated in cancer. We assessed the potential role of SET protein overexpression, a histone acetylation modulator accumulated in HNSCC, in gene regulation and protein activity of ALDH2 and GSTP1. SET was knocked down in HN13, HN12 and Cal27, and overexpressed in HEK293 cells; ethanol and cisplatin were the chemical agents. Cells with SET overexpression (HEK293/SET, HN13 and HN12) showed lower ALDH2 and GSTP1 mRNA levels and trichostatin A increased them (real-time PCR). Ethanol upregulated GSTP1 and ALDH2 mRNAs, whereas cisplatin upregulated GSTP1 in HEK293 cells. SET-chromatin binding revealed SET interaction with ALDH2 and GSTP1 promoters, specifically via SET NAP domain; ethanol and cisplatin abolished SET binding. ALDH2 and GSTP1 efficiency was assessed by enzymatic and comet assay. A lower ALDH2 activity was associated with greater DNA damage (tail intensity) in HEK293/SET compared with HEK293 cells, whereas HN13/siSET showed ALDH2 activity higher than HN13 cells. HN13/siSET cells showed increased tail intensity. Cisplatin-induced DNA damage response showed negative relationship between SET overexpression and BRCA2 recruitment. SET downregulated repair genes ATM, BRCA1 and CHEK2 and upregulated TP53. Cisplatin-induced cell-cycle arrest occurred in G0/G1 and S in HEK293 cells, whereas HEK293/SET showed G2/M stalling. Overall, cisplatin was more cytotoxic for HN13 than HN13/siSET cells. Our data suggest a role for SET in cellular detoxification, DNA damage response and genome integrity.

p300/CBP tyrosine phosphorylation in response to dna damage activated by c-Abl tyrosine kinase

The nuclear signaling that is triggered in response to DNA damage entails the recruitment and assembly of repair proteins and the induction of genes involved in the activation of cell cycle checkpoint, apoptosis or senescence. The extensive changes in chromatin structure underlying these processes suggest that chromatin-modifying enzymes could be relevant targets of DNA damage-activated signaling. The acetyltransferases p300 and CBP participate in DNA damage-activated responses, including local histone hyperacetylation, cell cycle regulation, and co-activation of DNA damage activated proteins, such as p53, p73 and BRCA1. However, the link between DNA damage and p300/CBP activation has not been identified.We have detected p300 tyrosine phosphorylation in response to DNA damage. We show that the DNA damage-activated cAbl tyrosine kinase enters the nuclei of cells exposed to genotoxic agents and phosphorylates p300 on a tyrosine residue within the bromodomain that is conserved in p300, CBP and many other bromodomain-containing proteins. Antibodies against tyrosine phosphorylated p300/CBP show a DNA damage-inducible nuclear staining, suggesting that p300 tyrosine phosphorylation is an event linking DNA damage and chromatin modifications.

DNA-Reparatur-assoziierte genetische und epigenetische Faktoren für die Zweittumorentstehung nach Tumoren im Kindes- und Jugendalter

Die Ursachen der Zweittumorentwicklung bei Personen, die eine Krebserkrankung in der Kindheit überlebten, sind weitgehend unklar. Strahlenexposition oder Chemotherapie führen in normalen somatischen Zellen zu DNA-Schäden, welche bei fehlerhafter Reparatur eine Karzinogenese auslösen können. Es ist denkbar, dass genetische Unterschiede z. B. in den Signalwegen der Zellzykluskontrolle und der DNA-Reparatur nach therapieinduzierten DNA-Schäden eine entscheidende Rolle bei der Zweittumorentwicklung spielen. Im Rahmen dieser Arbeit wurden 20 Personen, die eine Krebserkrankung in der Kindheit überlebten und einen unabhängigen Zweittumor entwickelten, mit 20 gematchten Kontrollpersonen ohne Zweittumorentwicklung verglichen. Die primären Fibroblasten der Patienten wurden auf somatische, genetische und/oder epigenetische Unterschiede in DNA-Reparaturnetzwerken untersucht. Die biologisch relevantesten Ergebnisse lieferten Proteinuntersuchungen mittels Antikörper-Microarrays. Hierbei wurde eine konstitutiv erniedrigte Menge an RAD9A und einigen anderen DNA-Reparatur-Proteinen (BRCA1, DDIT3, MSH6, p53, RAD51) in den Zweittumorpatienten im Vergleich zu den Eintumorpatienten festgestellt. Nach einer DNA-Schädigung durch 1 Gray Bestrahlung erhöhte sich die RAD9A-Proteinmenge, wobei die Zweittumorpatienten eine geringere Induktion als die Eintumorpatienten zeigten. Bei der Quantifizierung der mRNA-Expression mittels RTq-PCR wurde ein niedrigerer RAD9A-mRNA-Level sowohl in den unbehandelten und als auch in den 1 Gray bestrahlten Zellen der Zweittumorpatienten festgestellt. SNP-Array und Methylierungsanalysen konnten keine Auffälligkeiten im RAD9A-Lokus nachweisen. Diese Ergebnisse unterstützen die Hypothese, dass Modulationen von RAD9A und anderen Zellzyklusarrest- und DNA-Reparaturproteinen zum Risiko einer Zweittumorentwicklung in Kinderkrebspatienten beitragen. Bei einem diskordanten monozygoten Zwillingspaar wurde in ca. 20% der Zellen des Zweittumorzwillings eine Hypermethylierung des Tumorsuppressorgens BRCA1 festgestellt, die mit einer konstitutiv erniedrigten BRCA1-Proteinexpression einhergeht und einen möglichen Krebsrisikofaktor darstellt. Die partielle Deletion des Gens RSPO3, die wahrscheinlich als somatisches Zellmosaik beim Zweittumorzwilling vorliegt, korreliert mit einer niedrigeren RSPO3-mRNA-Expression und ist vermutlich auch mit einer erhöhten Krebsprädisposition assoziiert.

Life stage differences in mammary gland gene expression profile in non-human primates

Breast cancer (BC) is the most common malignancy of women in the developed world. To better understand its pathogenesis, knowledge of normal breast development is crucial, as BC is the result of disregulation of physiologic processes. The aim of this study was to investigate the impact of reproductive life stages on the transcriptional profile of the mammary gland in a primate model. Comparative transcriptomic analyses were carried out using breast tissues from 28 female cynomolgus macaques (Macaca fascicularis) at the following life stages: prepubertal (n = 5), adolescent (n = 4), adult luteal (n = 5), pregnant (n = 6), lactating (n = 3), and postmenopausal (n = 5). Mammary gland RNA was hybridized to Affymetrix GeneChip(®) Rhesus Macaque Genome Arrays. Differential gene expression was analyzed using ANOVA and cluster analysis. Hierarchical cluster analysis revealed distinct separation of life stage groups. More than 2,225 differentially expressed mRNAs were identified. Gene families or pathways that changed across life stages included those related to estrogen and androgen (ESR1, PGR, TFF1, GREB1, AR, 17HSDB2, 17HSDB7, STS, HSD11B1, AKR1C4), prolactin (PRLR, ELF5, STAT5, CSN1S1), insulin-like growth factor signaling (IGF1, IGFBP1, IGFBP5), extracellular matrix (POSTN, TGFB1, COL5A2, COL12A1, FOXC1, LAMC1, PDGFRA, TGFB2), and differentiation (CD24, CD29, CD44, CD61, ALDH1, BRCA1, FOXA1, POSTN, DICER1, LIG4, KLF4, NOTCH2, RIF1, BMPR1A, TGFB2). Pregnancy and lactation displayed distinct patterns of gene expression. ESR1 and IGF1 were significantly higher in the adolescent compared to the adult animals, whereas differentiation pathways were overrepresented in adult animals and pregnancy-associated life stages. Few individual genes were distinctly different in postmenopausal animals. Our data demonstrate characteristic patterns of gene expression during breast development. Several of the pathways activated during pubertal development have been implicated in cancer development and metastasis, supporting the idea that other developmental markers may have application as biomarkers for BC.