Relação entre governança corporativa e gerenciamento de resultados: um estudo em empresas brasileiras listadas na BM&FBOVESPA

The objective of this work was to identify a possible relation between corporate governance, through the structure of the boards of directors and the levels of earnings management of Brazilian public companies. The study is characterized as a descriptive, of a hypothetical-deductive nature, with quantitative approach guided by the estimation model proposed by Kang and Sivaramakrishnan (1995). The sample was comprised by 56 companies, analyzed in the period from 2011 to 2014. The information on the companies were extracted from Economatica databank. For the data analysis, we used multivariate techniques, such as Pearson correlation and panel data in POLS, Fixed Effects and Random Effects approaches. Furthermore, we applied the mean comparison test ANOVA. The results obtained show that, generally, the organizations tend to follow the profile of boards of directors recommended by the codes of corporative governance. However, the characteristics of the composition of the board, regarding its size and the duality of positions that are not yet sufficient to be considered capable of inhibiting the discretionary practice of the studied companies. Relative the control variables, only size and return on equity presented no significant relation with result management. It is important to highlight that literature point many factors that explain the different impacts caused by the formation of the administration board in different regions or countries. Among the factors pointed, we highlight the legal system of the country, the economic and political development, the importance of the capital market, and the level of accounting education.

Análise da expressão dos genes da subfamília DREB em resposta à seca em plantas de Coffea canephora Conilon

Coffea canephora is one of the most economically important coffee species and in Brazil, Conilon is the most widely cultivated plant of this species. Abiotic stresses such as temperature variations and drought periods are factors that significantly affect their production and tend to worsen with globally recognized climate changes. In an attempt to understand the molecular responses of coffee plants in water deficit conditions, recent studies have identified candidate genes (CGs) as CcDREB1D. This gene showed increased expression in response to drought in the leaves of clone 14 (drought tolerant) in relation to the clone 22 (sensitive to drought) of C. canephora Conilon. Based on these results, the identification of DREB genes and their subgroups (SGs) of C. canephora, the objective is to analyze in silico and also in vivo these genes expression in leaf and root of tolerant (14, 73 and 120) and sensitive clones (22) of C. canephora Conilon submitted or not to a water deficit. In silico expressions of all DREB genes were analyzed from the Coffee Genome Hub Database and in vivo expression was performed by the technique "reverse transcription-quantitative PCR" (RT-qPCR). In silico gene expression analysis was possible to identify DREB genes with potential responses to abiotic stresses, corroborating some validated in vivo results. In this analysis, several genes showed differential expression in response to drought among the SGs (IIV), the tolerant and sensitive clones and the leaf and root. These differentially expressed genes were identified as potential CGs and among them, it was found that most tolerant clones showed increased expression in relation to sensitive in response to drought, with higher expression levels for clones 14 and 73. These highest levels were observed in leaves compared to the roots and SG-I stood at greater number of genes expressed in response to drought. These results suggest that DREB CGs, as Cc05_g06840, Cc02_g03420 e Cc08_g13960, play an important role in the regulatory mechanisms of response to drought in C. canephora Conilon.

Four year multitrophic food web: source food web description, methodology accuracy and species diversity

Food webs have been used in order to understand the trophic relationship among organisms within an ecosystem, however the extension by which sampling efficiency could affect food web responses remain poorly understood. Still, there is a lack of long-term sampling data for many insect groups, mainly related to the interactions between herbivores and their host plants. In the first chapter, I describe a source food web based on the Senegalia tenuifolia plant by identifying the associated insect species and the interactions among them and with this host plant. Furthermore, I check for the data robustness from each trophic level and propose a cost-efficiently methodology. The results from this chapter show that the collected dataset and the methodology presented are a good tool for sample most insect richness of a source food web. In total the food web comprises 27 species belonging to four trophic levels. In the second chapter, I demonstrate the temporal variation in the species richness and abundance from each trophic level, as well as the relationship among distinct trophic levels. Moreover, I investigate the diversity patterns of the second and third trophic level by assessing the contribution of alfa and beta-diversity components along the years. This chapter shows that in our system the parasitoid abundance is regulated by the herbivore abundances. Besides, the species richness and abundances of the trophic levels vary temporally. It also shows that alfa-diversity was the diversity component that most contribute to the herbivore species diversity (2nd trophic level), while the contribution of alfa- and beta-diversity changed along the years for parasitoid diversity (3rd level). Overall, this dissertation describes a source food web and bring insights into some food web challenges related to the sampling effort to gather enough species from all trophic levels. It also discuss the relation among communities associated with distinct trophic levels and their temporal variation and diversity patterns. Finally, this dissertation contributes for the world food web database and in understanding the interactions among its trophic levels and each trophic level pattern along time and space