Neste trabalho, calibra-se um modelo que descreva a relação fluxo-velocidade em autoestradas e rodovias de pista dupla, considerando as recomendações tecidas pelos autores envolvidos na elaboração do modelo do Highway Capacity Manual - HCM 2010. Para tanto, foi utilizada uma amostra de 788.122 observações, coletadas por 25 estações em quatro rodovias paulistas: SP-348, SP-021, SP-280 e SP 270. A análise dos dados mostrou que, conforme preconizado pelo HCM 2010, existe um intervalo de fluxos no qual a velocidade se mantém constante, o que corresponde à velocidade de fluxo livre. Em seguida, verificou-se que a classificação do HCM 2010, entre autoestradas e rodovias de pista dupla, não se mostra adequada para a amostra estudada, tendo sido indicada a divisão entre rodovias urbanas e rurais. Para esses grupos, foram propostos valores representativos para a capacidade e calibrada a relação fluxo-velocidade. A comparação entre os dois conjuntos de curvas produzidos mostrou que a velocidade média da corrente de tráfego nas rodovias urbanas apresenta uma queda mais precoce e acentuada do que a observada em rodovias rurais. Além disso, os valores estimados para a capacidade C e para a velocidade na capacidade CS para rodovias urbanas são inferiores aos estimados para rodovias rurais. Em contraste com o modelo do HCM 2010, a principal diferença reside em valores significativamente inferiores para BP, o fluxo a partir do qual se nota um declínio da velocidade em função do aumento da densidade da corrente de tráfego. Outro ponto de destaque é que, embora os valores da capacidade sejam parecidos entre o modelo do HCM 2010 e o calibrado, a velocidade na capacidade CS é superior nas rodovias paulistas. / This work presents the calibration of a model describing the speed-flow relationship on freeways and multilane highways in the state of São Paulo, Brazil. The calibrated model follows the recommendations made by the authors involved in the development of the model used by the Highway Capacity Manual HCM 2010. The calibration used a sample of 788,122 observations, collected by 25 stations on four highways at São Paulo state: SP-348, SP-021, SP-280 and 270 SP. The analysis of the data showed that, as advocated by the HCM 2010, there is range of flows in which the average speed of the passenger cars remains constant and equal to the free flow speed. It was also found that the classification scheme used by HCM 2010, comprising freeways and multilane highways, is not adequate for highways in the state of São Paulo. A new classification scheme, which divides highways into urban or rural sections, is proposed. For these classes, representative values for the capacity were found, and the speed-flow relation was calibrated. The comparison between the two sets of curves produced showed that the average speed of the traffic stream on urban highways has an earlier and steeper drop than the observed on rural highways. In addition, the estimated values for the capacity C and for the speed at capacity CS for urban highways are lower compared to the estimated for rural highways. Compared to the model used by the HCM 2010, the main difference lies in the significantly lower values for BP, the traffic flow from which the average speed declines as the density of the traffic stream increases. Another important indication is that, although the capacity values are similar between the model used by the HCM 2010 and the calibrated model, the speed at capacity CS is higher at São Paulo state highways.

Neste trabalho, calibra-se um modelo que descreva a relação fluxo-velocidade em autoestradas e rodovias de pista dupla, considerando as recomendações tecidas pelos autores envolvidos na elaboração do modelo do Highway Capacity Manual - HCM 2010. Para tanto, foi utilizada uma amostra de 788.122 observações, coletadas por 25 estações em quatro rodovias paulistas: SP-348, SP-021, SP-280 e SP 270. A análise dos dados mostrou que, conforme preconizado pelo HCM 2010, existe um intervalo de fluxos no qual a velocidade se mantém constante, o que corresponde à velocidade de fluxo livre. Em seguida, verificou-se que a classificação do HCM 2010, entre autoestradas e rodovias de pista dupla, não se mostra adequada para a amostra estudada, tendo sido indicada a divisão entre rodovias urbanas e rurais. Para esses grupos, foram propostos valores representativos para a capacidade e calibrada a relação fluxo-velocidade. A comparação entre os dois conjuntos de curvas produzidos mostrou que a velocidade média da corrente de tráfego nas rodovias urbanas apresenta uma queda mais precoce e acentuada do que a observada em rodovias rurais. Além disso, os valores estimados para a capacidade C e para a velocidade na capacidade CS para rodovias urbanas são inferiores aos estimados para rodovias rurais. Em contraste com o modelo do HCM 2010, a principal diferença reside em valores significativamente inferiores para BP, o fluxo a partir do qual se nota um declínio da velocidade em função do aumento da densidade da corrente de tráfego. Outro ponto de destaque é que, embora os valores da capacidade sejam parecidos entre o modelo do HCM 2010 e o calibrado, a velocidade na capacidade CS é superior nas rodovias paulistas. / This work presents the calibration of a model describing the speed-flow relationship on freeways and multilane highways in the state of São Paulo, Brazil. The calibrated model follows the recommendations made by the authors involved in the development of the model used by the Highway Capacity Manual HCM 2010. The calibration used a sample of 788,122 observations, collected by 25 stations on four highways at São Paulo state: SP-348, SP-021, SP-280 and 270 SP. The analysis of the data showed that, as advocated by the HCM 2010, there is range of flows in which the average speed of the passenger cars remains constant and equal to the free flow speed. It was also found that the classification scheme used by HCM 2010, comprising freeways and multilane highways, is not adequate for highways in the state of São Paulo. A new classification scheme, which divides highways into urban or rural sections, is proposed. For these classes, representative values for the capacity were found, and the speed-flow relation was calibrated. The comparison between the two sets of curves produced showed that the average speed of the traffic stream on urban highways has an earlier and steeper drop than the observed on rural highways. In addition, the estimated values for the capacity C and for the speed at capacity CS for urban highways are lower compared to the estimated for rural highways. Compared to the model used by the HCM 2010, the main difference lies in the significantly lower values for BP, the traffic flow from which the average speed declines as the density of the traffic stream increases. Another important indication is that, although the capacity values are similar between the model used by the HCM 2010 and the calibrated model, the speed at capacity CS is higher at São Paulo state highways.

Myosin II, das Motorprotein des kardialen Muskels, besteht aus zwei schweren und vier leichten Ketten. Der Hebelarmbereich der schweren Myosinkette (MyHC) enthält das IQ-Konsensus-Motiv für die Bindung der essentiellen leichten Myosinkette (ELC), welche wesentlich für eine normale Kraftentwicklung des Myosinmoleküls ist. Im Rahmen dieser Arbeit wurden fünf, mit hypertropher Kardiomyopathie assoziierte, Mutationen im humanen essentiellen ventrikulären leichten Myosinketten (hVLC1)-Gen (MYL3) untersucht (E56G, A57G, E143K, M149V, R154H). Von keiner dieser Mutationen war der Pathomechanismus bekannt. Ziel der Arbeit war es, die Effekte der Mutationen im MYL3-Gen auf Proteinstruktur und Funktion zu untersuchen und daraufhin einen möglichen Pathomechanismus zu formulieren. Dazu erfolgten Strukturanalysen (CD-Spektren, Schmelzkurven, FLIM), Versuche auf Protein- und Zellebene (Protein-Protein-Interaktionsstudien, Sorting Assay) sowie Untersuchungen in vitro (Zell-Verkürzungsmessungen, isoliert perfundierte Herzen nach Langendorff) und in vivo (Echokardiographie) im transgenen Mausmodell. / Myosin II, the motor protein of cardiac muscle, is composed of two heavy chains (MyHC) and four non-covalently linked light chains (MLC). The lever arm of the MyHC contains the IQ motif that binds the essential myosin light chain (ELC), which is necessary for the normal force production of the myosin molecule. Five with HCM associated mutations in the human ventricular essential myosin light chain (hVLC1) -gen (MYL3) were investigated in this study (E56G, A57G, E143K, M149V, R154H). The pathomechanisms of the mutations were not known. Aim of the study was i) to test the hypothesis that mutations in the ventricular essential myosin light chain affect the protein structure, the binding to the IQ motif of MyHC and the force production of the myosin molecule as well as ii) to postulate an accompanying pathomechanism. Structural analyses (circular dichroism, melting curves, fluorescence lifetime imaging microscopy), functional investigations (surface plasmon resonance spectroscopy, sorting assay) and in vivo (echocardiography) and in vitro studies in a transgenic mouse model were performed.

O objetivo desta tese é adaptar os procedimentos do Highway Capacity Manual 2000, para que possam melhor estimar a qualidade de serviço nas condições de tráfego encontradas em rodovias de pista simples estudadas. Para isto, foi desenvolvido um programa para calibração automática do simulador TRARR, baseado em um algoritmo genético e implementado em Perl. A calibração, que usa dados de velocidade e porcentagem de veículos em pelotões, coletado por meio de filmagens nos locais estudados, foi feita simultaneamente para cinco trechos diferentes, para que o simulador fosse capaz de reproduzir as características de um trecho típico de rodovia de pista simples do estado de São Paulo. Para garantir a qualidade dos resultados das simulações realizadas com o modelo recalibrado, procedeu-se à sua validação, usando-se um conjunto independente da dados. Partindo-se do pressuposto que a estrutura geral do método deveria ser mantida, a adaptação do HCM-2000 consistiu em obter novos valores para os diversos fatores de ajuste usados no processo de estimativa do nível de serviço. Para tanto, foram reproduzidos os experimentos de simulação realizados no desenvolvimento do procedimento para análise de rodovias de pista simples, à exceção do uso da versão recalibrada TRARR no lugar do simulador TWOPAS. Os fatores de ajuste adaptados foram: fator de pico horário (PHF), fator de ajuste de rampas (fG), fator de equivalência veicular (ET), fator para o efeito de zonas de ultrapassagem proibida (fnp) e fator de ajuste para o efeito combinado da divisão direcional do tráfego e de zonas de ultrapassagem proibida (fd/np). Com exceção do PHF, todos os demais fatores foram obtidos a partir de resultados de simulação. Os níveis de serviço observados nos cinco trechos estudados foram comparados com estimativas obtidas com a versão original do HCM-2000 e com a adaptação proposta. Os resultados desta comparação mostram que as estimativas obtidas com a adaptação proposta são mais precisas que as obtidas usando-se os valores originais do HCM-2000. / This thesis objective is to adapt the Highway Manual Capacity 2000, procedures to produce better estimates of service level and capacity for two-lane highways studied, In order to do this, an automatic calibration system, implementing a genetic algorithm and coded in Perl, was used to obtain a recalibrated version of TRARR, a two-lane highway simulation model. The calibration, that which uses speed and percent vehicles traveling in platoons data collected by videotaping, was carried out simultaneously for five different road segments, so that the recalibrated model would be able to reproduce traffic behavior on a typical two-lane road in the state of São Paulo, Brazil. The recalibrated model was validated using a second, independent, set of data. Based on the assumption that the general framework of the methodology should be kept, the adaptation was achieved by obtaining new values for the adjustment factors used in the procedure. Thus, the simulation experiments used in the development of HCM-2000 methodology were reproduced using the recalibrated TRARR model. The following adjustment factors were adapted: peak hour factor (PHF), grade adjustment factor (fG), heavy vehicle equivalence factor (ET), adjustment factor for the effect of non-passing zones (fnp) and adjustment factor for combined effect of directional distribution of traffic and percentage of no-passing zones (fd/np). With the exception of the PHF, all factors were obtained from simulation results. In order to assess the accuracy of the proposed adaptation observed levels of service were compared to the estimates obtained through the original HCM-2000 procedure and the modified procedure. The results of these comparisons clearly show that the estimates obtained using the proposed values for the adjustment factors are closer to the observed levels of service than the estimates obtained using the original values for the adjustment factors.

Esta pesquisa apresenta um método de calibração da relação fluxo-velocidade do Highway Capacity Manual, HCM, para autoestradas e rodovias de pista dupla, a partir de recomendações tecidas pelos autores envolvidos na elaboração do manual. Para tanto, foi utilizado uma amostra com mais de 1.700.000 observações, coletadas por 34 sensores de tráfego, localizados em quatros rodovias do Estado de São Paulo: SP-280, SP-348, SP-270 e SP-021. O tratamento do banco de dados foi realizado por meio da filtragem dos dados originais através da aplicação de três critérios, que têm como finalidade remover observações consideradas inadequadas para a calibração do modelo. Os critérios baseiam-se em obter uma corrente de tráfego formada somente por veículos leves, com condições de operação normal e com observações referentes apenas ao regime de fluxo livre. A separação entre os regimes de fluxo livre e congestionado foi realizada por meio da densidade na capacidade. Para tanto, foi proposto um método de estimação da capacidade por meio da análise da variação da velocidade média da corrente em função do fluxo de tráfego. O pressuposto do método parte do conceito de que o colapso da corrente de tráfego é um fenômeno estocástico, caracterizado pela queda abrupta da velocidade média em função do aumento do fluxo de tráfego. O momento em que ocorre mudança abrupta de velocidade é associado ao colapso da corrente de tráfego e consequentemente à capacidade da via. Após a estimação da capacidade, a velocidade na capacidade foi determinada como a média das velocidades associadas ao fluxo na capacidade, enquanto que a densidade na capacidade foi obtida por meio da relação fundamental de tráfego. O método foi aplicado em um conjunto de 18 trechos de rodovias que atingem a capacidade e os resultados foram considerados satisfatórios após a comparação com valores encontrados na literatura. A calibração da relação fluxo-velocidade foi realizada individualmente para cada trecho de rodovia utilizado no estudo. A análise dos resultados mostra que as rodovias rurais apresentam em média valores maiores para velocidade de fluxo livre, capacidade, coeficiente de calibração, velocidade na capacidade e ponto de transição em relação as rodovias urbanas. Alem disso, o ponto de transição, que consiste em um patamar de velocidade de fluxo livre constante, apresenta valores significativamente menores em relação a literatura. / The following research aims to outline a calibration method of the speed-flow relationship presented in the Highway Capacity Manual (HCM). The method is applied on freeways and multilane highways, leveraging the guidelines made by the authors that developed the manual. In order to support the calibration, a sample with more than 1.700.000 observations was used, collected by 34 traffic sensors on four highways at São Paulo state (SP-280, SP-348, SP270, and SP-021). Furthermore, the data treatment and cleaning process aimed to remove observations considered inappropriate from a model calibration perspective by filtering the original data through three main criteria: traffic flow only composed by passenger car, traffic under normal operations condition, and only containing observations which free flow regime applies. The third filtering criteria was applied by splitting free flow and congested regime through the threshold of density at capacity. To this extent, a capacity estimation method was proposed, aiming to analyse the average speed variation in function of the traffic flow. The assumption of the method relies in the concept that the traffic flow breakdown is a stochastic process, characterized by the abrupt drop in the average speed as function of the traffic flow increase. The moment that abrupt drop occurs is related to traffic breakdown and, consequently, to the capacity freeway capacity. Once the capacity was estimated, the speed at capacity was defined as the average of the speed associated with the flow at capacity while the density at capacity was estimated through the fundamental relationship of the traffic flow. The method was applied in 18 delimited highway traffic perimeter that reached their corresponding capacity and the results were considered satisfactory after comparing against values found in the literature. Furthermore, the speed-flow relationships calibration was performed for each of them individually. The results show that rural highways have in average higher free flow speed, capacity, calibration coefficient, speed at capacity and breakpoint than urban highways. Moreover, the breakpoint, which consists a constant free flow speed level, presented significant lower values in comparison to the literature.

Esta pesquisa apresenta um método de calibração da relação fluxo-velocidade do Highway Capacity Manual, HCM, para autoestradas e rodovias de pista dupla, a partir de recomendações tecidas pelos autores envolvidos na elaboração do manual. Para tanto, foi utilizado uma amostra com mais de 1.700.000 observações, coletadas por 34 sensores de tráfego, localizados em quatros rodovias do Estado de São Paulo: SP-280, SP-348, SP-270 e SP-021. O tratamento do banco de dados foi realizado por meio da filtragem dos dados originais através da aplicação de três critérios, que têm como finalidade remover observações consideradas inadequadas para a calibração do modelo. Os critérios baseiam-se em obter uma corrente de tráfego formada somente por veículos leves, com condições de operação normal e com observações referentes apenas ao regime de fluxo livre. A separação entre os regimes de fluxo livre e congestionado foi realizada por meio da densidade na capacidade. Para tanto, foi proposto um método de estimação da capacidade por meio da análise da variação da velocidade média da corrente em função do fluxo de tráfego. O pressuposto do método parte do conceito de que o colapso da corrente de tráfego é um fenômeno estocástico, caracterizado pela queda abrupta da velocidade média em função do aumento do fluxo de tráfego. O momento em que ocorre mudança abrupta de velocidade é associado ao colapso da corrente de tráfego e consequentemente à capacidade da via. Após a estimação da capacidade, a velocidade na capacidade foi determinada como a média das velocidades associadas ao fluxo na capacidade, enquanto que a densidade na capacidade foi obtida por meio da relação fundamental de tráfego. O método foi aplicado em um conjunto de 18 trechos de rodovias que atingem a capacidade e os resultados foram considerados satisfatórios após a comparação com valores encontrados na literatura. A calibração da relação fluxo-velocidade foi realizada individualmente para cada trecho de rodovia utilizado no estudo. A análise dos resultados mostra que as rodovias rurais apresentam em média valores maiores para velocidade de fluxo livre, capacidade, coeficiente de calibração, velocidade na capacidade e ponto de transição em relação as rodovias urbanas. Alem disso, o ponto de transição, que consiste em um patamar de velocidade de fluxo livre constante, apresenta valores significativamente menores em relação a literatura. / The following research aims to outline a calibration method of the speed-flow relationship presented in the Highway Capacity Manual (HCM). The method is applied on freeways and multilane highways, leveraging the guidelines made by the authors that developed the manual. In order to support the calibration, a sample with more than 1.700.000 observations was used, collected by 34 traffic sensors on four highways at São Paulo state (SP-280, SP-348, SP270, and SP-021). Furthermore, the data treatment and cleaning process aimed to remove observations considered inappropriate from a model calibration perspective by filtering the original data through three main criteria: traffic flow only composed by passenger car, traffic under normal operations condition, and only containing observations which free flow regime applies. The third filtering criteria was applied by splitting free flow and congested regime through the threshold of density at capacity. To this extent, a capacity estimation method was proposed, aiming to analyse the average speed variation in function of the traffic flow. The assumption of the method relies in the concept that the traffic flow breakdown is a stochastic process, characterized by the abrupt drop in the average speed as function of the traffic flow increase. The moment that abrupt drop occurs is related to traffic breakdown and, consequently, to the capacity freeway capacity. Once the capacity was estimated, the speed at capacity was defined as the average of the speed associated with the flow at capacity while the density at capacity was estimated through the fundamental relationship of the traffic flow. The method was applied in 18 delimited highway traffic perimeter that reached their corresponding capacity and the results were considered satisfactory after comparing against values found in the literature. Furthermore, the speed-flow relationships calibration was performed for each of them individually. The results show that rural highways have in average higher free flow speed, capacity, calibration coefficient, speed at capacity and breakpoint than urban highways. Moreover, the breakpoint, which consists a constant free flow speed level, presented significant lower values in comparison to the literature.

O objetivo desta tese é adaptar os procedimentos do Highway Capacity Manual 2000, para que possam melhor estimar a qualidade de serviço nas condições de tráfego encontradas em rodovias de pista simples estudadas. Para isto, foi desenvolvido um programa para calibração automática do simulador TRARR, baseado em um algoritmo genético e implementado em Perl. A calibração, que usa dados de velocidade e porcentagem de veículos em pelotões, coletado por meio de filmagens nos locais estudados, foi feita simultaneamente para cinco trechos diferentes, para que o simulador fosse capaz de reproduzir as características de um trecho típico de rodovia de pista simples do estado de São Paulo. Para garantir a qualidade dos resultados das simulações realizadas com o modelo recalibrado, procedeu-se à sua validação, usando-se um conjunto independente da dados. Partindo-se do pressuposto que a estrutura geral do método deveria ser mantida, a adaptação do HCM-2000 consistiu em obter novos valores para os diversos fatores de ajuste usados no processo de estimativa do nível de serviço. Para tanto, foram reproduzidos os experimentos de simulação realizados no desenvolvimento do procedimento para análise de rodovias de pista simples, à exceção do uso da versão recalibrada TRARR no lugar do simulador TWOPAS. Os fatores de ajuste adaptados foram: fator de pico horário (PHF), fator de ajuste de rampas (fG), fator de equivalência veicular (ET), fator para o efeito de zonas de ultrapassagem proibida (fnp) e fator de ajuste para o efeito combinado da divisão direcional do tráfego e de zonas de ultrapassagem proibida (fd/np). Com exceção do PHF, todos os demais fatores foram obtidos a partir de resultados de simulação. Os níveis de serviço observados nos cinco trechos estudados foram comparados com estimativas obtidas com a versão original do HCM-2000 e com a adaptação proposta. Os resultados desta comparação mostram que as estimativas obtidas com a adaptação proposta são mais precisas que as obtidas usando-se os valores originais do HCM-2000. / This thesis objective is to adapt the Highway Manual Capacity 2000, procedures to produce better estimates of service level and capacity for two-lane highways studied, In order to do this, an automatic calibration system, implementing a genetic algorithm and coded in Perl, was used to obtain a recalibrated version of TRARR, a two-lane highway simulation model. The calibration, that which uses speed and percent vehicles traveling in platoons data collected by videotaping, was carried out simultaneously for five different road segments, so that the recalibrated model would be able to reproduce traffic behavior on a typical two-lane road in the state of São Paulo, Brazil. The recalibrated model was validated using a second, independent, set of data. Based on the assumption that the general framework of the methodology should be kept, the adaptation was achieved by obtaining new values for the adjustment factors used in the procedure. Thus, the simulation experiments used in the development of HCM-2000 methodology were reproduced using the recalibrated TRARR model. The following adjustment factors were adapted: peak hour factor (PHF), grade adjustment factor (fG), heavy vehicle equivalence factor (ET), adjustment factor for the effect of non-passing zones (fnp) and adjustment factor for combined effect of directional distribution of traffic and percentage of no-passing zones (fd/np). With the exception of the PHF, all factors were obtained from simulation results. In order to assess the accuracy of the proposed adaptation observed levels of service were compared to the estimates obtained through the original HCM-2000 procedure and the modified procedure. The results of these comparisons clearly show that the estimates obtained using the proposed values for the adjustment factors are closer to the observed levels of service than the estimates obtained using the original values for the adjustment factors.

This thesis fills a missing piece in research on multimodal performance measures for traffic on streets and highways. The Highway Capacity Manual (HCM) published by the Transportation Research Board (TRB) provides Level of Service (LOS) methodologies which enable engineers and planners to evaluate the overall performance of roadways and highways based on the physical characteristics of facilities. This allows for the evaluation of those facilities and offers a means for recognizing issues and planning, designing, implementing, and ultimately assessing improvements. Originally, level of service was developed for automotive traffic only, but with recent developments as part of the complete streets movement, the performance of infrastructure for alternative transportation modes have also started being assessed in this fashion. There are methodologies in HCM 2010 for bicycle traffic at signalized intersections, all-way stop-controlled intersections, roadway and highway segments, but as of yet, no bicycle level of service methodology exists for two-way stop-controlled intersections. This work attempts to fill this gap. The methodology utilized for this report includes video collection of sample two-way stop-controlled intersections throughout California, collection of survey responses from viewers of video, and linear regression of collected survey responses with physical attributes of each sample intersection as the explanatory variables. Data was analyzed from both combined and individual street movements to determine the final equation set. The final methodology involves two separate procedures for major and minor streets at TWSC intersections. Final factors deemed significant in bicycle level of service analysis include sight distances, speed limits, presence of bus stops, presence and type of bicycle infrastructure, street widths and types of lanes present, pavement quality, and traffic flows.

Background: The aim of this thesis was to study transcription patterns and extracellular signalling of the hypertrophic heart to better understand the mechanisms initiating, controlling and maintaining cardiac hypertrophy. Cardiac hypertrophy is a risk factor for cardiovascular morbidity and mortality. Hypertrophy of the myocardium is a state, independent of underlying disease, where the myocardium strives to compensate for an increased workload. This remodelling of the heart includes physiological changes induced by a changed gene expression, alteration of the extracellular matrix and diverse cell-to-cell signalling. Shedding microvesicles and exosomes are membrane released vesicles derived from the plasma membrane, which can mediate messages between cells and induce various cell-related processes in target cells. Methods and materials: Two different microarray studies on different materials were performed. In the first study, cardiac myectomies from 8 patients with hypertrophic obstructive cardiomyopathy (HOCM) and 5 controls without cardiac disease were used. In the second study, myocardial tissue from 6 aorta ligated and 6 sham operated (controls) rats at three different time points (1, 6 and 42 days post-surgically) were analysed. To reveal differences in gene expression the materials were analyzed with Illumina whole genome microarray and multivariate data analysis (PCA and OPLS-DA). Cultured cardiomyocytes (HL-1) were incubated with and without growth factors (TGF-β2 or PDGF BB). Microvesicles and exosomes were collected and isolated after differential centrifugations and ultracentrifugations of the cell culture medium. The microvesicles and exosomes were characterized with dynamic light scattering (DLS), flow cytometry, western blot, electron microscopy and Illumina whole genome microarray. Results: The two different microarray studies revealed differentially expressed gene transcripts and groups of transcripts. When comparing HOCM patients to controls significant down-regulation of the MYH6 gene transcript and two immediate early genes (IEGs, EGR1 and FOS), as well as significant up-regulation of the ACE2, JAK2 and HDAC5 gene transcripts were found. In the rat model, 5 gene groups showed interesting clustering after multivariate data analysis (OPLS-DA) associated with the hypertrophic development: “Atherosclerosis”, “ECM and adhesion molecules”, “Fatty acid metabolism”, “Glucose metabolism” and “Mitochondria”. The shedding microvesicles were rounded vesicles, 40-300 nm in size and surrounded by a bilayered membrane. Chromosomal DNA sequences were identified in the microvesicles. The microvesicles could be taken up by fibroblasts resulting in an altered gene expression in the fibroblasts. The exosomes from cultured cardiomyocytes (incubated with TGF-β2 or PDGF BB) had an average diameter of 50-80 nm, similar to the unstimulated control exosomes. A large, for all cardiomyocyte derived exosomes, common pool of mRNA seems stable and a smaller pool varied in mRNA content according to treatment of the cardiomyocyte. Of the common mRNA about 14% were ribosomal, 14% were of unknown locus and 5% connected to the function of the mitochondria. Conclusions: The microarray studies showed that transcriptional regulation at a stable stage of the hypertrophic development is a balance of pro and anti hypertrophic mechanisms and that diverse gene groups are differently regulated at different time points in the hypertrophic progression. OPLS-DA is a very useful and powerful tool when analyzing gene expression data, especially in finding clusters of gene groups not seen with traditional statistics. The extracellular vesicle studies suggests that microvesicles and exosomes released from cardiomyocytes contain DNA and can be involved in events in target cells by facilitating an array of processes including gene expression changes. Different treatment of the cardiomyocyte influence the content of the exosome produced, indicating that the signal function of the exosome might vary according to the state of the cardiomyocyte. / Bakgrund: Syftet med den här avhandlingen var att studera transkriptions-mönster och extracellulär signalering vid hjärthypertrofi för att bättre förstå de mekanismer som startar, styr och underhåller tillväxten. Hjärthypertrofi, onormal tillväxt av hjärtmuskeln, är en riskfaktor för andra hjärt-kärlsjukdomar och dödlighet. Hypertrofi av hjärtmuskeln är ett tillstånd, oberoende av bakomliggande sjukdom, där hjärtmuskeln strävar efter att kompensera för ökad arbetsbelastning. Denna omställning av hjärtat innefattar fysiologiska förändringar orsakade av ett förändrat genuttryck, modifiering av miljön utanför cellen och ändrad cell-till-cell signalering. Mikrovesiklar och exosomer är små membranomslutna bubblor som frisätts från cellmembranet, ut i cellens omgivning. De kan förmedla budskap mellan celler och påverka olika processer i målceller. Metoder och material: Avhandlingen innefattar två olika microarraystudier på olika material. I den första studien användes hjärtbiopsier från 8 patienter med hypertrofisk obstruktiv kardiomyopati (HOCM) och 5 kontroller utan hjärtsjukdom. I det andra projektet användes hjärtvävnad från 6 aortaligerade och 6 skenopererade (kontroller) råttor vid tre olika tidpunkter (1, 6 och 42 dagar efter kirurgiskt ingrepp). För att påvisa skillnader i genuttryck analyserades proverna med Illumina helgenom microarray och multivariat dataanalys. Avhandlingens andra del innehåller två studier om mikrovesiklar och exosomer. Odlade hjärtmuskelceller (HL-1) stimulerades med tillväxt-faktorer (TGF-β2 eller PDGF BB) och ostimulerade celler användes som kontroll. Mikrovesiklar och exosomer renades fram med centrifugeringar och ultracentrifugering av cellodlingsmediet för att sedan karakteriseras med olika metoder för att studera storlek, ytmarkörer och innehåll. Illumina helgenom microarray användes för att studera microvesiklarnas och exosomernas mRNA innehåll. Resultat: I de två olika microarraystudierna hittades gentranskript och grupper av gentranskript som skiljde sig mellan kontroller och den hypertrofa hjärtvävnaden. När HOCM patientproverna jämfördes med kontroller hittades nedreglering av MYH6, EGR1 och FOS samt uppreglering av ACE2, JAK2 och HDAC5. Efter multivariat dataanalys av materialet från råtta, hittades 5 grupper av gentranskript med intressanta mönster som kunde kopplas till den hypertrofiska utvecklingen av hjärtmuskeln: "Ateroskleros", "ECM och adhesionsmolekyler", "Fettsyrametabolism", "Glukosmetabolis-men" och "Mitokondrien". Mikrovesiklarna hade en diameter på 40-300 nm och innehöll kromosomala DNA-sekvenser. När mikrovesiklarna överfördes till en annan celltyp (fibroblaster) resulterade det i ett förändrat genuttryck i fibroblasterna. Exosomer från hjärtmuskelcellerna som odlats med eller utan tillväxtfaktor hade en diameter på 50-80 nm. En stor pool av olika gentranskript var gemensam för alla exosomer oavsett stimulering eller ej. En mindre pool av gentranskript varierade i innehåll mellan de stimulerade och ostimulerade hjärtmuskelcellerna. I den gemensamma gentranskript poolen var ca 14 % ribosomala, ca 14 % var okända och ca 5 % var associerade till mitokondrien och dess funktion. Slutsats: Microarraystudierna visade att transkriptionsreglering i ett stabilt skede av hypertrofiutvecklingen är en balans mellan pro- och anti-hypertrofiska mekanismer och att olika gengrupper var olika reglerade vid olika tidpunkter i hjärtmuskeltillväxten. OPLS-DA är ett mycket användbart och kraftfullt verktyg när man analyserar genexpressionsdata, särskilt för att hitta grupper av gen-transkript som är svåra att upptäcka med traditionell statistik. Microvesikel- och exosomstudierna visade att mikrovesiklar och exosomer som frisätts från hjärtmuskelceller innehåller både DNA och RNA och kan vara inblandade i händelserna i målceller genom att underlätta en rad processer, inklusive ändringar av genuttryck. Olika stimulering av hjärtmuskelcellen kan påverka innehållet i exosomernas som produceras, vilket indikerar att exosomernas signalfunktion kan variera beroende på hjärtmuskelcellens tillstånd.

Die Vergrößerung des linken Vorhofes (LA) hat bei Katzen mit Herzerkrankungen eine prognostische Bedeutung und wird mit einem erhöhten Risiko der Entwicklung einer Links-herzinsuffizienz, von supraventrikulären Arrhythmien, von systemischer Thrombembolie und von plötzlichem Herztod in Verbindung gebracht. Bei Katzen mit idiopathischer (HCM) oder sekundärer linksventrikulärer (LV) Hypertrophie kommt es infolge einer LV diastolischen Dysfunktion zu Veränderungen von Größe und Funktion des LA. Ziel der prospektiven Studie war es, die Größe und Funktion des LA in einer Population gesunder Katzen und Katzen mit Kardiomyopathie elektrokardiographisch, radiologisch sowie echokardiographisch zu charakterisieren und die Tiergruppen sowie die diagnostischen Methoden miteinander zu vergleichen. Zur Untersuchung des LA wurden 59 Katzen berücksichtigt. Die Kontrollgruppe bestand aus 26 gesunden Katzen und die Gruppe der kranken Tiere aus 33 Katzen mit HCM oder einer sekundären LV-Hypertrophie. Davon waren 18 Katzen asymptomatisch und 15 hatten eine Linksherzinsuffizienz oder einen aortalen Thrombembolismus. Im EKG wurde die Dauer der P-Welle als diagnostisches Kriterium der LA-Größe untersucht. Bei der radiologischen Untersuchung wurde die Größe des LA in zwei orthogonalen Projektionsebenen subjektiv beurteilt. Außerdem wurde eine neue quantitative Messung des LA im latero-lateralen Strahlengang etabliert. Diese vergleicht die Größe des LA mit der Länge der Thorakalwirbel und wurde als vertebrale Vorhofgröße (LA-VHS) bezeichnet. Mit der transthorakalen zweidimensionalen (2D)- und M-Mode Echokardiographie konnte die Größe des LA durch unterschiedliche Messungen bestimmt werden. Im rechts-parasternalen Vierkammerblick wurden der maximale antero-posteriore Durchmesser des LA (LADs) gemessen und als echokardiographische Bezugsvariable („Gold Standard“) zur Charakterisierung der LA-Größe genutzt. Aus derselben Anschallung wurden die maximale apico-basale Länge des LA und die maximale Vorhoffläche ermittelt. In der rechts-parasternalen Darstellung des LV-Ausflusstraktes wurde unter Anwendung des M-Modes der LA dargestellt und das LA/Ao-Verhältnis berechnet. In der rechts-parasternalen kurzen Achse wurde der maximale Durchmesser des LA (LAmax) bestimmt und sowohl die so genannte „schwedische Methode“ als auch der M-Mode zur Bestimmung der Vorhofgröße und Berechnung der LA/Ao-Verhältnisse angewandt. Der LA wurde als vergrößert definiert, sobald LADs größer 1,60 cm war. Die Ergebnisse der weiteren echokardiographischen Messmethoden zur Charak-terisierung der LA-Größe wurden mit LADs verglichen und folgende Grenzwerte zur Diagnose einer Vergrößerung des LA für die einzelnen unterschiedlichen Messmethoden ermittelt: aus der rechts-parasternalen langen Achse die maximale LA-Länge 1,97 cm, die maximale LA-Fläche 2,80 cm2 und LA/Ao (M-Mode) 1,55, aus der rechts-parasternalen kurzen Achse LA/Ao (M-Mode) 1,54, LA/Ao („schwedische Methode“) 1,44 und LAmax 1,60 cm. Trotz ähnlicher Grenzwerte der LA/Ao-Verhältnisse konnte gezeigt werden, dass die Messwerte der unterschiedlichen Methoden untereinander abweichen und daher diagnostisch nicht austauschbar sind. Die Ergebnisse der elektrokardiographischen und radiologischen Messungen der LA-Größe wurden hinsichtlich ihrer diagnostischen Wertigkeit mit den echokardiographischen Unter-suchungen zur Diagnose einer LA-Vergrößerung verglichen. Das EKG (P-Welle) war wenig sensitiv aber spezifisch bei der Diagnosestellung einer LA-Vergrößerung. Die subjektive radiologische Beurteilung der LA-Größe hatte eine deutlich höhere diagnostische Treffsicherheit, wobei jedoch die Bestimmung der LA-VHS sowohl die größte Sensitivität als auch Spezifität zeigte. Zur Einschätzung der globalen Vorhoffunktion wurden die echokardiographischen Indices LA-Verkürzungsfraktion und LA-Flächenverkürzung herangezogen. Beide Variablen waren in der Gruppe der symptomatischen Katzen deutlich vermindert, verglichen mit der Kontrollgruppe und der Gruppe der asymptomatischen Katzen mit LV-Hypertrophie. Die Reservoirfunktion des LA wurde anhand der S-Welle und des S/D-Verhältnisses des Pulmonalvenenflusses beurteilt. In der Gruppe der symptomatischen Katzen mit LV-Hypertrophie lag das S/D-Verhältnis unter 1,0, was für eine gestörte Reservoirfunktion des LA sprach. Die Weiterleitungsfunktion des LA, charakterisiert durch die D-Welle des Pulmonalvenenflusses war in der Gruppe der symptomatischen Katzen im Gegensatz zu den beiden anderen Gruppen signifikant vermindert. Die Vorhofkontraktion („Booster-Funktion“) wurde anhand der transmitralen A-Welle und der AR-Welle des Pulmonalvenenflusses beurteilt. Hinsichtlich der AR-Welle lag kein Unterschied der Maximalgeschwindigkeit zwischen den drei Tiergruppen vor, jedoch war die Dauer der AR-Welle in der Gruppe der symptomatischen Katzen verlängert. Dieses wurde als Hinweis auf eine erhöhte Nachlast des LA (verminderte LV Dehnbarkeit) und/oder gestörte systolische Funktion des LA gewertet. Die Untersuchung der Blutflussgeschwindigkeit im linken Herzohr zeigte eine deutliche Abnahme in der Gruppe der symptomatischen Katzen und galt als hinweisend für eine gestörte Pumpfunktion und Blutstase. Es wurde ein Zusammenhang zwischen verminderter Blutflussgeschwindigkeit im linken Herzohr (< 0,23 m/s) und dem Risiko eines aortalen Thrombembolismus gefunden. Zusammenfassend kann festgestellt werden, dass eine LV-Hypertrophie bei Katzen zu einer Vergrößerung des LA, verminderter Funktion des LA und Blutstase in LA und im linken Herzohr führt. Somit bestätigt sich die Vermutung, dass der LA sowohl einen diagnostischen als auch prognostischen Wert bei der Untersuchung von Katzen mit HCM oder sekundärer LV-Hypertrophie hat. Die vorliegende Studie liefert einen Beitrag zur echokardiographischen Standarisierung der LA-Größe und LA-Funktion bei der Katze. Weitere Untersuchungen, insbesondere die invasive Validierung echokardiographischer Indices der Größe und Funktion des LA sind notwendig.