Genomic studies of Babesia bigemina and Anaplasma marginale infection levels in cattle
Carregando...
Arquivos
Data
2021-02-25
Autores
Romero, Andrea Renata da Silva
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Carrapatos são responsáveis pela propagação de patógenos como a Babesia bigemina e Anaplasma marginale que causam as doenças babesiose e anaplamose bovina, respectivamente. A presença constante do carrapato nos rebanhos nacionais acarreta perdas econômicas e dificuldade na inclusão de raças taurinas, por serem animais mais susceptíveis a doenças parasitárias. Assim, o objetivo desse estudo foi estimar componentes de variância, acurácia de predição dos valores genéticos genômicos (GEBV), e realizar estudo de associação genômica ampla (GWAS) para os níveis de infecção da B. bigemina (BigIL) e A. marginale (AmIL). No capítulo dois foram feitas as análises para BigIL utilizando 1.882 animais (1.716 Braford e 166 Hereford) com genótipos imputados para painel de ~777.000 marcadores. Os componentes de variância foram obtidos por abordagem bayesiana. Para as estimativas de valores genéticos genômicos (GEBV) os fenótipos de contagem de carrapato (CC) e níveis de infecção por B. bovis (BovIL) foram incluídos no modelo multivariado visando ganhos em acurácia de predição da BigIL. As análises de predição genômica e GWAS foram realizadas utilizando o melhor preditor linear não viesado genômico (GBLUP). Foi observada baixa herdabilidade para BigIL (0,090) indicando maior influência ambiental do que genética aditiva. A estimativa da correlação genética entre BigIL e CC foi baixa (0.240), em contraste com a correlação entre BigIL e BovIL que foi alta (0,624). Os valores de correlação genética podem ter contribuído com aumento da acurácia quando comparamos o modelo univariado (0.246±0.189) e multivariado (0.606±0.180). No GWAS foi possível observar o a herança poligênica da BigIL. Nas regiões próximas dos SNP que explicaram maior parte da variância foram identificados 35 genes, os quais têm sido relacionados com resposta imune nos cromossomos 1, 4, 5, 7, 10, 13, 18, 20, 27, 29. No capítulo três foi realizada quantificação do número de cópias da A. marginale no sangue dos animais pela técnica de qPCR. Os componentes de variância foram estimados por análise Bayesiana e o método empregado para GWAS e estimação dos valores GEBV foi o melhor preditor linear não viesado genômico (GBLUP) usando modelo multivariado. As estimativas de herdabilidade e acurácia de predição foram baixas 0,090 e 0,180, respectivamente. A correlação genética entre AmIL com BigIL foi alta 0,793, enquanto as correlações entre AmIL com BovIL e contagem de carrapato foram próximas a zero. Ao explorar as top 10 regiões do GWAS, foram encontrados genes candidatos relacionados com interação entre hospedeiro e patógeno e resposta à inflamação, localizadas nos cromossomos 2, 5, 8, 10, 13, 15, 17, 20, 24, e 29. Foram identificadas sobreposições com QTL previamente descritos para susceptibilidade a tuberculose em bovinos e QTL com escore de células somáticas. Assim, os resultados obtidos no presente trabalho podem contribuir para estabelecimento de estratégias de seleção de animais que consigam manter níveis mais baixos de infecções por agentes causadores da Tristeza Parasitária Bovina.
Ticks are responsible for the spread of pathogens such as Babesia bigemina and Anaplasma marginale that cause the diseases babesiosis and anaplamosis, respectively. The constant presence of the tick in national herds leads to economic losses and difficulty in the inclusion of taurine breeds, as they are animals more susceptible to parasitic diseases than indicine cattle. Thus, the objective of this study was to estimate components of variance, accuracy of prediction, and to conduct a genome-wide association study (GWAS) for the levels of infection of B. bigemina (BigIL) and A. marginale (AmIL). In chapter two, analyses for BigIL were performed using 1,882 animals (1,716 Braford and 166 Hereford) with genotypes imputed to a panel of ~777,000 markers. The variance components were obtained using the Bayesian approach. For the estimation of genomic breeding values (GEBV), tick count (CC) phenotype and levels of B. bovis infection (BovIL) were included in the multivariate model aiming at gains in BigIL prediction accuracy. Genomic prediction and GWAS analyses were performed using genomic best linear unbiased prediction method (GBLUP). Low heritability was observed for BigIL (0.090) indicating greater environmental influence than genetic. The estimate of the genetic correlation between BigIL and CC was low (0.240), in contrast to the correlation between BigIL and BovIL which was high (0.624). The genetic correlation values may have contributed to an increase in accuracy when comparing the univariate (0.246±0.189) and multivariate (0.606±0.180) models. GWAS results showed a polygenic inheritance of BigIL. In the regions close to the SNPs that explained most of the additive variance were identified 35 genes previously related to the immune response located on chromosomes 1, 4, 5, 7, 10, 13, 18, 20, 27, 29. In chapter three, the variance components were estimated by Bayesian analysis and the method used for GWAS and estimation of the values was the best linear biased genomic predictor (GBLUP) using the multivariate model. The estimates of heritability and prediction accuracy were 0.090 and 0.180, respectively. The genetic correlation between AmIL and BigIL was high 0.793, while the correlations between AmIL and BovIL and tick count were close to zero. Exploring the top 10 regions of GWAS, candidate genes were found related to host-pathogen interaction and response to inflammation, located on chromosomes 2, 5, 8, 10, 13, 15, 17, 20, 24, and 29, overlapping with QTL previously described for susceptibility to tuberculosis in cattle and QTL with somatic cell score were identified. Thus, the results obtained in the present work can contribute to the establishment of selection strategies that result in animals that are able to maintain lower levels of infections by agents that cause Bovine Parasitic Sadness.
Ticks are responsible for the spread of pathogens such as Babesia bigemina and Anaplasma marginale that cause the diseases babesiosis and anaplamosis, respectively. The constant presence of the tick in national herds leads to economic losses and difficulty in the inclusion of taurine breeds, as they are animals more susceptible to parasitic diseases than indicine cattle. Thus, the objective of this study was to estimate components of variance, accuracy of prediction, and to conduct a genome-wide association study (GWAS) for the levels of infection of B. bigemina (BigIL) and A. marginale (AmIL). In chapter two, analyses for BigIL were performed using 1,882 animals (1,716 Braford and 166 Hereford) with genotypes imputed to a panel of ~777,000 markers. The variance components were obtained using the Bayesian approach. For the estimation of genomic breeding values (GEBV), tick count (CC) phenotype and levels of B. bovis infection (BovIL) were included in the multivariate model aiming at gains in BigIL prediction accuracy. Genomic prediction and GWAS analyses were performed using genomic best linear unbiased prediction method (GBLUP). Low heritability was observed for BigIL (0.090) indicating greater environmental influence than genetic. The estimate of the genetic correlation between BigIL and CC was low (0.240), in contrast to the correlation between BigIL and BovIL which was high (0.624). The genetic correlation values may have contributed to an increase in accuracy when comparing the univariate (0.246±0.189) and multivariate (0.606±0.180) models. GWAS results showed a polygenic inheritance of BigIL. In the regions close to the SNPs that explained most of the additive variance were identified 35 genes previously related to the immune response located on chromosomes 1, 4, 5, 7, 10, 13, 18, 20, 27, 29. In chapter three, the variance components were estimated by Bayesian analysis and the method used for GWAS and estimation of the values was the best linear biased genomic predictor (GBLUP) using the multivariate model. The estimates of heritability and prediction accuracy were 0.090 and 0.180, respectively. The genetic correlation between AmIL and BigIL was high 0.793, while the correlations between AmIL and BovIL and tick count were close to zero. Exploring the top 10 regions of GWAS, candidate genes were found related to host-pathogen interaction and response to inflammation, located on chromosomes 2, 5, 8, 10, 13, 15, 17, 20, 24, and 29, overlapping with QTL previously described for susceptibility to tuberculosis in cattle and QTL with somatic cell score were identified. Thus, the results obtained in the present work can contribute to the establishment of selection strategies that result in animals that are able to maintain lower levels of infections by agents that cause Bovine Parasitic Sadness.
Descrição
Palavras-chave
Anaplasmose, Babesiose, Bovinos, Genômica, Polimorfismo de nucleotídeo único