Zhou Wu

The phenotype of individuals is determined by the genome together with the environment. The diverse phenotypic variation of domesticated animals is reflected in their genome. With this thesis, I employed genomic analyses to obtain valuable insights into the underlying genetic causes of size variation in chickens. The genetic causes of two dwarf phenotypes (autosomal dwarfism and bantam) in chickens were identified, prioritized, and validated. I addressed the genetic basis of the dwarf phenotypes in terms of the underlying genetic variants, gene expression, and the human-mediated bantamization history. I provide a comprehensive understanding of the function of dwarf related genes, which can potentially be utilized in poultry breeding.

In chapter 2, I describe a fine-mapping experiment to identify the causal gene and genetic variant underlying autosomal dwarfism (adw) in chickens. Within the earlier detected candidate region on chromosome 1 (52–56 Mb), a novel nonsense mutation NP_001006244.1:p.(Trp59*) was identified in the transmembrane protein 263 gene (TMEM263). The effect of the associated nonsense mutation is expected to truncate the transmembrane protein within the membrane-spanning domain, resulting in a loss-of-function alteration. The results point towards a potentially causative role of TMEM263 in growth reduction and provide the molecular basis of the development in autosomal dwarf chickens.

Chapter 3 focuses on a study to identify the genetic basis of the bantam phenotype in Dutch traditional chicken breeds by using Genome-Wide Association Studies and a meta-analysis. The Dutch breeds can be divided into three different clusters. Different bantam associated genomic regions were identified depending on different bantam clusters of breeds, which include the growth-related genes HMGA2 and PRDM16. The bantam associated regions clearly show distinct bantam introgression signals and haplotype diversity across the bantam clusters. The observation of heterogeneity indicates that the bantam phenotype in the Dutch chicken breeds is a complex trait caused by multiple underlying genes and variants. Collectively, the genomic analyses on the bantam phenotype offer valuable insights into the human-mediated bantam crossbreeding history.

The objective of chapter 4 is to provide a comprehensive understanding of the bantam phenotype from a gene expression perspective. RNA-seq data of embryonic samples derived from bantam and normal-sized chickens was generated to identify differentially expressed genes (DEGs) associated with size variation at two developmental stages (E5 and E15). Overall, the gene expression data shows a developmental stage-specific pattern across the samples. I was able to detect DEGs associated with growth in bantam chickens as early as E5. DEGs compared between breeds or developmental stages showed limited overlap, in agreement with the expected heterogeneity of the bantam phenotype. Many of the detected genes are involved in growth-related signaling pathways (e.g., FGF) and biological processes (myogenesis and osteogenesis) in the expression study.

In chapter 5, I describe the population structure of Dutch traditional chicken breeds. Using the Dutch chicken breeds as a model, I demonstrate the complexity of the admixed subdivided population structure. This complexity was dissected and attributed to human-mediated selection and crossbreeding, of which I highlight the influence of selection based on different management purposes and for phenotypic diversification. In particular, I identified signatures of selection involved in phenotypic selection between Dutch traditional chicken breeds, such as loci associated with the bantam trait and the plumage color. In addition, I describe a case study (of Drenthe fowl bantam) providing further genomic insight into the ongoing process of bantamization and the effect of human-mediated crossbreeding.

Finally, in the general discussion (chapter 6), I address the challenges and opportunities for the genomic study and the utilization of dwarf phenotypes. With regard to genomic analyses, I emphasize the importance of accounting for population structure in local traditional breeds as well as the phenotypic and genetic heterogeneity in dwarfisms. Moreover, I discuss how to translate the methodological approaches and analytical strategies used to study naturally occurring introgression into studying human-mediated bantamization. In addition, I discuss an experimental attempt to knock-out the ortholog of the adw candidate gene (i.e. TMEM263) in zebrafish using CRISPR-Cas. I end my discussion by highlighting the potential implementation of dwarf phenotypes in poultry breeding by addressing the options to introduce a dwarf phenotype into a population, with specific emphasis on the option of using genome-editing.