Project Summary Increasing lines of evidence in mammals have shown that certain acquired traits during paternal environmental exposure could be “memorized” in sperm and transmitted to the future generations, implicating epigenetic inheritance via the sperm. Particularly, diet-induced metabolic disorders in the father are heritable in mammals, suggesting this type of epigenetic inheritance has a long-term impact in many metabolic-related human diseases. To date, the exact sperm “epigenetic carrier” that responds to paternal dietary changes and transmits the intergenerational phenotype has remained elusive, but presumably involves DNA methylation, chromatin status or non-coding RNAs. Recently, in a high-fat diet (HFD) mouse model, we showed that highly enriched, tRNA-derived small RNAs (tsRNAs) from sperm are altered in both expression profiles and RNA modification after paternal HFD exposure. We also showed that sperm tsRNAs, along with their RNA modifications, are required for intergenerational transmission of paternally acquired metabolic disorders (Chen et.al. Science, 2016). This discovery raises the open question of how sperm tsRNAs, along with their RNA modifications, mediate the embryonic developmental programming that affects the offspring phenotype (Chen et.al. Nat Rev Genet, 2016). To understand the underlying mechanisms, we propose to (1) identify the target of sperm tsRNAs by injecting sperm tsRNAs (from HFD and control males) into zygotes, followed by comparative single-cell embryo RNA-seq of 2- to 4-cell embryos and bioinformatics analysis. (2) Because significant alterations of m5C (5-methylcytidine) in sperm tsRNAs from HFD males have been found, we will determine whether the m5C cytosine RNA methyltransferase, DNMT2, is essential for sperm tsRNAs to transmit acquired metabolic disorder to offspring, by utilizing HFD model in Dnmt2-/- and Dnmt2+/+ male mice, and injecting their sperm tsRNAs into normal zygote followed by examining the metabolic phenotype of F1 offspring. (3) Moreover, in the Dnmt2-/- versus Dnmt2+/+ HFD model, we will examine the changes of RNA expression profiles by RNA-seq, and the RNA modification profiles in sperm tsRNAs by a high-throughput approach (based on Liquid chromatography-tandem mass spectrometry, LC-MS/MS) to quantify multiple RNA modifications, which not only will detect the changes of m5C, but also discover novel RNA modification changes in the presence or absence of DNMT2. Data from the proposed study will provide insights into the mechanism by which sperm tsRNAs mediate intergenerational inheritance of acquired metabolic disorders and facilitate our understanding of the etiology of human diseases originated from diet-based transgenerational effect.