The diversity in morphological features and cell types among multicellular animals necessitates precise regulation mechanisms to ensure differential gene expression patterns. At the heart of regulatory programs are transcription factors (TF) and cis regulatory elements (CREs), the interplay of which guides the spatiotemporal specificity and dynamics of transcription. Tissue-specific gene expression is remarkably conserved during embryonic development. In the developing heart, for example, tissue patterning from fish to mammals is governed by a core set of TFs. While expression patterns and the coding sequences for TFs are highly conserved, the DNA sequences of regulatory elements associated with developmental genes are highly diversified. Tracing the orthology of CREs based on sequence similarity is therefore challenging, especially between distantly related species. One hypothesis to explain this apparent contrast is the preservation of regulatory function despite sequence divergence in CREs. Another explanation is the syntenic arrangement of the regulatory elements and their target genes, which are often conserved in diverse taxonomic groups, also contribute to the conserved expression profiles. However, current methods lack the power to detect homology based on synteny. Furthermore, while individual reports of functional conservation exist, it remains unclear how widespread this phenomenon might be for highly divergent regulatory sequences. A systematic approach to identify divergent orthologous regulatory elements and evaluate their function is therefore necessary. Mammals and birds diverged over 300 million years ago from their last common ancestor and independently evolved a four-chamber heart. This setup provides a unique model system to explore conserved regulatory programs in the context of highly divergent DNA. In this work, I address this challenge by profiling the regulatory genome in the embryonic hearts of chicken and mouse at equivalent developmental stages. Comparative analysis reveals conserved tissue-specific gene expression profiles and 3D chromatin structure, contrasting the low degree of sequence homology in CREs as estimated by pairwise sequence alignment. To uncover orthologous CREs independent of sequence alignability between chicken and mouse, I adapt and optimize a new algorithm called Interspecies Point Projection (IPP). IPP works by using conserved synteny in developmental loci to map corresponding genomic locations between highly diverged genomes. Compared to traditional alignment-based tools, this framework uncovers up to 5-fold more putative orthologs in chicken, and up to 9-fold in more distantly related vertebrate species. This improvement is due to uncovering both sequence-conserved and sequence-divergent orthologous CREs. The functional equivalence of identified orthologs is validated with equivalent functional genomic data combined with in-depth sequence and machine learning analyses. Despite lack of alignability, divergent orthologs exhibit conserved signatures in terms of active chromatin profiles, predicted tissue-specificity, and shared TF binding specificities, similarly to their sequence-conserved counterparts. I further validate their functional conservation using in vivo reporter assays, where identified orthologous chicken enhancers drive conserved and specific expression patterns in the developing mouse heart. Finally, I show that divergent orthologs have a higher degree of binding site shuffling in their sequences, indicating regulatory information can be conserved through maintenance of TFBS composition. Such sequence turnover allows flexible adaptation of the regulatory genome during evolution but obstructs detection of functional orthologs by alignment-based comparative studies. Taken together, these findings challenge the notion that sequence homology is required for conserved regulatory activity and uncover widespread functional conservation of regulatory sequences across large evolutionary distances.
