Identification and characterization of a protein complex involved in chromatin architecture and the evolution of Drosophila species

Lukacs, Andrea (2024): Identification and characterization of a protein complex involved in chromatin architecture and the evolution of Drosophila species. Dissertation, LMU München: Medizinische Fakultät

Abstract

Centromeres and pericentromeric heterochromatin are involved in a number of essential functions including cell division, silencing of repetitive DNA elements and spatial organization of the genome. The clustering of these regions around chromocenters during interphase is required for their proper function and involves a complex and finely tuned network of protein-protein interactions. Despite their critical importance, centromeric and pericentromeric chromatin are poorly conserved even among closely related species and are, in fact, often involved in the formation of species. This apparent paradox is often the result of intrinsically fast-evolving repetitive DNA elements embedded into heterochromatin. The rapid evolution of such repetitive elements poses a threat to cellular fitness thereby exerting a selective pressure on the silencing machinery responsible for their management. Such process typically leads to the rapid coevolution of heterochromatin proteins acting like suppressors of fast-evolving DNA elements. This coevolution can, in turn, lead two populations once belonging to the same species to diverge by developing species-specific heterochromatin, that can eventually function as a postzygotic barrier between the two newly generated species. The fruit fly sibling species Drosophila melanogaster (D.mel) and Drosophila simulans (D.sim) have been a privileged model for studying the formation of species since over a century and provide an excellent example of how hybrid incompatibilities can arise within chromatin. Hybrids from D. melanogaster mothers and D. simulans fathers fail to develop because of the detrimental genetic interaction of the three hybrid incompatibility (HI) genes Hmr, Lhr and gfzf. Genetic studies of the three HI genes have revealed pleiotropic phenotypes, with Hmr and Lhr mutations disrupting oogenesis, female fertility and repetitive elements silencing and gfzf mutants having a broad spectrum of phenotypes including defects in cell cycle regulation. In addition to their genetic interaction, HMR and LHR proteins interact within a complex network of protein-protein interactions that is important for the architecture and function of centromeric and pericentromeric chromatin. For GFZF, on the other hand, few molecular data are available and there is no evidence of molecular interactions with the other two HI genes. Although the last decade has brought enormous progress in the field, the molecular details underlying both the divergent evolution of these hybrid incompatibility factors in the respective pure species and their genetic interactions in interspecific hybrids are still poorly understood. A major twist in the field has been the finding that the expression of HMR and LHR proteins has diverged during the evolution in D.mel and D.sim species and that the two proteins are both overrepresented in hybrids. However, while these findings revealed the importance of a proper quantitative balance of HMR and LHR, the HMR/LHR protein-protein interaction network could only be studied in overexpressing conditions so far. Characterizing the HMR/LHR protein complex in native conditions could therefore pave the way, on the one hand for understanding how these two proteins interact to mediate normal pericentromeric and centromeric functions in pure species, and, on the other hand, how their interaction network is altered in hybrids where they are overexpressed. To address these questions, in the first publication, we used affinity purification coupled with Mass Spectrometry (AP-MS) and revealed for the first time the existence of a stable six-subunit HMR/LHR protein complex in native conditions. In addition to HMR and LHR, the complex contains the two nucleolar proteins nucleoplasmin (NLP) and nucleophosmin (NPH), as well as the two non-characterized proteins, CG33213 and CG4788. For these last two proteins our publication provided the first molecular characterization and we named them Buddy Of HMR 1 (BOH1) and Buddy Of HMR 2 (BOH2), respectively. In addition, as a resource for the field, we published a detailed description of the intricate network of interactions (interactome) involving each complex component. After identifying the complex we went further and generated two different mutants targeting two different Hmr domains, to dissect how HMR interacts with other complex components and how disrupting such interactions affects HMR localization and function. Our results suggest that the integrity of the HMR/LHR complex is necessary for both HMR physiological function in pure species and its toxic function in hybrids. Next, we started from the HMR/LHR native complex and induced HMR/LHR overexpression to mimic a hybrid background in a cell culture system and asked how the HMR protein-protein interaction network is altered upon their overexpression. A range of new chromatin interactors, from architectural proteins like insulators to zinc-finger DNA binding proteins, appear to specifically interact when HMR is in excess, suggesting that these may be binding with low affinity and therefore only observed when HMR amount is not limiting, such as in hybrids. Finally, we set out to study HMR subnuclear localization with respect to CENP-A and HP1a, a centromeric and a pericentromeric marker, respectively. Our findings allow us to build a model that reconciles previous controversies and suggests that HMR is neither centromeric nor pericentromeric but it is rather sitting in the middle, bridging these two types of chromatin by forming a complex that interacts with both. In the second publication, we focused on the third and less known HI factor and asked how GFZF localizes in D.mel, D.sim and hybrids and how it molecularly interacts with HMR to cause hybrid incompatibility. We used in situ hybridization in polytene chromosomes to describe for the first time GFZF localization in D.mel, D.sim and hybrids. In addition, here we show the first evidence of a molecular interaction between GFZF and HMR by looking at their respective localization in both polytene chromosomes and cell lines where HMR was present in physiological amounts or overexpressed with LHR (hybrid-mimicking condition). Strikingly, while the two HI proteins HMR and GFZF occupy distinct and non-overlapping territories in pure species, their localization merges in the hybrid background.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Themengebiete: 600 Technik, Medizin, angewandte Wissenschaften ; 600 Technik, Medizin, angewandte Wissenschaften > 610 Medizin und Gesundheit
• Fakultäten: Medizinische Fakultät
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 25. März 2024
• 1. Bericht­erstatter:in: Imhof, Axel