In Vitro Characterization of Protein:Nucleic Acid Liquid-Liquid Phase Separation by Microscopy Methods and Nanoparticle Tracking Analysis.

Uncontrolled assembly/disassembly of physiologically formed liquid condensates is linked to irreversible aggregation. Hence, the quest for understanding protein-misfolding disease mechanism might lie in the studies of protein:nucleic acid coacervation. Several proteins with intrinsically disordered regions as well as nucleic acids undergo phase separation in the cellular context, and this process is key to physiological signaling and is related to pathologies. Phase separation is reproducible in vitro by mixing the target recombinant protein with specific nucleic acids at various stoichiometric ratios and then examined by microscopy and nanotracking methods presented herein. We describe protocols to qualitatively assess hallmarks of protein-rich condensates, characterize their structure using intrinsic and extrinsic dyes, quantify them, and analyze their morphology over time. Analysis by nanoparticle tracking provides information on the concentration and diameter of high-order protein oligomers formed in the presence of nucleic acid. Using the model protein (globular domain of recombinant murine PrP) and DNA aptamers (high-affinity oligonucleotides with 25 nucleotides in length), we provide examples of a systematic screening of liquid-liquid phase separation in vitro.

Prion protein complexed to a DNA aptamer induce behavioral and synapse dysfunction in mice.

Conversion of the cellular prion protein (PrPC) into the scrapie form (PrPSc) is the leading step to the development of transmissible spongiform encephalopathies (TSEs), still incurable neurodegenerative disorders. Interaction of PrPC with cellular and synthetic ligands that induce formation of scrapie-like conformations has been deeply investigated in vitro. Different nucleic acid (NA) sequences bind PrP and convert it to ß-sheet-rich or unfolded species; among such NAs, a 21-mer double-stranded DNA, D67, was shown to induce formation of PrP aggregates that were cytotoxic. However, in vivo effects of these PrP-DNA complexes were not explored. Herein, aggregates of recombinant full-length PrP (rPrP23-231) induced by interaction with the D67 aptamer were inoculated into the lateral ventricle of Swiss mice and acute effects were investigated. The aggregates had no influence on emotional, locomotor and motor behavior of mice. In contrast, mice developed cognitive impairment and hippocampal synapse loss, which was accompanied by intense activation of glial cells in this brain region. Our results suggest that the i.c.v. injection of rPrP:D67 aggregates is an interesting model to study the neurotoxicity of aggregated PrP in vivo, and that glial cell activation may be an important step for behavioral and cognitive dysfunction in prion diseases.

Rabbit PrP Is Partially Resistant to in vitro Aggregation Induced by Different Biological Cofactors.

Prion diseases have been described in humans and other mammals, including sheep, goats, cattle, and deer. Since mice, hamsters, and cats are susceptible to prion infection, they are often used to study the mechanisms of prion infection and conversion. Mammals, such as horses and dogs, however, do not naturally contract the disease and are resistant to infection, while others, like rabbits, have exhibited low susceptibility. Infection involves the conversion of the cellular prion protein (PrPC) to the scrapie form (PrPSc), and several cofactors have already been identified as important adjuvants in this process, such as glycosaminoglycans (GAGs), lipids, and nucleic acids. The molecular mechanisms that determine transmissibility between species remain unclear, as well as the barriers to transmission. In this study, we examine the interaction of recombinant rabbit PrPC (RaPrP) with different biological cofactors such as GAGs (heparin and dermatan sulfate), phosphatidic acid, and DNA oligonucleotides (A1 and D67) to evaluate the importance of these cofactors in modulating the aggregation of rabbit PrP and explain the animal's different degrees of resistance to infection. We used spectroscopic and chromatographic approaches to evaluate the interaction with cofactors and their effect on RaPrP aggregation, which we compared with murine PrP (MuPrP). Our data show that all cofactors induce RaPrP aggregation and exhibit pH dependence. However, RaPrP aggregated to a lesser extent than MuPrP in the presence of any of the cofactors tested. The binding affinity with cofactors does not correlate with these low levels of aggregation, suggesting that the latter are related to the stability of PrP at acidic pH. The absence of the N-terminus affected the interaction with cofactors, influencing the efficiency of aggregation. These findings demonstrate that the interaction with polyanionic cofactors is related to rabbit PrP being less susceptible to aggregation in vitro and that the N-terminal domain is important to the efficiency of conversion, increasing the interaction with cofactors. The decreased effect of cofactors in rabbit PrP likely explains its lower propensity to prion conversion.

Phase separation of p53 precedes aggregation and is affected by oncogenic mutations and ligands.

Mutant p53 tends to form aggregates with amyloid properties, especially amyloid oligomers inside the nucleus, which are believed to cause oncogenic gain-of-function (GoF). The mechanism of the formation of the aggregates in the nucleus remains uncertain. The present study demonstrated that the DNA-binding domain of p53 (p53C) underwent phase separation (PS) on the pathway to aggregation under various conditions. p53C phase separated in the presence of the crowding agent polyethylene glycol (PEG). Similarly, mutant p53C (M237I and R249S) underwent PS; however, the process evolved to a solid-like phase transition faster than that in the case of wild-type p53C. The data obtained by microscopy of live cells indicated that transfection of mutant full-length p53 into the cells tended to result in PS and phase transition (PT) in the nuclear compartments, which are likely the cause of the GoF effects. Fluorescence recovery after photobleaching (FRAP) experiments revealed liquid characteristics of the condensates in the nucleus. Mutant p53 tended to undergo gel- and solid-like phase transitions in the nucleus and in nuclear bodies demonstrated by slow and incomplete recovery of fluorescence after photobleaching. Polyanions, such as heparin and RNA, were able to modulate PS and PT in vitro. Heparin apparently stabilized the condensates in a gel-like state, and RNA apparently induced a solid-like state of the protein even in the absence of PEG. Conditions that destabilize p53C into a molten globule conformation also produced liquid droplets in the absence of crowding. The disordered transactivation domain (TAD) modulated both phase separation and amyloid aggregation. In summary, our data provide mechanistic insight into the formation of p53 condensates and conditions that may result in the formation of aggregated structures, such as mutant amyloid oligomers, in cancer. The pathway of mutant p53 from liquid droplets to gel-like and solid-like (amyloid) species may be a suitable target for anticancer therapy.

Liquid-liquid phase separation and fibrillation of the prion protein modulated by a high-affinity DNA aptamer.

Structural conversion of cellular prion protein (PrPC) into scrapie PrP (PrPSc) and subsequent aggregation are key events associated with the onset of transmissible spongiform encephalopathies (TSEs). Experimental evidence supports the role of nucleic acids (NAs) in assisting this conversion. Here, we asked whether PrP undergoes liquid-liquid phase separation (LLPS) and if this process is modulated by NAs. To this end, two 25-mer DNA aptamers, A1 and A2, were selected against the globular domain of recombinant murine PrP (rPrP90-231) using SELEX methodology. Multiparametric structural analysis of these aptamers revealed that A1 adopts a hairpin conformation. Aptamer binding caused partial unfolding of rPrP90-231 and modulated its ability to undergo LLPS and fibrillate. In fact, although free rPrP90-231 phase separated into large droplets, aptamer binding increased the number of droplets but noticeably reduced their size. Strikingly, a modified A1 aptamer that does not adopt a hairpin structure induced formation of amyloid fibrils on the surface of the droplets. We show here that PrP undergoes LLPS, and that the PrP interaction with NAs modulates phase separation and promotes PrP fibrillation in a NA structure and concentration-dependent manner. These results shed new light on the roles of NAs in PrP misfolding and TSEs.