FUS, HNRNPA1, and other RNA-binding proteins implicated in ALS and other neurodegenerative diseases have a floppy tail. These intrinsically disordered regions (IDRs) allow proteins to undergo liquid phase separation and form liquid droplets. Studies of this phenomenon, which has been linked to protein aggregation, have been largely limited to in vitro experiments. Now, in the January 12 Cell, scientists led by Clifford Brangwynne, Princeton University, New Jersey, report they have developed a tool to manipulate IDRs in living cells. With a flash of light, they could coax FUS and HNRNPA1 to condense into droplets. Flicking the light on and off many times and increasing its intensity forces the proteins in those droplets to aggregate, they report. This technique may help scientists understand how FUS and other IDR-containing proteins go from liquid-like droplets to toxic aggregates.

“This is a very powerful method that provides a quantum leap forward in our ability to study and manipulate granules made from RNA-binding proteins,” said Benjamin Wolozin, Boston University, who was not involved in the work. For the first time, the method allows scientists to rigorously examine phase separation in cells, he said. It would be useful to vet in a disease model, he added. 

Light-induced clustering: A few seconds after turning on blue light, light-activated IDRs start sticking together (white dots). More accumulate with time. [Courtesy of Cell, Shin et al., 2017.]

Liquid phase separation—akin to how oil forms droplets in water—gives rise to membrane-less organelles such as nucleoli and stress granules. IDR-containing proteins also form toxic aggregates in patients with amyotrophic lateral sclerosis, frontotemporal dementia, and multisystem proteinopathy. First author Yongdae Shin and colleagues set out to develop a tool that would allow them to study liquid, gel, and aggregated phases in vivo. A gel is a semisolid, jelly-like phase that is more structured than a liquid, but not as rigid as an aggregate.

The researchers took advantage of a photoreceptor called cryptochrome 2, which is found in the flowering plant Arabidopsis thaliana. In response to blue light, Cry2 changes conformation and readily associates with other Cry2 molecules to form clusters (Bugaj et al., 2013). Researchers have used this property to study interactions between intramembrane proteins and cell signaling (Kennedy et al., 2010). 

The researchers fused the light-responsive portion of Cry2 to the IDRs of FUS and HNRNPA1, as well as to the RNA-processing protein DDX4, creating what they called optoIDRs. Once activated by light, Cry2 self-associated, bringing IDRs closer together, much like they would when bound to RNA. Concentrated, IDRs tend to undergo liquid-liquid phase separation to form droplets. In response to blue light, optoIDRs did the same, readily forming liquid droplets in mouse fibroblasts. This optoDroplet system not only worked cell-wide when illuminating the cytoplasm all at once (see image above), it also promoted local droplets when the light beam was focused on just a portion of the cell.

Shin and colleagues then used blue light to control the concentration of light-activated molecules to determine the optimal conditions for droplet formation. They found a concentration threshold, or phase boundary, for all three optoIDRs. Above that concentration, the proteins clustered together to form droplets; below it, molecules stayed in solution. The threshold for optoDDX4 was twofold lower than that for optoFUS, implying that the DDX4 IDRs have stronger intermolecular interactions that allow the protein to more readily form droplets.

How far Shin exceeded the phase boundary affected dynamics within the protein clusters. When he used light to activate the minimum concentration of proteins—“shallow” crossing of the boundary—round droplets formed that readily dissipated after the light was turned off (see image below). Even with repeatedly turning the light on and off every 30 minutes, dynamic droplets formed and then dissolved completely. Photobleaching experiments suggested that optoIDRs fluidly moved in and out of these droplets, implying that they were liquid-like.

However, if Shin pushed “deeper” past the phase boundary by turning up the intensity of blue light, irregularly-shaped aggregates of protein started to form. These stuck around for several minutes (see image below). Though these initially fell apart when the light was turned off, the more intense the light, the more stable they became with subsequent on/off cycles. Photobleaching suggested these clusters took a more stable, gel-like form, as molecules were slower to enter and exit. Some clusters seemed permanent, persisting for as long as the researchers ran the experiment—more than six hours. It seems they had formed aggregates, which could be related to the pathological inclusions seen in ALS patients, said Shin. 

Phase Schematic.

IDRs (red tail) fused with Cry2 (orange cylinder) form droplets under blue light. Modest light elicits round, fluid droplets (top right), while more intense light creates irregular-shaped gels (bottom right). [Courtesy of Cell, Shin et al., 2017.]

The study challenges recent reports that aggregates form directly from dynamic liquid droplets (Sep 2016 news; Patel et al., 2015; Molliex et al., 2015). Rather, it suggests that aggregates arise from more stable gels, Shin said. Gels may hold molecules in place long enough to form amyloid-like fibrils, he told Alzforum. Brangwynne’s group plans to analyze whether the permanent clusters they observed are indeed amyloid fibrils, and how they form from gels. The researchers also plan to see how liquid and gel states contribute to cell function, and whether aggregates are toxic.

Nicolas Fawzi of Brown University in Providence, Rhode Island, agreed that it would be important to look at these phases on the molecular or atomic level, as it’s still not clear how these states differ from a molecular perspective (Burke et al., 2015). “The structure of the IDRs in the liquid and gel states is important to understand how they convert to irreversible aggregates,” he said. He still believes aggregates could form from liquids, since the authors used only portions of RNA-binding proteins, loaded them with bulky tags, and examined aggregate formation on a short timescale. “If you let it stay liquid for hours, days, years, a liquid might nucleate an aggregate.”

Fawzi was curious to know how changing the phase of these proteins might tune their function within a cell. If scientists fused the light-sensitive portion of Cry2 with full-length RNA-binding proteins, they might be able to study that, he said.—Gwyneth Dickey Zakaib


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. Helical Tail Holds Sway Over TDP-43 Packaging

Paper Citations

  1. . Optogenetic protein clustering and signaling activation in mammalian cells. Nat Methods. 2013 Mar;10(3):249-52. Epub 2013 Feb 3 PubMed.
  2. . Rapid blue-light-mediated induction of protein interactions in living cells. Nat Methods. 2010 Dec;7(12):973-5. Epub 2010 Oct 31 PubMed.
  3. . A Liquid-to-Solid Phase Transition of the ALS Protein FUS Accelerated by Disease Mutation. Cell. 2015 Aug 27;162(5):1066-77. PubMed.
  4. . Phase separation by low complexity domains promotes stress granule assembly and drives pathological fibrillization. Cell. 2015 Sep 24;163(1):123-33. PubMed.
  5. . Residue-by-Residue View of In Vitro FUS Granules that Bind the C-Terminal Domain of RNA Polymerase II. Mol Cell. 2015 Oct 15;60(2):231-41. Epub 2015 Oct 8 PubMed.

Further Reading


  1. . ALS/FTD Mutation-Induced Phase Transition of FUS Liquid Droplets and Reversible Hydrogels into Irreversible Hydrogels Impairs RNP Granule Function. Neuron. 2015 Nov 18;88(4):678-90. Epub 2015 Oct 29 PubMed.
  2. . Intrinsic disorder in proteins involved in amyotrophic lateral sclerosis. Cell Mol Life Sci. 2017 Apr;74(7):1297-1318. Epub 2016 Nov 12 PubMed.

Primary Papers

  1. . Spatiotemporal Control of Intracellular Phase Transitions Using Light-Activated optoDroplets. Cell. 2017 Jan 12;168(1-2):159-171.e14. Epub 2016 Dec 29 PubMed.