FIGURE SUMMARY
Title

Mutant Human FUS Is Ubiquitously Mislocalized and Generates Persistent Stress Granules in Primary Cultured Transgenic Zebrafish Cells

Authors
Acosta, J.R., Goldsbury, C., Winnick, C., Badrock, A.P., Fraser, S.T., Laird, A.S., Hall, T.E., Don, E.K., Fifita, J.A., Blair, I.P., Nicholson, G.A., Cole, N.J.
Source
Full text @ PLoS One

Whole mount and cell cultures of FUS-GFP transgenic zebrafish.

(A) Transgenic zebrafish larvae whole mounts showed cytosolic mislocalization of mutant human FUS in FUS-R521C-GFP in comparison to FUS-WT-GFP which was restricted to cell nuclei. (B) FUS-R521C-GFP showed greater cytosolic distribution in comparison to FUS-WT-GFP in zebrafish primary cell cultures. (C) Confocal images of 48 hpf transgenic zebrafish spinal cord further demonstrate mislocalization of mutant FUS-521C-GFP (green) in motor neurons (red) (arrows). Images are maximum projections captured using a Leica SPE5 confocal microscope. Sagittal sections (upper images) of Tg(s1020tGAL4: UASmCherry) (Scott and Baier, 2009) and transverse sections (lower images) of Tg(HB9: mK02caax) membrane localised mk02 expressed in motorneurons by HB9 promoter (Flanagan-Steet et al 2005) with either FUS-WT-GFP or FUS-521C-GFP as indicated. Scale bar = 20 μm.

Mislocalization of mutant FUS-GFP was also found in motor neurons.

(A) FUS-R521C-GFP was similarly mislocalized to the cytosol in motor neurons (labeled with 39.4D5 for islet1 and islet2 homeodomain marker). (B) Quantification of FUS-GFP signal in nucleus vs. cytosol in 39.4D5 marked motor neurons demonstrated that FUS-R521C-GFP was significantly more cytosolic compared to FUS-WT-GFP (post-hoc Tukey HSD. ** = P<0.01, N.S. = >0.05, n = 40 for all samples. Error bars represent SE. Scale bars = 20 μm.

Ubiquitous FUS-GFP SG assembly in zebrafish cells.

(A) FUS-GFP SGs formed in cultured transgenic zebrafish cells after heat-shock at 40°C for 30 mins (arrows). However, SGs were more abundant in mutant FUS-R521C-GFP cultures (right panels) than in FUS-WT-GFP. (B) Motor neurons (39.4D5-labeled cells) were not particularly susceptible to SG assembly. (C) SGs were reversible when cells were allowed to recover at 37°C for another 30 mins. Some persistent SGs were still present particularly in FUS-R521C-GFP cultures. Motor neurons labeled with 39.4D5 readily reversed SGs. (D) Stress granules (SGs) were also induced by sodium arsenite (Na3AsO3; 0.2 mM) treatment. SGs formed in mutant (arrows) but not in the FUS-WT-GFP line after Na3AsO3 treatment for 1 hr. Similar to heat-shocked cells, 39.4D5-labeled cells were no more susceptible to chemical-induced SG formation. (E) Chemical-induced FUS-GFP containing SGs were reversible in both lines. Reversibility also occurred readily in 39.4D5 labelled motor neurons.

Punctuate staining with SG marker eIF3e was commonly found adjacent to or surrounding FUS-GFP SGs.

Scale bars = 10 µm; Insets = 1 μm.

Zebrafish cell culturing protocol supports the growth and differentiation of motor neurons. (A) Cell cultures from transgenic zebrafish embryos expressing GFP under the motor neuron promoter islet-1 (islet1: GFP) demonstrated that motor neurons represented <10% of the cells in culture and exhibited extensive differentiation with axonal growth and branching (arrow). (B) Zebrafish neural cell-associated antibodies obtained from the Developmental Studies Hybridoma Bank (University of Iowa) were used: 39.4D5 [anti-islet-1/2] – primary motor neuron-specific transcription factor; Zn12 [anti-L2/HNK-1] – neural cell adhesion molecule (labels many different neural subtypes); 3A10 [anti-neurofilament] - derived from a neurofilament-associated antigen and labels a subset of hindbrain spinal cord projecting neurons such as Mauthner neurons (Brand et al. 1996) but appears not to label islet 1/2 expressing motor neurons; Zn8 [anti-neurolin] - expressed by secondary but not primary motoneurons during zebrafish development. This labeling demonstrated a mix of different neural subtypes in the cultures. (C) FUS-GFP was expressed in the cell soma of motor neurons and was not extensively transported into neurites. Scale bar = 20 μm.

Images confirming the presence of SGs. (A) Heat-shocked cells treated prior with the SG inhibitor cycloheximide (CHX) did not form FUS-GFP containing SGs. (B) Islet1: GFP cells did not form SGs after heat-shock, indicating that the GFP tag was not involved in SG formation. (C) Non-GFP fused cultures of FUS-R521C counterstained with anti-human FUS show the presence of aggregates of similar appearance to SGs tagged with GFP. (D) The majority of non-transgenic cells did not form SGs that could be stained with anti-FUS. Scale bars = 10 μm.

Co-localization of FUS-GFP SGs and FUS (stained with polyclonal anti-FUS antibody, red) confirms the presence of human FUS in SGs. Scale bar = 20 μm; insets = 1 μm. Brand M, Heisenberg CP, et al. (1996) Mutations in zebrafish genes affecting the formation of the boundary between midbrain and hindbrain. Development 1236179–190.

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ PLoS One