Isoform-specific expression and ratio changes accompany oxidant-induced peripherin aggregation in a neuroblastoma cell line

BR A I N R ES E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research Report

Isoform-specific expression and ratio changes accompany oxidant-induced peripherin aggregation in a neuroblastoma cell line Jesse R. McLean⁎, Janice Robertson Centre for Research in Neurodegenerative Diseases, University of Toronto, Toronto, ON, Canada

A R T I C LE I N FO

AB S T R A C T

Article history:

The type III intermediate filament peripherin is found associated with pathological

Accepted 16 September 2011

inclusions present within motor neurons of patients with amyotrophic lateral sclerosis

Available online 22 September 2011

(ALS). Peripherin intra-isoform associations contribute to filament network formation at defined stoichiometric ratios. Distinct biochemical signatures characterize peripherin isoform

Keywords:

expression in traumatic neuronal injury and motor neuron disease, while disruptions to per-

ALS

ipherin alternative splicing or translation are associated with inclusion formation. In our

Oxidation

efforts to identify pathological relationships between peripherin isoform expression and in-

Inclusion

clusion formation, we provide evidence of peripherin isoform-specific expression and ratio

Aggregation

changes with concomitant, dose-dependent inclusion formation in response to oxidative

Splicing

stress. Upon increasing exposure to physiologically relevant levels of hydrogen peroxide in

Peripherin

Neuro-2a cells, we observed a significant increase and decrease in peripherin isoforms Per-58 and Per-45, respectively, with peripherin-specific perikaryal aggregation of filaments 10–15 μm in diameter. Interestingly, peripherin-immunoreactive inclusions showed no overt carbonylation, suggesting that aggregation may serve a physiologically relevant role during oxidative stress. These findings provide novel insight into the biological significance of peripherin isoforms and inclusion formation, with relevance to the pathology of ALS. © 2011 Elsevier B.V. All rights reserved.

1.

Introduction

Amyotrophic lateral sclerosis is a devastating late-onset neurodegenerative disorder caused by the rapid progressive degeneration of upper and lower motor neurons. A major neuropathological hallmark of familial and sporadic ALS is the presence of perikaryal inclusions and axonal swellings immunoreactive for the type III intermediate filament (IF) protein peripherin (Corbo and Hays, 1992; He and Hays, 2004; Migheli et al., 1993; Mizuno et al., 2011; Wong et al., 2000; Xiao et al.,

2008). The mechanism by which peripherin is associated with these inclusions remains unknown, however, it may be related to abnormal changes in peripherin isoform expression (McLean et al., 2008; Robertson et al., 2003; Swarup et al., 2011; Xiao et al., 2008). While peripherin filament networks assemble as a result of intra-isoform associations among alternatively spliced or translated variants at defined stoichiometric ratios (McLean et al., 2010; Xiao et al., 2006), disruptions to normal isoform expression or the generation of abnormal splice variants have been associated with the formation of periph-

⁎ Corresponding author at: Neuroregeneration Laboratories, McLean Hospital/Harvard Medical School, 115 Mill Street, Belmont, MA 02478, USA. Fax: +1 617 855 2522. E-mail address: [email protected] (J.R. McLean). 0006-8993/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2011.09.032

58

BR A IN RE S E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

erin inclusions (Gros-Louis et al., 2004; Leung et al., 2004; McLean et al., 2008; Robertson et al., 2003; Xiao et al., 2008). As a neuronal IF protein, peripherin is thought to play an integral role in the development of the neuronal cytoskeleton and to provide mechanical support to terminally differentiated neurons (Helfand et al., 2003; Troy et al., 1992). Peripherin expression is a highly regulated process mediated by a number of cis- and trans-acting factors acting throughout the gene region (McLean and Robertson, 2011). While little is known about the influence of exogenous factors on normal peripherin expression, in vitro studies have identified certain neurotrophins (NGF and FGF) (Choi et al., 2001; Leonard et al., 1987, 1988; Parysek and Goldman, 1987; Portier et al., 1983; Thompson et al., 1992) and proinflammatory cytokines (LIF and IL-6) (Lecomte et al., 1998; Sterneck et al., 1996) as transcriptional inducers of peripherin expression. In this regard, it is not surprising that enhanced peripherin expression occurs after traumatic neuronal injury (Beaulieu et al., 2002; Kriz et al., 2005; Oblinger et al., 1989; Troy et al., 1990) and in ALS (Robertson et al., 2003; Xiao et al., 2008). Interestingly, differential isoform-specific expression and ratio changes are observed after sciatic crush and middle cerebral artery occlusion, in transgenic mouse models of motor neuron disease, including Per, Per;L−/−, SOD1 and TDP-43 mice, and in ALS (McLean et al., 2010; Robertson et al., 2003; Swarup et al., 2011; Xiao et al., 2008). As these injurious or disease conditions are likely to encompass aspects of cellular stress as a result of oxidative damage, we sought to identify whether exposure to physiologically relevant increases in hydrogen peroxide (H2O2) concentration was capable of modifying peripherin isoform-specific expression and morphology.

2.

Results

2.1. Peripherin isoform-specific expression and ratio changes accompany exposure to increasing concentrations of H2O2 in N2a cells To assess peripherin isoform expression in response to oxidative stress, we exposed N2a cells to physiologically relevant increasing concentrations of H2O2 (Halliwell et al., 2000). We used N2a cells for their robust peripherin expression (Djabali et al., 1999; Portier et al., 1983) and sensitivity to oxidative stress (Calderon et al., 1999; Fernaeus et al., 2005; Rojo et al., 2008). A lethal dose-50 (LD50) was achieved in N2a cells after administration of ~250 μM of H2O2 (n = 3; Fig. 1, i). With increasing concentrations of H2O2, peripherin protein expression changes were evident on Western blot, as indicated by bands corresponding to Per-58, Per-56, and Per-45, as well as two unknown bands at ~50 and ~40 kDa (Fig. 1, ii). Quantitative analysis of total peripherin protein expression showed a significant dose-dependent ~23–46% decrease in the 10– 100 μM range when compared to baseline expression levels (0 μM), while a return to ~79% of baseline expression levels occurred between the 250 and 500 μM range (n = 3; Fig. 1, iii). Significant isoform-specific expression changes were observed in the 100–500 μM range, resulting in a ratio shift (Fig. 1, iv); here, Per-58 accounted for ~57–61% of total isoform content in the 100–500 μM range, as compared to ~ 43–50% in the 0–50 μM

range, while Per-45 accounted for ~11–15% as compared to ~16–26% in the same ranges.

2.2. Dose-dependent, peripherin-specific aggregation accompanies H2O2 administration We next examined peripherin filament morphology in response to oxidative stress in N2a cells. With increasing concentrations of H2O2, we noticed a loss in peripherin filament networks and the formation of peripherin-immunoreactive perikaryal inclusions in the 100–500 μM range (Fig. 2, i, arrows). Interestingly, by phase contrast (Fig. 2, i, gray panels), we observed a retraction of peripherin-immunoreactive neuritic processes (chevron arrowheads) as low as 10 μM. It is of note, that the cytoskeletal structure of adherent cells at higher H2O2 concentrations remained intact. As such, we were interested in identifying a cytoskeletal marker that did not form inclusions. Double labeling of peripherin with the microtubule protein α-tubulin at 500 μM revealed widespread cytoskeletal distribution of α-tubulin with no apparent structural abnormalities (Fig. 2, ii), indicating that inclusion formation was a specific event involving certain proteins, including peripherin. Quantitation of the number of peripherin-immunoreactive inclusions revealed a significant increase in the number of N2a cells with inclusions in the 250–500 μM range when compared to 0 μM (n = 3; Fig. 2, iii); here, we observed that ~ 19 and ~60% of N2a cells contained inclusions at 250 and 500 μM, respectively, when compared to ~1.6% at 0 μM. We assessed the ultrastructure of these peripherinimmunoreactive inclusions by electron microscopy. With exposure to increasing concentration of H2O2, N2a cells displayed several cellular abnormalities indicative of oxidative stress, including irregular membrane contour, mitochondrial swelling, and internal vacuolization (Fig. 3, i, top panels). A prominent feature at higher H2O2 concentrations was the formation of large focal inclusions within the cytoskeleton (Fig. 3, i, top panels, arrowheads). These inclusions were composed of aggregates of bundled filaments, with each filament having a discernable width of 10–15 nm, suggesting the presence of IFs (Fig. 3, i, bottom panels, arrows). We observed a positive relationship between the size of these aggregates and the H2O2 dose (Fig. 3, i, compare bottom panels). We performed an immunogold labeling and found that peripherin was highly specific for these aggregate structures (Fig. 3, ii, bottom panels).

2.3. Peripherin inclusions are not immunoreactive for carbonylated residues A major consequence of cellular oxidation by reactive oxygen species (ROS) is the process of protein carbonylation, the nonenzymatic addition of aldehydes or ketones to specific amino acid residues. As the typical fate of most carbonylated proteins is either degradation or aggregation (Nystrom, 2005), we wanted to identify whether peripherin inclusions were carbonylated. Using DNP as a marker for oxidation, we identified an increase in the number of carbonylated cytoskeletal proteins in N2a cells with exposure to increasing concentrations of H2O2 (Fig. 4, i). Surprisingly, upon double labeling with peripherin and DNP at 500 μM, both peripherin filaments (Fig. 4, ii, arrows) and inclusions (Fig. 4 ii, arrowheads) were not carbonylated.

BR A I N R ES E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

59

Fig. 1 – N2a cell viability and peripherin isoform expression. (i) N2a cell viability was analyzed by MTT assay after exposure to increasing concentrations of H2O2 to determine an appropriate dosing strategy. (ii) Total protein extractions from H2O2-treated N2a cells were visualized by Western blot using peripherin and GAPDH antibodies. (iii) Quantitative analysis revealed changes in total peripherin protein expression relative to GAPDH expression. (iv) Significant isoform-specific ratio changes were observed in the 100–500 μM range, with increasing and decreasing Per-58 and Per-45 expression, respectively. For MTT and total protein expression: * = p < 0.05. For isoform-specific expression: p < 0.05 for * = Per-58; # = Per-45. Bars represent standard error mean.

3.

Discussion

Peripherin-immunoreactive inclusions are pathological hallmarks found within motor neurons of most ALS patients (Corbo and Hays, 1992; He and Hays, 2004; Migheli et al., 1993; Mizuno et al., 2011; Wong et al., 2000; Xiao et al., 2008). Understanding how they are formed is likely to reveal some shared pathological features between familial and sporadic ALS (Xiao et al., 2006). This study provides evidence that oxidative stress is associated with changes in peripherin isoform expression and aggregation. Upon exposure to increasing physiologically relevant concentrations of H2O2, we observed changes in the expression of Per-58 and Per-45 that were accompanied by a dose-dependent increase in peripherin aggregation. These data may have relevance to the pathophysiology of ALS, where the co-occurrence of oxidative stress, peripherin-immunoreactive inclusions, and peripherin isoform expression abnormalities are found. We have previously demonstrated that peripherin intraisoform ratio changes result in the formation of peripherin inclusions in vitro and in vivo (McLean et al., 2008; Robertson et al., 2003; Xiao et al., 2008). This may occur through the generation of abnormal alternative splice variants, such as Per-61 in human mutant SOD1 and TDP-43 mice (Robertson et al., 2003; Swarup et al., 2011), or through disruptions in normal intraisoform stoichiometry, such as Per-28 or Per-45 expression

changes in ALS (McLean et al., 2008; Xiao et al., 2008). In the current study, we observed two distinct changes in peripherin isoform expression upon exposure to increasing concentrations of H2O2: an initial loss in total protein expression at lower H2O2 doses; and, an increase in total protein expression accompanied by intra-isoform ratio changes and perikaryal aggregation at higher doses. The loss in protein expression may be explained, in part, by the retraction of peripherin-immunoreactive N2a neurites in response to oxidative stress. The rise in total protein expression may occur as existing peripherin filaments aggregate and are, therefore, unable to turnover and/or degrade while new peripherin is generated. From our previous work, we have shown that Per-45 is constitutively expressed with the full-length isoform Per-58 and is necessary to stabilize filament networks (McLean et al., 2008). Here, we show that exposure to H2O2 is sufficient to modify peripherin isoform expression and induce aggregation, presumably through abnormal intra-isoform associations arising from stoichiometric differences between Per-58 and Per-45. While we do not yet understand the reason for these intra-isoform changes, we have provided evidence that peripherin isoform expression changes and aggregation may be triggered by oxidative stress. While oxidative injury is a contributing factor to the pathogenesis of ALS (Robberecht, 2000; Simpson et al., 2003), the extent to which ROS contribute to impaired neuronal function is uncertain. Here, exposure to increasing concentrations of H2O2

60

BR A IN RE S E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

Fig. 2 – Peripherin inclusion formation following H2O2 exposure in N2a cells. (i) A polyclonal peripherin antibody (red) and DAPI nucleic acid stain (blue) revealed the formation of peripherin-immunoreactive inclusions in N2a cells at higher H2O2 concentrations (arrows). Phase contrast shows a dose-dependent retraction of peripherin-immunoreactive neurites with increasing concentrations of H2O2 (DIC; gray panels). (ii) Labeling with monoclonal α-tubulin (green), polyclonal peripherin (red), and DAPI nucleic acid stain (blue) shows that other cytoskeletal proteins are preserved despite the presence of large intracellular inclusions at 500 μM H2O2. (iii) Quantitative analysis of peripherin inclusions revealed a significant increase in their formation from 250 to 500 μM H2O2. * = p < 0.05. Bars represent standard error mean.

in N2a cells resulted in irregular membrane contour, mitochondrial swelling, and internal vacuolization. Perhaps the most prominent abnormality, however, was the formation of large diameter (10–15 μm) perikaryal inclusions. These inclusions were comprised of aggregates of IF-sized (10–15 nm) linear bundles that were heavily immunoreactive to peripherin upon immunogold labeling. Interestingly, the microtubule protein α-tubulin did not display any structural abnormalities akin to peripherin, suggesting that peripherin aggregation is conditional upon a specific set of criteria and not necessarily a reflection of the effects of non-specific oxidative stress. The inclusions we report here are similar to the filament-like ultrastructure sometimes found within Lewy body-like inclusions, Bunina bodies, hyaline conglomerate inclusions, and skein-like inclusions in ALS (Sasaki and Maruyama, 1993, 1994). Certainly, it is reasonable to suspect that peripherin-immunoreactive inclusions, which have thus far been reported in all sub-types of ALS inclusions, including Bunina bodies, ubiquitinated inclusions, hyaline conglomerate inclusions, and axonal spheroids (Corbo and Hays, 1992; He and Hays, 2004; Migheli et al., 1993; Mizuno et al., 2011; Wong et al., 2000), may form as a result of dynamic intraisoform associations in response to stress. Additionally, further investigation into peripherin interactions with other IFs, including the neurofilaments and vimentin, as well as their response(s) to different pathological stressors, may shed light into the mechanisms of inclusion formation (Xiao et al., 2006). Protein carbonylation is considered to be a normal physiological process that prepares abnormally modified or misfolded proteins for degradation. To this end, cells are able to minimize the damaging effects of ROS in a non-enzymatic manner involving proteolysis, however, protein aggregation has also been observed as a consequence of increased carbonylation (Levine, 2002; Nystrom, 2005). Extensive protein carbonylation has been reported in the human mutant SOD1G93A mouse (Andrus et al., 1998) and within motor neurons of ALS patients (Ferrante et al., 1997; Niebroj-Dobosz et al., 2004). We looked to identify whether peripherin backbone moieties were targets of protein carbonylation during oxidative stress. Surprisingly, despite widespread cellular DNP reactivity, peripherin aggregates and remaining filaments were largely carbonyl free. While it is unlikely that peripherin carbonylated residues would exist buried within the aggregate, as we are using a polyclonal antibody that recognizes multiple epitopes and because we have previously shown that peripherin may act as a partition between distinct protein layers within ALS inclusions (Sanelli et al., 2007), we are cautious with interpreting this data. That such a prominent cytoskeletal protein should aggregate, yet show no overt carbonylation in response to oxidative stress remains an enigma and runs contrary to general theories about cellular degradative mechanisms in response to ROS (Levine, 2002; Nystrom, 2005). Indeed, the pathological significance of nIF inclusion formation remains controversial, with theories implicating them as either neuroprotective, providing a cytoskeletal scaffold from which other damaged or toxic proteins may be sequestered (Julien et al., 2005; Strong, 2003) or harmful, with downstream effects detrimental to normal cellular function (Caughey and Lansbury, 2003; Lansbury and Lashuel, 2006); unfortunately, most conclusions remain speculative at this time.

BR A I N R ES E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

61

Fig. 3 – Ultrastructure of N2a cells following H2O2 administration. (i) N2a cells treated with H2O2 shows signs of oxidative damage, including irregular membrane contour, mitochondrial swelling, and internal vacuolization (top panels). The normal scattered arrangement of 10–15 nm filaments (bottom panel, arrows, 0 μM condition) arranges into bundled linear sheets (bottom panel, arrows, 100 μM condition) in a dose-dependent manner to form aggregates 10–15 μm in diameter (bottom panel, 500 μM condition). Arrowheads in the top panels indicate the area of N2a cells depicted in the bottom panels. (ii) Immunogold labeling with polyclonal peripherin antibody identifies peripherin as constituent of these aggregates in the 500 μM condition as compared to the diffuse labeling in the 0 μM condition. Specificity was confirmed by omitting the primary antibody (−ve Ab, 500 μM condition). Black bar: 10 μm; white bar: 100 nm.

4.

Conclusion

The selective activation or inhibition of alternative isoforms in response to various pathological stressors is gaining interest in a number of neurodegenerative diseases (Anthony and Gallo, 2010). This study may provide an inroad into helping us understand the relationship(s) among ALS pathologies and peripherin abnormalities, including changes in isoform expression and inclusion formation.

supplemented with 10% fetal bovine serum and 100 mg/mL streptomycin. H2O2 [30% in phosphate-buffered saline (PBS), pH 7.4] was purchased from Sigma-Aldrich (St. Louis, MO). Cell viability was assessed using 3-(4,5-dimethylthiazol-2-yl)2,5,diphenyl-tetrazolium bromide (MTT). Briefly, N2a cells were incubated with MTT (5 mg/mL in PBS; Sigma-Aldrich) for 2 h at 37 °C and the absorbance at 550 nm determined following solubilization with isopropanol. Oxidative stress was induced by treating N2a cells with 0, 10, 25, 50, 100, 250 or 500 μM of H2O2 for 6 h. 5.2.

5.

Experimental procedures

5.1.

Cell culture and reagents

The mouse neuroblastoma cell line Neuro-2a (N2a) was maintained at 37 °C under an atmosphere of 5% CO2 in Opti-Modified Eagle's medium (Invitrogen, Carlsbad, CA)

Immunoblotting

Total protein lysates were harvested in 62.5 mmol/L Tris–HCl (pH 6.8) containing 2% sodium dodecyl sulfate (SDS) and protease inhibitors (P-8340; Sigma-Aldrich). Samples were assayed for protein concentration using the bicinchoninic acid assay and then diluted in 2× loading buffer [160 mmol/L Tris–HCl, pH 6.8, 30% glycerol, 4% SDS, 10% β-mercaptoethanol, and

62

BR A IN RE S E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

Fig. 4 – Immunocytochemical localization of protein carbonyls in H2O2-treated N2a cells. (i) N2a cells treated with H2O2 show increased carbonylation as detected by a monoclonal DNP antibody (red). Intense peripherin labeling (Per; green) was observed at higher H2O2 concentrations as a result of protein aggregation. The primary antibody was omitted to identify non-specific DNP effects (−ve DNP, 500 μM). N2a cells were also treated with NaBH4 and FeSO4 at 500 μM to block or enhance carbonylation as positive and negative controls, respectively. (ii) At higher magnification, DNP (red) was distributed widely throughout the cells at 500 μM, while dark areas of non-DNP labeling were also observed (chevron arrows). Double labeling of peripherin (green) with DNP (red) revealed that peripherin filaments (arrow) or aggregates (arrowheads) did not colocalize with DNP. Phase contrast confirms an intact cytoskeleton at 500 μM (DIC; gray panel). Blue = DAPI; bar = 20 μM.

0.02% bromophenol blue] and boiled for 5 min. 10 μg protein loadings were analyzed on 10% SDS-polyacrylamide gels and then blotted to polyvinylidene fluoride membrane. For immunoblotting, membranes were blocked with 3% skimmed milk powder in Tris-buffered saline (TBS) containing 0.2% Tween-20 for 1 h at room temperature, then diluted 1:5000 with polyclonal peripherin antibody in blocking solution overnight at 4 °C. A monoclonal glyceraldehyde 3-phosphate

dehydrogenase (GAPDH) antibody (H86504M; Biodesign, Saco, ME), diluted 1:5000 in blocking solution, was used as the internal loading control. Antibody binding was revealed using horseradish peroxidase-conjugated anti-rabbit or anti-mouse IgG and an enhanced chemiluminescence detection system (NEN Life Science Products, Woodbridge, ON). To quantitate peripherin isoform expression following H2O2 treatment, immunoblots were scanned to ImageJ software (National

BR A I N R ES E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

Institutes of Health, Bethesda, MD) for densometric analysis. Using the integrated density function, after background subtraction, we measured the sum of the pixels from Per-58, Per56, Per-45, or unidentified peripherin bands as a function of GAPDH expression and the control condition (0 μM). Isoform ratios were calculated by dividing individual isoform expression values by total isoform content. Statistical tests were performed with Prism 4.0 software (GraphPad, La Jolla, CA). For comparisons of the changes in peripherin isoform expression, we used one-way ANOVA and Tukey's post-test, with p < 0.05 considered significant. 5.3.

Immunocytochemistry

N2a cells grown on glass coverslips were washed in PBS (pH 7.4) and fixed in 4% paraformaldehyde for 30 min at room temperature. Cells were rehydrated with PBS and permeabilized with 0.1% Triton X (TX)-100. Cells were then blocked for 30 min in 5% bovine serum albumin (BSA) with 0.3% TX100 in PBS at room temperature. For detection of protein oxidation, carbonyl groups were converted into dinitrophenyl hydrazones by reaction with 1 mg/mL 2,4-dinitrophenylhydrazine (DNPH) (Sigma-Aldrich) prepared in 2 N HCl for 30 min. Cells were labeled with rabbit polyclonal peripherin (AB1530; 1:1000; Chemicon), mouse monoclonal dinitrophenyl (DNP) (D8406; 1:500; Sigma-Aldrich), or α-tubulin (T9026; 1:1000; Sigma-Aldrich) diluted in blocking solution for 1 h at room temperature. Cells were then labeled for 30 min with mouse and rabbit IgGs conjugated to Alexa Fluors 488 (green) and 594 (red), respectively, diluted 1:300 in blocking solution. Cells were counterstained with DAPI Nucleic Acid Stain (Invitrogen) diluted 1:100 for 10 min at room temperature. Cells were viewed using a Leica DM6000 digital microscope (Leica Microsystems, Germany) and images captured using an Orca-ER digital camera (Hamamatsu, Japan) with Openlab software (Improvision, England). To quantitate inclusion formation, at least 10 random field-of-views were analyzed across three independent experiments and performed a one-way ANOVA and Tukey's post-test, with p < 0.05 considered significant. Chemical and immunochemical controls were used to define carbonyl-specific binding. Chemical reduction of free carbonyls was performed by incubating 500 μM-treated H2O2 sections with 25 mM sodium borohydride (NaBH4) in 80% methanol for 30 min at room temperature before incubation with DNPH (negative control). Enhanced protein carbonylation was achieved in the 500 μMtreated H2O2 sections treated with 100 μM FeSO4 (positive control). Immunospecificity of the DNP antibody was assessed by omitting the primary antibody. 5.4.

Electron microscopy

Pelleted N2a cells treated with 0, 100, or 500 μM of H2O2 were fixed for 1 h with 2.5% glutaraldehyde in cacodylate buffer [0.1 M sodium cacodylate [(CH3)2AsO2Na·3H2O]; pH 7.3–7.4] and 5 mM CaCl2 (pH 6.8) at room temperature and then treated with 0.1% tannic acid in primary fixative for 30 min at room temperature. Samples were then washed in cacodylate buffer and postfixed in 1% cacodylate-buffered osmium tetroxide for 30 min at room temperature. After several washes

63

with cacodylate buffer, samples were dehydrated in graded ethanol washes and embedded in Epon-Araldite epoxy resin at 37 °C (SPI Supplies/Structure Probe, West Chester, PA). Thin sections (80 nm thickness) were cut on a Reichert-Jung E microtome (Leica Microsystems) and collected on electrolytic copper 200 mesh grids (SPI Supplies/Structure Probe). Sections were stained with uranyl acetate and lead citrate and examined with a Hitachi H7000 transmission electron microscope (Hitachi High-Technologies Canada, Toronto, ON) at an accelerating voltage of 75 kV. For immunogold labeling, pelleted N2a cells were initially fixed in 0.2% glutaraldehyde, 4% paraformaldehyde, and 2% uranyl acetate in cacodylate buffer for 30 min at room temperature. Samples were washed and dehydrated as described above and embedded in LR White resin (SPI Supplies/Structure Probe) at 45 °C. Thin sections were washed with TBS containing 1% BSA (pH 7.2) and then incubated with rabbit polyclonal peripherin antibody (AB1530; 1:1000) diluted 1:100 in 0.1% BSA in 0.1% TBS (pH 7.2) overnight at 4 °C. The sections were washed in 0.1% BSA in 0.1% TBS with 0.1% Tween-20 (pH 7.2) on a shaker, followed by a subsequent wash at pH 8.2 (to prevent dissociation of gold particles). Control grids omitting the primary antibody were included. The sections were then immersed in goat anti-rabbit IgG conjugated to 10 nm colloidal gold particles (Sigma-Aldrich) diluted 1:50 in 1% BSA in 0.1% TBS for 1 h. The grids were then washed in 0.1% BSA in 0.2% TBS with 0.1% Tween-20 and washed again, but with 0.05% sodium azide added. Thin sections were then fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.3) for 10 min at room temperature, washed in distilled water, and lightly counterstained with uranyl acetate and lead citrate.

Acknowledgments The authors wish to thank Steven Doyle at the Microscopy Imaging Laboratory (University of Toronto, Toronto, ON, Canada) for preparing cellular pellets for ultrastructural analysis. JR is a Canada Research Chair in the Molecular Mechanisms of ALS. This work was supported by grants from the Canadian Institutes of Health Research (to JR), the ALS Society of Canada (to JR), and the Natural Sciences and Engineering Research Council of Canada (to JM).

REFERENCES

Andrus, P.K., Fleck, T.J., Gurney, M.E., Hall, E.D., 1998. Protein oxidative damage in a transgenic mouse model of familial amyotrophic lateral sclerosis. J. Neurochem. 71, 2041–2048. Anthony, K., Gallo, J.M., 2010. Aberrant RNA processing events in neurological disorders. Brain Res. 1338, 67–77. Beaulieu, J.M., Kriz, J., Julien, J.P., 2002. Induction of peripherin expression in subsets of brain neurons after lesion injury or cerebral ischemia. Brain Res. 946, 153–161. Calderon, F.H., Bonnefont, A., Munoz, F.J., Fernandez, V., Videla, L.A., Inestrosa, N.C., 1999. PC12 and neuro 2a cells have different susceptibilities to acetylcholinesterase–amyloid complexes, amyloid25-35 fragment, glutamate, and hydrogen peroxide. J. Neurosci. Res. 56, 620–631.

64

BR A IN RE S E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

Caughey, B., Lansbury, P.T., 2003. Protofibrils, pores, fibrils, and neurodegeneration: separating the responsible protein aggregates from the innocent bystanders. Annu. Rev. Neurosci. 26, 267–298. Choi, D.Y., Toledo-Aral, J.J., Lin, H.Y., Ischenko, I., Medina, L., Safo, P., Mandel, G., Levinson, S.R., Halegoua, S., Hayman, M.J., 2001. Fibroblast growth factor receptor 3 induces gene expression primarily through Ras-independent signal transduction pathways. J. Biol. Chem. 276, 5116–5122. Corbo, M., Hays, A.P., 1992. Peripherin and neurofilament protein coexist in spinal spheroids of motor neuron disease. J. Neuropathol. Exp. Neurol. 51, 531–537. Djabali, K., Piron, G., de Nechaud, B., Portier, M.M., 1999. alphaB-crystallin interacts with cytoplasmic intermediate filament bundles during mitosis. Exp. Cell Res. 253, 649–662. Fernaeus, S., Reis, K., Bedecs, K., Land, T., 2005. Increased susceptibility to oxidative stress in scrapie-infected neuroblastoma cells is associated with intracellular iron status. Neurosci. Lett. 389, 133–136. Ferrante, R.J., Browne, S.E., Shinobu, L.A., Bowling, A.C., Baik, M.J., MacGarvey, U., Kowall, N.W., Brown Jr., R.H., Beal, M.F., 1997. Evidence of increased oxidative damage in both sporadic and familial amyotrophic lateral sclerosis. J. Neurochem. 69, 2064–2074. Gros-Louis, F., Lariviere, R., Gowing, G., Laurent, S., Camu, W., Bouchard, J.P., Meininger, V., Rouleau, G.A., Julien, J.P., 2004. A frameshift deletion in peripherin gene associated with amyotrophic lateral sclerosis. J. Biol. Chem. 279, 45951–45956. Halliwell, B., Clement, M.V., Long, L.H., 2000. Hydrogen peroxide in the human body. FEBS Lett. 486, 10–13. He, C.Z., Hays, A.P., 2004. Expression of peripherin in ubiquinated inclusions of amyotrophic lateral sclerosis. J. Neurol. Sci. 217, 47–54. Helfand, B.T., Mendez, M.G., Pugh, J., Delsert, C., Goldman, R.D., 2003. A role for intermediate filaments in determining and maintaining the shape of nerve cells. Mol. Biol. Cell 14, 5069–5081. Julien, J.P., Millecamps, S., Kriz, J., 2005. Cytoskeletal defects in amyotrophic lateral sclerosis (motor neuron disease). Novartis Found. Symp. 264, 183–192 (discussion 192–6, 227–30). Kriz, J., Beaulieu, J.M., Julien, J.P., Krnjevic, K., 2005. Up-regulation of peripherin is associated with alterations in synaptic plasticity in CA1 and CA3 regions of hippocampus. Neurobiol. Dis. 18, 409–420. Lansbury, P.T., Lashuel, H.A., 2006. A century-old debate on protein aggregation and neurodegeneration enters the clinic. Nature 443, 774–779. Lecomte, M.J., Basseville, M., Landon, F., Karpov, V., Fauquet, M., 1998. Transcriptional activation of the mouse peripherin gene by leukemia inhibitory factor: involvement of STAT proteins. J. Neurochem. 70, 971–982. Leonard, D.G., Ziff, E.B., Greene, L.A., 1987. Identification and characterization of mRNAs regulated by nerve growth factor in PC12 cells. Mol. Cell. Biol. 7, 3156–3167. Leonard, D.G., Gorham, J.D., Cole, P., Greene, L.A., Ziff, E.B., 1988. A nerve growth factor-regulated messenger RNA encodes a new intermediate filament protein. J. Cell Biol. 106, 181–193. Leung, C.L., He, C.Z., Kaufmann, P., Chin, S.S., Naini, A., Liem, R.K., Mitsumoto, H., Hays, A.P., 2004. A pathogenic peripherin gene mutation in a patient with amyotrophic lateral sclerosis. Brain Pathol. 14, 290–296. Levine, R.L., 2002. Carbonyl modified proteins in cellular regulation, aging, and disease. Free Radic. Biol. Med. 32, 790–796. McLean, J.R., Robertson, J., 2011. Peripherin pathology. In: Nixon, R.A., Yuan, A. (Eds.), Cytoskeleton of the Nervous System, vol. 3. Springer, New York, pp. 201–224. McLean, J., Xiao, S., Miyazaki, K., Robertson, J., 2008. A novel peripherin isoform generated by alternative translation

is required for normal filament network formation. J. Neurochem. 104, 1663–1673. McLean, J., Liu, H.N., Miletic, D., Weng, Y.C., Rogaeva, E., Zinman, L., Kriz, J., Robertson, J., 2010. Distinct biochemical signatures characterize peripherin isoform expression in both traumatic neuronal injury and motor neuron disease. J. Neurochem. 114, 1177–1192. Migheli, A., Pezzulo, T., Attanasio, A., Schiffer, D., 1993. Peripherin immunoreactive structures in amyotrophic lateral sclerosis. Lab. Invest. 68, 185–191. Mizuno, Y., Fujita, Y., Takatama, M., Okamoto, K., 2011. Peripherin partially localizes in Bunina bodies in amyotrophic lateral sclerosis. J. Neurol. Sci. 302, 14–18. Niebroj-Dobosz, I., Dziewulska, D., Kwiecinski, H., 2004. Oxidative damage to proteins in the spinal cord in amyotrophic lateral sclerosis (ALS). Folia Neuropathol. 42, 151–156. Nystrom, T., 2005. Role of oxidative carbonylation in protein quality control and senescence. EMBO J. 24, 1311–1317. Oblinger, M.M., Wong, J., Parysek, L.M., 1989. Axotomy-induced changes in the expression of a type III neuronal intermediate filament gene. J. Neurosci. 9, 3766–3775. Parysek, L.M., Goldman, R.D., 1987. Characterization of intermediate filaments in PC12 cells. J. Neurosci. 7, 781–791. Portier, M.M., de Nechaud, B., Gros, F., 1983. Peripherin, a new member of the intermediate filament protein family. Dev. Neurosci. 6, 335–344. Robberecht, W., 2000. Oxidative stress in amyotrophic lateral sclerosis. J. Neurol. 247 (Suppl. 1), I1–I6. Robertson, J., Doroudchi, M.M., Nguyen, M.D., Durham, H.D., Strong, M.J., Shaw, G., Julien, J.P., Mushynski, W.E., 2003. A neurotoxic peripherin splice variant in a mouse model of ALS. J. Cell Biol. 160, 939–949. Rojo, A.I., Sagarra, M.R., Cuadrado, A., 2008. GSK-3beta down-regulates the transcription factor Nrf2 after oxidant damage: relevance to exposure of neuronal cells to oxidative stress. J. Neurochem. 105, 192–202. Sanelli, T., Xiao, S., Horne, P., Bilbao, J., Zinman, L., Robertson, J., 2007. Evidence that TDP-43 is not the major ubiquitinated target within the pathological inclusions of amyotrophic lateral sclerosis. J. Neuropathol. Exp. Neurol. 66, 1147–1153. Sasaki, S., Maruyama, S., 1993. A fine structural study of Onuf's nucleus in sporadic amyotrophic lateral sclerosis. J. Neurol. Sci. 119, 28–37. Sasaki, S., Maruyama, S., 1994. Immunocytochemical and ultrastructural studies of the motor cortex in amyotrophic lateral sclerosis. Acta Neuropathol. 87, 578–585. Simpson, E.P., Yen, A.A., Appel, S.H., 2003. Oxidative stress: a common denominator in the pathogenesis of amyotrophic lateral sclerosis. Curr. Opin. Rheumatol. 15, 730–736. Sterneck, E., Kaplan, D.R., Johnson, P.F., 1996. Interleukin-6 induces expression of peripherin and cooperates with Trk receptor signaling to promote neuronal differentiation in PC12 cells. J. Neurochem. 67, 1365–1374. Strong, M.J., 2003. The basic aspects of therapeutics in amyotrophic lateral sclerosis. Pharmacol. Ther. 98, 379–414. Swarup, V., Phaneuf, D., Bareil, C., Robertson, J., Rouleau, G.A., Kriz, J., Julien, J.P., 2011. Pathological hallmarks of amyotrophic lateral sclerosis/frontotemporal lobar degeneration in transgenic mice produced with TDP-43 genomic fragments. Brain 134, 2610–2626. Thompson, M.A., Lee, E., Lawe, D., Gizang-Ginsberg, E., Ziff, E.B., 1992. Nerve growth factor-induced derepression of peripherin gene expression is associated with alterations in proteins binding to a negative regulatory element. Mol. Cell. Biol. 12, 2501–2513. Troy, C.M., Muma, N.A., Greene, L.A., Price, D.L., Shelanski, M.L., 1990. Regulation of peripherin and neurofilament expression in regenerating rat motor neurons. Brain Res. 529, 232–238.

BR A I N R ES E A RCH 1 4 22 ( 20 1 1 ) 5 7 –65

Troy, C.M., Greene, L.A., Shelanski, M.L., 1992. Neurite outgrowth in peripherin-depleted PC12 cells. J. Cell Biol. 117, 1085–1092. Wong, N.K., He, B.P., Strong, M.J., 2000. Characterization of neuronal intermediate filament protein expression in cervical spinal motor neurons in sporadic amyotrophic lateral sclerosis (ALS). J. Neuropathol. Exp. Neurol. 59, 972–982.

65

Xiao, S., McLean, J., Robertson, J., 2006. Neuronal intermediate filaments and ALS: a new look at an old question. Biochim. Biophys. Acta 1762, 1001–1012. Xiao, S., Tjostheim, S., Sanelli, T., McLean, J.R., Horne, P., Fan, Y., Ravits, J., Strong, M.J., Robertson, J., 2008. An aggregate-inducing peripherin isoform generated through intron retention is upregulated in amyotrophic lateral sclerosis and associated with disease pathology. J. Neurosci. 28, 1833–1840.