Skip to main content
NCBI home page
Brain Pathology logoLink to Brain Pathology
. 2009 Sep 21;20(3):589–597. doi: 10.1111/j.1750-3639.2009.00337.x

FUS‐Immunoreactive Intranuclear Inclusions in Neurodegenerative Disease

John Woulfe 1,2,✉, Douglas A Gray 1,3, Ian RA Mackenzie 4
PMCID: PMC8094734  PMID: 19832837

Abstract

Neuronal intranuclear inclusions (NIIs) are a histopathological hallmark of several neurodegenerative disorders. However, the role played by NIIs in neurodegenerative pathogenesis remains enigmatic. Defining their molecular composition represents an important step in understanding the pathophysiology of these disorders. Recently, a nuclear protein, “fused‐in‐sarcoma” (FUS) was identified as the pathological protein in two forms of frontotemporal lobar degeneration (FTLD‐IF, formerly known as neuronal intermediate filament inclusion disease, and FTLD‐UPS, formerly known as atypical FTLD‐U), both of which are characterized by the presence of NII. The objective of the present study was to determine the range of neurodegenerative disorders characterized by FUS‐positive NIIs. Immunostaining for FUS revealed intense reactivity of NIIs in FTLD‐IF and FTLD‐UPS as well as in Huntington's disease, spinocerebellar ataxias 1 and 3, and neuronal intranuclear inclusion body disease. In contrast, there was no FUS staining of NIIs in inherited forms of FTLD‐TDP caused by GRN and VCP mutations, fragile‐X‐associated tremor ataxia syndrome, or oculopharyngeal muscular dystrophy. In a cell culture model of Huntington's disease, NIIs were intensely FUS‐positive. NII‐bearing cells displayed loss of the normal diffuse nuclear pattern of FUS staining. This suggests that sequestration of nuclear FUS by NIIs may interfere with its normal nuclear localization.

Keywords: CNS tumors, diagnostic and prognostic markers, glioma, meningioma

INTRODUCTION

Many neurodegenerative disorders are characterized pathologically by the accumulation of abnormal protein aggregates (20). The latter are often present in the form of intracellular “inclusion bodies”. A subset of neurodegenerative disorders are defined by the presence of such inclusion bodies within the nucleus of brain cells [reviewed in (30)]. Neuronal intranuclear inclusions (NIIs) are the hallmark of Huntington's disease (HD) and the other polyglutamine (polyQ) repeat diseases. However, NIIs are also an important pathological finding in other neurodegenerative diseases including those caused by pathogenic polyalanine (polyA) expansions, such as oculopharyngeal muscular dystrophy (OPMD), neuronal intranuclear inclusion body disease (NIIBD), neuroferritinopathy and multiple system atrophy (MSA). Recently, it was demonstrated that ubiquitin‐positive NIIs are a feature of the pathology of the frontotemporal dementias, referred to as frontotemporal lobar degeneration (FTLD) 1, 14, 21, 29. Specifically, NIIs are a characteristic histopathological feature in familial FTLD with tau‐negative, ubiquitin‐positive inclusions (FTLD‐U) caused by mutations in the genes encoding progranulin and valosin containing protein (VCP) (22) and in FTLD‐IF (formerly known as neuronal intermediate filament disease; NIFID) 1, 13.

The role played by NIIs in neurodegeneration remains to be elucidated and identification of the principle protein components is a key step in this process. Recently, abnormally cleaved, ubiquitinated and phosphorylated transactive response element DNA binding protein 43 (TDP‐43) has been identified as the principle protein constituent of the neuronal cytoplasmic inclusion bodies (NCI) in most cases of FTLD‐U (now referred to as FTLD‐TDP) (18) as well as amyotrophic lateral sclerosis (ALS) (18), providing further evidence that FTD and ALS are pathogenetically related diseases and part of a clinicopathological spectrum (11). Indeed, mutations in TDP‐43 have been described in ∼3% of familial and sporadic ALS cases, thereby attesting to its pathogenic significance (23). The formation of TDP‐43‐immunoreactive NIIs is associated with a loss of the normal diffuse nuclear localization of TDP‐43, suggesting that NII formation may sequester nuclear TDP‐43 and abrogate its normal function.

There remains a subset of FTLD in which the principle protein comprising the cellular inclusions remains to be defined. These include FTLD‐IF, basophilic inclusion body disease (BIBD) and cases with neuronal inclusions that are only identified with immunohistochemistry against components of the ubiquitin proteasome system (FTLD‐UPS). The later group includes sporadic cases recently published under the nomenclature of “atypical” FTLD‐U (aFTLD‐U) and familial cases caused by mutations in the CHMP2B gene. aFTLD‐U is characterized clinically by an early age of onset and severe psychobehavioral abnormalities in the absence of significant language or motor deficits. Pathologically, it is defined by the presence of ubiquitin‐positive NCIs and unusual, curvilinear NIIs which do not stain for tau, synuclein, intermediate filaments or TDP‐43. Instead, Neumann and co‐workers recently demonstrated that the NIIs and cytoplasmic inclusions in aFTLD‐U are immunoreactive for the fused in sarcoma (FUS) protein (16). Intriguingly, FUS shares many structural and functional features with TDP‐43. Both are nuclear proteins with RNA and DNA binding properties and a role in RNA splicing. Moreover, mutations in the FUS gene, located on chromosome 16q12, were recently identified as the cause of familial ALS (FALS) type 6 8, 27. In those studies, description of the post‐mortem findings included NCI that were immunoreactive for FUS, but not for TDP‐43 (27). In addition to aFTLD‐U, both FTLD‐IF (17) and BIBD (David G. Munoz, pers. comm.) are also characterized by FUS‐immunoreactive inclusions. Thus, there appears to be emerging evidence for the existence of a dichotomous biochemistry of NIIs in neurodegenerative diseases based on whether they label for TDP‐43 or FUS.

By analogy with TDP‐43, it is tempting to speculate that the sequestration of FUS into NIIs in aFTLD‐U, FTLD‐IF and BIBD may have a role in neurodegeneration. Based on a previous report of FUS‐positive NIIs in HD (3), we were interested to determine whether FUS may be a molecular component of NIIs in other TDP‐43‐negative neurodegenerative disorders. Thus, we immunostained brain sections of patients with a variety of neurodegenerative diseases characterized by the presence of NIIs as well as a mouse model of spinocerebellar ataxia (SCA)1. In order to more closely investigate the consequences of NII formation on the physiological localization of FUS, we studied FUS immunostaining in a cell culture model of HD.

MATERIALS AND METHODS

Immunohistochemistry of human brain

The demographic data of the patients examined in this study, as well as the areas of the brain examined in each case, are shown in Table 1. The listed diagnoses were rendered based on established consensus diagnostic criteria. For HD, SCA1 and SCA3 cases, pathogenic CAG repeat expansions were confirmed in the huntingtin, ataxin 1 and ataxin 3 genes, respectively. VCP and GRN mutations were confirmed in the FTLD‐TDP cases. GRN and FUS mutations were excluded in one aFTLD‐U case and in one FTLD‐IF case.

Table 1.

Patient demographics and NII immunostaining properties. Abbreviations: AbN = abnormal; BG = basal ganglia; ctx = cortex; DRPLA = dentatorubropallidoluysian atrophy; F = female; fvFTD = frontal variant frontotemporal dementia; FTLD‐U = frontotemporal lobar degeneration, ubiquitin‐positive; FUS = fused in sarcomal; HC = hippocampus; HD = Huntington's disease; M = male; MND = motor neuron disease; MSA = multiple system atrophy; NIIBD = neuronal intranuclear inclusion body disease; OPMD = oculopharyngeal muscular dystrophy; Polyq = polyglutamine; SCA = spinocerebellar ataxia; SBMA = spinal and bulbar muscular atrophy; VCP = valosin valosin containing protein.

Diagnosis Sex Onset Age at tissue diagnosis Clinical diagnosis Genetics Section Immunostain
p62 ubiquitin FUS TDP‐43 syn Polyq
aFTLD‐U
 aFTLDU F 38 49 fvFTD, childhood epilepsy GRN, FUS‐none found HC, frontal ctx No Yes Yes No Nd No
 aFTLDU F 37 45 fvFTD, depression None HC, frontal ctx No Yes Yes No Nd No
 aFTLDU F 40 51 fvFTD, poor memory, falls None HC, frontal ctx No Yes Yes No Nd No
 aFTLDU F 32 39 fvFTD None HC, frontal ctx No Yes Yes No Nd No
 aFTLDU M 29 36 fvFTD None HC, frontal ctx No Yes Yes No Nd No
 aFTLDU F 36 42 fvFTD, chorea None HC, frontal ctx No Yes Yes No Nd No
FTLD‐IF
 NIFID F 25 29 FTD, MND None HC, frontal ctx No Yes Yes No Nd No
 NIFID F 34 41 FTD, MND FUS‐none found HC, frontal ctx No Yes Yes No Nd No
 NIFID M 58 61 FTD None HC, frontal ctx No Yes Yes No Nd No
FTLD‐TDP
  GRN M 57 61 FTD GRN Frontal ctx Yes Yes No Yes Nd No
  GRN F 69 77 FTD GRN Frontal ctx Yes Yes No Yes Nd No
  VCP F ? 56 IBMPFD VCP Frontal ctx Yes Yes No Yes Nd No
  VCP F ? 60 IBMPFD VCP Frontal ctx Yes Yes No Yes Nd No
Polyglutamine
 HD F 28 43 HD Huntingtin BG Yes Yes Yes No Nd Yes
 HD M ? 49 HD Huntingtin BG Yes Yes Yes No Nd Yes
 SCA‐3 M ? 58 Machado‐Joseph's Ataxin 3 Pons Yes Yes Yes No Nd Yes
 SCA‐3 M ? 50 Machado‐Joseph's Ataxin 3 Pons Yes Yes Yes No Nd Yes
 SCA‐1 M 32 59 SCA, ophthalmoplegia, neuropathy Ataxin 1 Pons Yes Yes Yes No Nd Yes
Other
 NIIBD M ? 58 Tremor, ataxia, gastrointestinal dysfunction, dementia SBMA, DRPLA, FXTAS‐none found HC Yes Yes Yes No Nd Some
 NIIBD M 68 78 Dementia, aphasia, dysphagia, incontinence, gait abN None HC Yes Yes Yes No Nd Some
 MSA M ? 72 Ataxia None Pons No Some No No Yes No
 FXTAS M 65 78 FXTAS Pre‐mutation in fragile X Midbrain Yes Yes No No Nd No
 OPMD F ? 54 Oculopharyngeal muscular dystrophy Polya repeat expansion in PABPN1 Deltoid muscle Yes Yes No No Nd No
Open in a new tab

All immunohistochemistry was performed on 5‐µm thick sections of formalin fixed, paraffin embedded tissue using the Ventana BenchMark® XT automated staining system (Ventana, Tuscon, AZ, USA) and developed with aminoethylcarbizole. The primary antibodies employed recognized FUS (Sigma‐Aldrich anti‐FUS, St. Louis, MO, USA; 1:25 with initial overnight incubation at room temperature, following microwave antigen retrieval), ubiquitin (DAKO anti‐ubiquitin, Carpinteria, CA, USA; 1:500, following microwave antigen retrieval), p62 (BD Transduction Laboratories p62 Lck ligand, Mississauga, ON, Canada; 1:500 following microwave antigen retrieval), TDP‐43 (ProteinTech Group anti‐TARDBP, Chicago, IL, USA; 1:1000 following microwave antigen retrieval), expanded polyQ repeat regions (Chemicon 1C2, Temecula, CA, USA; 1:1000, 24 h at room temperature following formic acid pre‐treatment) and alpha‐synuclein (Intermedico, Markham, ON, Canada; 1:50, 30 minute at room temperature following antigen retrieval in de‐cloaker at 15 psi for 5 minutes).

Immunohistochemistry of mouse brain

Cerebella were excised from B05 transgenic mice carrying a mutant Atx‐1 allele with 82 CAG repeats (gift from Dr H. Orr, University of Minnesota). Cerebella were fixed in 10% phosphate‐buffered formalin overnight and paraffin embedded by the Morphology Unit of the Department of Pathology and Laboratory Medicine of the University of Ottawa. Tissues were sectioned using a microtome at a thickness of 5 µm. The sections were deparaffinized and then heated in 10‐mM sodium citrate buffer (pH 7.6) in microwave oven for 10 minutes to unmask the antigen. Endogenous peroxide activity was blocked by incubation in methanol containing 3% hydrogen peroxide for 20 minutes. The sections were incubated for 30 minutes with 1.5% normal goat serum in 0.1M PBS (pH 7.4) to block nonspecific binding. They were then incubated overnight at 4°C with rabbit anti‐FUS (Bethyl Laboratories, Montgomery, TX, USA) at 1:500 diluted in blocking solution or the 1C2 antibody (Chemicon) at 1:1000. After being washed with PBS, the sections were exposed for 30 minutes to biotinylated goat anti‐rabbit or horse anti‐mouse antibodies diluted 1:200 in blocking solution. The reaction product was visualized by the avidin biotin peroxidase method with an ABC Elite kit (Vector Labs, Burlingame, CA, USA) and DAB substrate from KPL Inc. (Gaithersburg, MD, USA). Cell nuclei were counterstained with hematoxylin. Digital images of stained sections were obtained using a ScanScope instrument (Aperio Technologies Inc., Vista, CA, USA).

Cell culture, transfection and immunofluorescence analysis

Human U87MG cells (a gift from Dr. W. Cavenee, Ludwig Institute for Cancer Research, La Jolla, CA, USA) were cultured on glass coverslips at 37°C and 5% CO2 in Dulbecco's modified Eagle's medium (Hyclone DMEM/High Glucose, Thermo Scientific, Logan, UT, USA) supplemented with 10% (v/v) of a 2:1 mixture of donor bovine serum and fetal bovine serum. Cells were transfected with plasmids containing exon 1 of the mutant human huntingtin gene fused to the yellow fluorescent reporter protein (gift from Dr. R. Kopito, Stanford University) using GeneJuice Transfection Reagent (Novagen, Madison, WI, USA) as per the supplier's protocol. At 48‐h post‐transfection, the cells were fixed in 4% paraformaldehyde (pH 7.2∼7.4) for 20 minutes. The cells were then incubated in 5% normal goat serum containing 0.2% Triton X‐100 in 0.1M PBS (pH 7.4) for 30 minutes to block nonspecific binding and permeabilize after washing with 0.1M PBS (pH 7.4). They were then incubated overnight at 4°C with an anti‐polyQ at a 1:4000 dilution (1C2 monoclonal antibody, Chemicon) or a rabbit anti‐FUS antibody (Bethyl Laboratories) at 1:500 diluted in the blocking solution. Binding of the 1C2 primary antibody was detected with a goat anti‐mouse secondary antibody conjugated to FITC (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA). The FUS‐specific antibody was detected using a goat anti‐rabbit secondary conjugated to Cy3 (Jackson ImmunoResearch Laboratories, Inc.). Nuclei were stained by mounting coverslips in Vectashield solution (Vector Labs) containing DAPI (4′,6 diamidino‐2‐phenylindole). Cell images were obtained using a Zeiss Axioskop microscope and Zeiss Axovision software (Gottingen, Germany).

RESULTS

FUS immunostaining of human brain

PolyQ disorders

FUS immunostaining of NIIs in polyQ repeat diseases is shown in Figure 1. In sections of basal ganglia from HD patients, round, oblong or lentiform NIIs as well as rare, smaller cytoplasmic inclusions in the caudate nucleus, putamen and adjacent insular cortex were immunoreactive for ubiquitin, p62 and polyQ but negative for TDP‐43. NIIs displayed intense immunostaining for FUS (Figure 1A). Rare NCIs were also FUS‐positive. Qualitatively, many more NIIs were identified in FUS‐immunostained sections than in those stained for ubiquitin or p62. In some nuclei with NIIs, diffuse nuclear FUS staining appeared to be preserved whereas in others, FUS immunostaining was confined to the NII. In many cases with preserved diffuse nuclear staining, visualization of NIIs was obscured. To enhance the visualization and demonstration of NIIs in Figure 1 and Figure 2A,B (see next), reduced concentrations of anti‐FUS primary antibody were employed.

Figure 1.

Figure 1

Open in a new tab

FUS‐immunoreactive NIIs in: (A) striatal neurons in HD, (B) neurons of the pontine nuclei in SCA1 and (C) SCA3, and (D) CA3 hippocampal neurons in NIIBD. Scale bars = 20 microns.

Figure 2.

Figure 2

Open in a new tab

A. FUS‐immunoreactive curvilinear (left) and vermiform (right) NIIs in dentate granule neurons in aFTLD‐U. B. FUS‐positive curvilinear (arrows) and annular (arrowhead) NIIs as well as cytoplasmic inclusions in dentate granule neurons in FTLD‐IF. C, D. p62‐immunopositive NIIs (arrows) and short dystrophic neurites in frontal cortex of FTLD‐TDP with GRN and VCP mutations, respectively. These p62‐positive inclusions were completely FUS‐negative. Scale bars = 10 microns in A, 30 microns in B, 20 microns in C and D.

In sections of pons from SCA1 and SCA3 patients, scattered neurons in the pontine nuclei displayed solitary, round NIIs positive for ubiquitin, p62 and polyQ. In SCA3, rare neurons displayed multiple dot‐like cytoplasmic inclusions as well. NIIs were also identified in the locus coeruleus in SCA3. All of these inclusions showed intense staining for FUS (Figure 1B,C). As in HD, some NII‐bearing neurons displayed a normal background pattern of diffuse nuclear FUS staining whereas in others this was lost. By further analogy with HD, FUS immunohistochemistry was of superior sensitivity relative to ubiquitin or p62 in detecting the NIIs in SCA1 and SCA3 cases.

NIIBD

Sections of mesial temporal lobe from NIIBD cases were characterized by the presence of round or oblong, solitary NIIs measuring 4–10 microns in diameter. These were concentrated in the CA2 and CA3 sectors of the hippocampus but rare NIIs were also identified in the CA1 sector, entorhinal cortex and adjacent temporal neocortex. The NIIs were visible on HE stain and immunostained for ubiquitin, p62, and were negative for TDP‐43 and polyQ. FUS immunostaining revealed intense immunoreactivity of the NIIs (Figure 1D). In the vast majority of hippocampal neurons, with or without NIIs, there was no, or only very light, diffuse nuclear FUS staining.

FTLD

As reported previously (12), cases of aFTLD‐U were characterized by the presence of ubiquitin‐immunoreactive small, round, cytoplasmic inclusions in neocortical and hippocampal neurons as well as unusual curvilinear or vermiform NIIs in cortical and dentate granule cells. These inclusions were negative for tau, alpha‐synuclein, neurofilament protein, polyQ and TDP‐43. Curiously, they were also negative for p62 sequestosome. We confirmed the previously reported immunostaining of these unusual NIIs in aFTLD‐U for FUS (Figure 2A) (16). Cytoplasmic inclusions were also FUS‐positive. In neurons lacking inclusions, nuclei showed intense, diffuse immunostaining for FUS. Notably, FUS staining of NIIs was not associated with an appreciable decrease in this diffuse staining in the remainder of the nucleus.

We also confirmed the previous report of FUS‐immunoreactivity of NIIs in FTLD‐IF (Figure 2B) (17). Two morphologically distinct types of NII were identified; curvilinear or vermiform NII reminiscent of those seen in aFTLD‐U, and large round NII. As in aFTLD‐U, the vermiform NII failed to stain for p62 sequestosome. The vermiform NIIs showed consistent FUS staining while only a minority of the round NIIs were FUS‐positive (best seen with immunofluorescence). As in aFTLD‐U, nuclei with FUS‐positive NIIs did not show an appreciable diminution in diffuse nuclear FUS staining.

Sections of frontal neocortex from familial FTLD‐TDP cases associated with mutations in GRN showed NCI and dystrophic neurites predominantly in layer 2 as well as scattered round or lentiform NIIs, immunoreactive for ubiquitin, p62 (Figure 2C), and TDP‐43. Those from familial cases associated with VCP mutations showed ubiquitin‐, p62‐, and TDP‐43‐immunoreactive short neurites in layer 2 and numerous round, lentiform or rod‐shaped NIIs (Figure 2D). In both cases, inclusion‐bearing neurons exhibited a loss of the normal diffuse nuclear pattern of TDP‐43 staining. NCIs and NIIs were completely negative for FUS and polyQ.

Other NII diseases

In the basis pontis of the MSA case, there were abundant ubiquitin‐ and synuclein‐immunoreactive oligodendroglial cytoplasmic inclusions and NCIs. In addition, there were numerous synuclein‐immunostained NIIs in the neurons of the pontine nuclei (Figure 3A). These were complex, curvilinear, “web‐like” structures that are rarely stained for ubiquitin and never stained for p62. In addition, rod‐shaped oligodendroglial nuclear inclusions were also identified which displayed a similar immunophenotype. None of these inclusion types showed immunostaining for TDP‐43, polyQ or FUS. The latter showed the normal diffuse pattern of staining within the nuclei of pontine neurons (Figure 3B).

Figure 3.

Figure 3

Open in a new tab

A. Synuclein‐immunoreactive “spiderweb‐like” NII (arrow) in a pontine neuron in MSA. B. FUS‐immunostaining of the same section shows the normal diffuse nuclear pattern of staining in pontine neurons with no staining of NIIs. C. p62 immunopositive glial intranuclear inclusions in mesencephalic tegmentum in FXTAS. These inclusions were FUS‐negative. D. Diffuse nuclear FUS staining in myonuclei in OPMD. Scale bars = 20 microns.

In sections of midbrain from FXTAS cases, ubiquitin and p62 immunostaining revealed intensely stained round or linear inclusions within the nuclei of glial cells in the mesencephalic tegmentum and in the cerebral peduncle (Figure 3C). These inclusions were negative for TDP‐43, polyQ and FUS. Notably, ubiquitin‐ and p62‐immunostained Marinesco bodies were identified in pigmented neurons of the substantia nigra and these were also FUS‐negative.

Finally, skeletal muscle from a patient with oculopharyngeal muscular dystrophy showed rare, scattered ubiquitin‐ and p62‐positive inclusions within myonuclei. FUS immunostaining revealed intense, diffuse staining of myonuclei with no detectable staining of the inclusions (Figure 3D).

FUS immunostaining of mouse brain

B05 transgenic mice carry a mutant ataxin‐1 allele with 82 CAG repeats. We have characterized the pathological and behavioral features of these mice (26). They represent a valid model of human SCA1. In sections of cerebellum from these mice, the vast majority of Purkinje cells contained predominantly round, solitary NIIs which were immunoreactive for ubiquitin and polyQ as well as for FUS (Figure 4).

Figure 4.

Figure 4

Open in a new tab

Cerebellar sections from the B05 transgenic mouse model of SCA1. A. Control section in which only secondary antibody was used. B. PolyQ‐immunopositive NIIs were detected in the majority of Purkinje cells (arrows). C. Immunoreactivity of Purkinje cell NIIs for FUS (arrows). Scale bar in A = 50 microns.

FUS immunostaining of huntingtin inclusions in vitro

We have employed transfection of human glioblastoma (U87MG) cells with plasmids containing exon 1 of the mutant human huntingtin gene fused to the yellow fluorescent reporter protein as an in vitro tool to study the dynamic aspects of NII formation in HD. Control cells were transfected with an empty vector. The transfection efficiency in our system was approximately 40%. FUS immunostaining revealed diffuse nuclear immunoreactivity in all control cells, but only in inclusion‐negative cells in cells transfected with mutant huntingtin. Roughly half of all cells transfected with human mutant huntingtin developed NIIs or, more commonly, perinuclear inclusions by 48‐h post‐transfection. These were predominantly solitary, spherical and immunostained for ubiquitin and polyQ (Figure 5). All perinuclear and intranuclear inclusions also showed intense immunostaing for FUS (Figure 5). In many cases, FUS immunostaining appeared to be concentrated at the periphery of inclusions. Importantly, all inclusion‐bearing cells displayed either a complete absence of diffuse nuclear background staining or a marked diminution relative to cells without inclusions (Figure 5).

Figure 5.

Figure 5

Open in a new tab

Representative image of human U87MG cells cultured on coverslips and transfected with a plasmid encoding the first exon of the human huntingtin gene containing a polyQ tract of 103 residues. A. Phase contrast image. B. Image of merged color channels including DAPI‐stained nuclei (blue) FUS (red) and polyQ (green). The bright yellow dot is an NII immunoreactive for polyQ and FUS (arrow). C. DAPI nuclear counterstain. D. NII detected with a polyQ‐specific antibody. E. FUS protein. Note the sequestration of nuclear FUS protein into the inclusion (evident by comparing nuclear intensity of FUS in the inclusion‐containing cells vs. untransfected adjacent cells). The intense brightness of the inclusion is evident from the exposure time utilized in obtaining the image in panel D (6 ms) as compared with the DAPI image (panel C, 8 ms) and FUS image (panel E, 500 ms). Scale bar in A = 10 microns.

DISCUSSION

The classification of neurodegenerative diseases is predicated on elucidation of the pathogenic protein species that define each disease or family of diseases. For FTLD, there has been considerable progress in recent years in constructing this protein‐based classification (1). Thus, most FTLDs can now be classified as tau‐opathies (FTLD‐tau), TDP‐43 proteinopathies (FTLD‐TDP) and neuronal intermediate filament‐opathies (FTLD‐IF). The recent demonstration that the tau‐, synuclein‐, neuronal intermediate filament‐, and TDP‐43‐negative inclusions in aFTLD‐U are positive for FUS (16) potentially adds a novel diagnostic category, FTLD‐FUS, to this classification scheme. One theme that is becoming evident as this classification emerges is that in tau‐negative cases of FTLD‐U, there appears to be a mutually exclusive reciprocal staining pattern of inclusions for either TDP‐43 or for FUS. The mechanisms underlying this dichotomy remain to be defined. However, the fact that cellular abnormalities related to TDP‐43 or FUS can result in biochemically distinct, but clinically and morphologically similar disorders is not surprising in light of the striking structural and functional homologies between FUS and TDP‐43 (9). The results of the present study confirm this reciprocal relationship with respect to FUS and TDP‐43 staining in neurodegenerative disorders in which NIIs are a histopathological hallmark. Strong FUS staining of NIIs was demonstrated in the polyQ repeat disorders HD, SCA1 and SCA3, as well as in a mouse model of SCA1 and in cells expressing mutant huntingtin, suggesting that FUS is a component of polyQ NIIs. Confirmation of this will require examination of the entire range of polyQ repeat disorders. NIIs in OPMD, a polyA repeat disorder, as well as the synuclein‐positive NIIs in MSA, were all FUS‐negative. In addition, FUS failed to stain NIIs in FXTAS. These latter NIIs have been shown to contain, among other protein constituents, pathologically expanded RNA species (5). This indicates that FUS, an RNA binding protein, is not being recruited to NIIs secondarily by virtue of its RNA‐binding properties.

In the polyQ repeat disorders, NIIs are comprised of the abnormal polyQ‐expanded protein that may be truncated (as in the case of huntingtin in HD). The NIIs contain a host of additional proteins including SUMO1, heat shock proteins, components of the ubiquitin‐proteasome system and a variety of transcription factors [see 28, 30 for review]. Our demonstration of FUS in NIIs in polyQ repeat disorders adds to this growing list and extends previous evidence by Doi et al (3) for the presence of FUS in HD NIIs (3). These authors demonstrated that FUS could not bind monomeric expanded huntingtin but instead that the recruitment of FUS occurred at an “early stage” of aggregate formation. In electron microscopic studies in vitro, they demonstrated that polyQ huntingtin formed amorphous aggregates before becoming mature amyloid fibrils and showed that FUS bound to the former. They suggested that FUS bound to some conformation of exposed polyQ. Such a conformational affinity of FUS for pathological polyQ aggregates and not for polyQ‐expanded huntingtin per se would be consistent with our evidence that FUS may a constituent of all polyQ NIIs. Our demonstration of FUS staining of NIIs in the B05 mouse model of SCA1, and in our in vitro model of HD NII formation, indicates that it will be possible to study the role of FUS in NII formation in animal and in vitro models of NII diseases.

The presence of FUS in polyQ NIIs may have relevance with respect to FUS staining of NIIs in NIIBD. NIIBD is characterized by the presence of widespread ubiquitinated NIIs [reviewed in (30)]. Like polyQ NIIs, they have been shown to contain a variety of proteins including proteasome subunits, heat shock proteins, SUMO1, transcription factors and glucocorticoid receptor (15). Interestingly, they also contain proteins involved in membrane trafficking, including NSF, dynamin 1 and Unc‐18‐1 (19). Importantly, studies have reported immunostaining of at least a proportion of NIIs using the monoclonal anti‐polyQ antibody 1C2 not only in NIIBD 10, 24, 25, but also in FTLD‐IF (6) (although this was not our experience). While the possibility that FUS staining of NIIs in NIIBD and FTLD‐IF reflects the presence of polyQ is conceivable, it is unlikely that these are primarily polyQ disorders given their largely sporadic nature. Instead, native or post‐translationally modified FUS probably represents the primary protein constituent of NIIs in FTLD‐IF, aFTLD‐U and possibly NIIBD, whereas it is a secondary molecular component of NIIs in the polyQ disorders.

The results of the present study may have implications for understanding the role of NIIs in neurodegenerative pathogenesis. Whether inclusion formation is cytotoxic, cytoprotective or incidental has been hotly debated (20). A cytotoxic role for NIIs in FTLD has been postulated based on the loss of the normal diffuse nuclear pattern of TDP‐43 immunoreactivity in inclusion‐bearing cells, providing morphological evidence for a loss of normal nuclear TDP‐43 function through sequestration in NIIs (2). Given the structural and functional homologies between TDP‐43 and FUS, it is tempting to postulate that the sequestration of FUS in NIIs demonstrated in the present study reflects a relative loss of FUS function with detrimental consequences for the cell. In this context, it is interesting that FUS‐null mouse hippocampal neurons display abnormal spine morphology mimicking those described in human HD (4) and mouse models of HD (7). Doi and co‐workers provided biochemical evidence for a decrease in soluble FUS protein in a cell model of HD in which FUS accumulated in inclusion bodies (3). Moreover, there is evidence for a relative shift of FUS from the nucleus to the cytoplasm in familial ALS associated with FUS mutations 8, 27. Our immunohistochemical studies of human tissue revealed a loss of normal physiological nuclear FUS staining in some, but not all, inclusion‐bearing cells. In contrast, our in vitro studies provided strong evidence for a negative correlation between NII formation and diffuse nuclear FUS staining. Mislocalization of FUS was detected in 100% of inclusion‐bearing cells. The factors underlying the discrepancy between our human in vivo and in vitro results remain to be elucidated but may relate to technical factors including the temporal dynamics of inclusion formation. Whereas the time course of NII formation in vivo remains to be defined, inclusion formation in our cell culture model is extremely rapid. If inclusion formation is slower in vivo, it is more likely that some inclusion‐bearing cells will be caught “in a state of transition” wherein nuclear FUS has not yet been completely sequestered within the NII. Altogether, it is tempting to render an analogy with TDP‐43 and speculate that the sequestration of FUS into NIIs results in mislocalization and loss of function of nuclear FUS. To address this, it will be interesting in future mechanistic studies to determine whether FUS overexpression can rescue the neurodegenerative phenotype in FUS‐positive diseases.

In summary, the results of the present study contribute substantially to our ongoing efforts to construct a protein‐based classification of neurodegenerative diseases. By analogy with TDP‐43 in FTLD‐TDP, it is tempting to speculate that FUS is the major protein comprising the unique NIIs in aFTLD‐U and FTLD‐IF. In contrast, its presence in polyQ and NIIBD NIIs is probably secondary to intranuclear aggregation of another protein. However, the demonstration that FUS is a component of NIIs in the polyQ neurodegenerative diseases as well as NIIBD may have important implications for understanding the cellular pathogenesis of these disorders and the role of NIIs therein. Our in vitro results suggest that FUS incorporation into NIIs may abrogate its normal intranuclear function. Future studies should address this possibility.

ACKNOWLEDGMENTS

The authors wish to acknowledge the excellent technical assistance of Mei Zhang and Margaret Luk. This study was supported by Canadian Institutes of Health Research grants #57737 awarded to DAG and JW and #74580 awarded to IM; Physicians Services Incorporated Foundation grant 09‐01 awarded to JW; and the Pacific Alzheimer Research Foundation (IM). The authors would also like to thank Dr Marc DelBigio who supplied the tissue from the FXTAS case.

REFERENCES

  • 1. Cairns NJ, Bigio EH, Mackenzie IR, Neumann M, Lee VM, Hatanpaa KJ et al (2007) Neuropathologic diagnostic and nosologic criteria for frontotemporal lobar degeneration: consensus of the consortium for frontotemporal lobar degeneration. Acta Neuropathol 114:5–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Davidson Y, Kelley T, Mackenzie IR, Pickering‐Brown S, Du Plessis D, Neary D et al (2007) Ubiquitinated pathological lesions in frontotemporal lobar degeneration contain the TAR DNA‐binding protein, TDP‐43. Acta Neuropathol 113:521–533. [DOI] [PubMed] [Google Scholar]
  • 3. Doi H, Okamura K, Bauer PO, Furukawa Y, Shimizu H, Kurosawa M et al (2008) RNA‐binding protein TLS is a major nuclear aggregate‐interacting protein in huntingtin exon 1 with expanded polyglutamine‐expressing cells. J Biol Chem 283:6489–6500. [DOI] [PubMed] [Google Scholar]
  • 4. Ferrante RJ, Kowall NW, Richardson EP Jr (1991) Proliferative and degenerative changes in striatal spiny neurons in Huntington's disease: a combined study using the section‐Golgi method and calbindin D28k immunocytochemistry. J Neurosci 11:3877–3887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Iwahashi CK, Yasui DH, An HJ, Greco CM, Tassone F, Nannen K et al (2006) Protein composition of the intranuclear inclusions of FXTAS. Brain 129:256–271. [DOI] [PubMed] [Google Scholar]
  • 6. Josephs KA, Holton JL, Rossor MN, Braendgaard H, Ozawa T, Fox NC et al (2003) Neurofilament inclusion body disease: a new proteinopathy? Brain 126:2291–2303. [DOI] [PubMed] [Google Scholar]
  • 7. Klapstein GJ, Fisher RS, Zanjani H, Cepeda C, Jokel ES, Chesselet MF, Levine MS (2001) Electrophysiological and morphological changes in striatal spiny neurons in R6/2 Huntington's disease transgenic mice. J Neurophysiol 86:2667–2677. [DOI] [PubMed] [Google Scholar]
  • 8. Kwiatkowski TJ Jr, Bosco DA, Leclerc AL, Tamrazian E, Vanderburg CR, Russ C et al (2009) Mutations in the FUS/TLS gene on chromosome 16 cause familial amyotrophic lateral sclerosis. Science 323:1205–1208. [DOI] [PubMed] [Google Scholar]
  • 9. Lagier‐Tourenne C, Cleveland DW (2009) Rethinking ALS: the FUS about TDP‐43. Cell 136:1001–1004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Lieberman AP, Robitaille Y, Trojanowski JQ, Dickson DW, Fischbeck KH (1998) Polyglutamine‐containing aggregates in neuronal intranuclear inclusion disease. Lancet 351:884. [DOI] [PubMed] [Google Scholar]
  • 11. Mackenzie IR, Feldman HH (2005) Ubiquitin immunohistochemistry suggests classic motor neuron disease, motor neuron disease with dementia, and frontotemporal dementia of the motor neuron disease type represent a clinicopathologic spectrum. J Neuropathol Exp Neurol 64:730–739. [DOI] [PubMed] [Google Scholar]
  • 12. Mackenzie IR, Foti D, Woulfe J, Hurwitz TA (2008) Atypical frontotemporal lobar degeneration with ubiquitin‐positive, TDP‐43‐negative neuronal inclusions. Brain 131:1282–1293. [DOI] [PubMed] [Google Scholar]
  • 13. Mackenzie IR, Neumann M, Bigio EH, Cairns NJ, Alafuzoff I, Kril J et al (2009) Nomenclature for neuropathologic subtypes of frontotemporal lobar degeneration: consensus recommendations. Acta Neuropathol 117:15–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Mann DM, South PW, Snowden JS, Neary D (1993) Dementia of frontal lobe type: neuropathology and immunohistochemistry. J Neurol Neurosurg Psychiatry 56:605–614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. McFadden K, Hamilton RL, Insalaco SJ, Lavine L, Al‐Mateen M, Wang G, Wiley CA (2005) Neuronal intranuclear inclusion disease without polyglutamine inclusions in a child. J Neuropathol Exp Neurol 64:545–552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Neumann M, Rademakers R, Roeber S, Baker M, Kretzschmar HA, Mackenzie IRA (2009) Frontotemporal lobar degeneration with FUS pathology. Brain (in press). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Neumann M, Roeber S, Kretzschmar HA, Rademakers R, Baker M, Mackenzie IRA (2009) Abundant FUS‐immunoreactive pathology in neuronal intermediate filament inclusion disease. Acta Neuropathol (in press). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, Chou TT et al (2006) Ubiquitinated TDP‐43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science 314:130–133. [DOI] [PubMed] [Google Scholar]
  • 19. Pountney DL, Raftery MJ, Chegini F, Blumbergs PC, Gai WP (2008) NSF, Unc‐18‐1, dynamin‐1 and HSP90 are inclusion body components in neuronal intranuclear inclusion disease identified by anti‐SUMO‐1‐immunocapture. Acta Neuropathol 116:603–614. [DOI] [PubMed] [Google Scholar]
  • 20. Ross CA, Poirier MA (2005) Opinion: What is the role of protein aggregation in neurodegeneration? Nat Rev Mol Cell Biol 6:891–898. [DOI] [PubMed] [Google Scholar]
  • 21. Rosso SM, Kamphorst W, De Graaf B, Willemsen R, Ravid R, Niermeijer MF et al (2001) Familial frontotemporal dementia with ubiquitin‐positive inclusions is linked to chromosome 17q21‐22. Brain 124:1948–1957. [DOI] [PubMed] [Google Scholar]
  • 22. Sampathu DM, Neumann M, Kwong LK, Chou TT, Micsenyi M, Truax A et al (2006) Pathological heterogeneity of frontotemporal lobar degeneration with ubiquitin‐positive inclusions delineated by ubiquitin immunohistochemistry and novel monoclonal antibodies. Am J Pathol 169:1343–1352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Sreedharan J, Blair IP, Tripathi VB, Hu X, Vance C, Rogelj B et al (2008) TDP‐43 mutations in familial and sporadic amyotrophic lateral sclerosis. Science 319:1668–1672. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Takahashi J, Fukuda T, Tanaka J, Minamitani M, Fujigasaki H, Uchihara T (2000) Neuronal intranuclear hyaline inclusion disease with polyglutamine‐immunoreactive inclusions. Acta Neuropathol 99:589–594. [DOI] [PubMed] [Google Scholar]
  • 25. Takahashi J, Tanaka J, Arai K, Funata N, Hattori T, Fukuda T et al (2001) Recruitment of nonexpanded polyglutamine proteins to intranuclear aggregates in neuronal intranuclear hyaline inclusion disease. J Neuropathol Exp Neurol 60:369–376. [DOI] [PubMed] [Google Scholar]
  • 26. Tsirigotis M, Tang MY, Beyers M, Zhang M, Woulfe J, Gray DA (2006) Delayed spinocerebellar ataxia in transgenic mice expressing mutant ubiquitin. Neuropathol Appl Neurobiol 32:26–39. [DOI] [PubMed] [Google Scholar]
  • 27. Vance C, Rogelj B, Hortobagyi T, De Vos KJ, Nishimura AL, Sreedharan J et al (2009) Mutations in FUS, an RNA processing protein, cause familial amyotrophic lateral sclerosis type 6. Science 323:1208–1211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Woulfe J (2008) Nuclear bodies in neurodegenerative disease. Biochim Biophys Acta 1783:2195–2206. [DOI] [PubMed] [Google Scholar]
  • 29. Woulfe J, Kertesz A, Munoz DG (2001) Frontotemporal dementia with ubiquitinated cytoplasmic and intranuclear inclusions. Acta Neuropathol 102:94–102. [DOI] [PubMed] [Google Scholar]
  • 30. Woulfe JM (2007) Abnormalities of the nucleus and nuclear inclusions in neurodegenerative disease: a work in progress. Neuropathol Appl Neurobiol 33:2–42. [DOI] [PubMed] [Google Scholar]

Articles from Brain Pathology are provided here courtesy of Wiley

RESOURCES