Evidence of Multisystem Disorder in Whole-Brain Map of Pathological TDP-43 in Amyotrophic Lateral Sclerosis

Background  Pathological 43-kDa transactivating responsive sequence DNA-binding protein (TDP-43) has been identified recently as the major disease protein in amyotrophic lateral sclerosis (ALS), and in frontotemporal lobar degeneration with ubiquitinated inclusions, with or without motor neuron disease, but the distribution of TDP-43 pathology in ALS may be more widespread than previously described.

Objective  To determine the extent of TDP-43 pathology in the central nervous systems of patients with clinically confirmed and autopsy confirmed diagnoses of ALS.

Design  Performance of an immunohistochemical whole–central nervous system scan for evidence of pathological TDP-43 in ALS patients.

Setting  An academic medical center.

Participants  We included 31 patients with clinically and pathologically confirmed ALS and 8 control participants.

Main Outcome Measures  Immunohistochemistry and double-labeling immunofluorescence to assess the frequency and severity of TDP-43 pathology.

Results  In addition to the stereotypical involvement of upper and lower motor neurons, neuronal and glial TDP-43 pathology was present in multiple areas of the central nervous systems of ALS patients, including in the nigro-striatal system, the neocortical and allocortical areas, and the cerebellum, but not in those of the controls.

Conclusions  These findings suggest that ALS does not selectively affect only the pyramidal motor system, but rather is a multisystem neurodegenerative TDP-43 proteinopathy.

Pathological 43-kDa transactivating responsive sequence DNA-binding protein (TDP-43) has been identified as the major disease protein in amyotrophic lateral sclerosis (ALS), and in frontotemporal lobar degeneration (FTLD) with ubiquitin-positive inclusions, with or without motor neuron disease. Indeed, TDP-43 inclusions in the central nervous system (CNS) appear to be diagnostic of these disorders, and may underlie some of their mechanisms of neurodegeneration.^1-3 Although the functions of TDP-43 are poorly understood, TDP-43 is a ubiquitously expressed, highly conserved nuclear protein that appears to be involved in transcriptional repression and alternative splicing, and may act as a scaffold for nuclear bodies through interactions with survival motor neuron proteins.¹ Amyotrophic lateral sclerosis is historically considered to be the prototypical pyramidal motor system neurodegenerative disease, and the presence of additional features has usually been considered to negate the diagnosis of classic ALS, therefore indicating ALS-Plus syndrome, according to El Escorial clinical diagnostic criteria.⁴ These criteria suggests that the diagnosis of ALS requires evidence of lower and upper motor neuron degeneration, the progressive spread of symptoms, and the absence of other disease processes that might explain the degeneration of upper and lower motor neurons.⁵ The whole-brain distribution of CNS degeneration and pathology in ALS is not well described. Therefore, we performed an immunohistochemical whole-brain scan for evidence of pathological TDP-43 in a large cohort of sporadic ALS patients. We found TDP-43 pathology in multiple brain areas, suggesting that ALS is a disease that not only affects the pyramidal motor system, but instead is a multisystem neurodegenerative TDP-43 proteinopathy.

We examined a cohort of ALS patients as well as control participants without overt neurological or cognitive dysfunction. Clinically diagnosed ALS patients were included whose diagnosis was confirmed at autopsy and was consistent with the El Escorial criteria. In 5 patients, the motor neuron disease was accompanied by cognitive impairment (ALS-dementia). The participants were longitudinally followed in the Penn ALS Center; CNS samples came from the Center for Neurodegenerative Disease Research Brain Bank at the Penn center following neuropathology diagnostic assessment, as described.¹ Informed consent for autopsy was obtained from the legal representative in accordance with the local institutional review board.

Fresh tissues from the brain and spinal cord were fixed in 10% neutral buffered formalin, or in 70% ethanol with 150-mmol sodium chloride, paraffin-embedded, and cut into 6-μm sections. Sections were stained with hematoxylin and eosin, while immunohistochemistry was performed by means of the avidin–biotin complex detection technique (Vectastatin ABC kit; Vector Laboratories, Burlingame, California) with 3,3-diaminobenzidine as chromogen, using previously described methods.^1,6 The following primary antibodies were used: mouse monoclonal anti-ubiquitin 1510 (1:100 000) (Chemicon International, Temecula, California); rabbit polyclonal anti-TDP-43 (1:4 500) (Protein-Tech Group, Chicago, Illinois); and mouse monoclonal anti-TDP-43 (1:10 000) (Novus Biologicals, Littleton, Colorado). Sections were boiled in a pressure cooker in citrate antigen unmasking solution (Vector Laboratories) and counterstained with hematoxylin after immunohistochemistry. Double-labeling immunofluorescence was performed using the same primary antibodies and Alexa Fluor 488 and 594 conjugated secondary antibodies (Molecular Probes, Eugene, Oregon) followed by coverslipping with Vectashield-DAPI-mounting medium (Vector Laboratories), as described.¹ Digital images of immunohistochemical preparations were obtained using an Olympus BX 51 (Tokyo, Japan) microscope equipped with bright-field and fluorescence light sources with a digital camera, the DP71 (Olympus, Orangeburg, New York), and DP manager (Olympus). Digital images of immunofluorescence preparations were obtained using a Nikon TE2000 microscope (Nikon Instruments, Inc, Melville, New York) and were captured with a CoolSNAP Monochrome camera (Photometrics, Tucson, Arizona) and MetaMorph software (Molecular Devices, Downingtown, Pennsylvania). Neuropathological workup excluded all neurodegenerative diseases other than ALS, as described.^1,6

All data were analyzed using SPSS 15.0 for Windows (SPSS, Chicago, Illinois). The disease duration of the ALS patients was calculated as the median (interquartile range). The t test was used to compare the mean (SD) age at death of ALS patients and control participants. Severity ratings of TPD-43 pathology were performed using an arbitrary ordinal scale (0, none; 1, mild; 2, moderate; 3, severe).⁶ The median rating scores were calculated from grouped data across all patients for each region. Grouped data denotes that 1 stage follows continuously into the other; therefore, they represent classes rather than clearly distinguishable values on a numerical scale.

We examined 31 ALS patients (13 women and 18 men). Median disease duration was 2.0 years (interquartile range, 1.0-3.0 years), and mean (SD) age at death was 58.7 (10.1) years. We also scrutinized 8 control participants (5 women and 3 men); mean (SD) age at death was 64.8 (6.6) years, without overt neurological or cognitive dysfunction. There was no significant difference in age at death when comparing ALS patients with control participants (t₃₇ = -1.592, P = .12).

In addition to the stereotypical involvement of upper and lower motor neurons, pathological TDP-43 was present in multiple brain areas far beyond the pyramidal motor system, including the nigrostriatal system, neocortical and allocortical areas, and the cerebellum to a variable extent. Remarkably, none of the areas examined here were devoid of TDP-43 pathology. The frequency of TDP-43 pathology (percentage of cases in which TDP-43 was observed) is depicted in Figure 1, which also shows a color-coded topographical distribution scheme of pathology, based on severity ratings (Table). The median values of the ratings were calculated for all investigated cases. Cellular inclusions of all types were encountered throughout the CNS in the ALS patients; none were observed in the control participants. Although the number of ALS cases with associated dementia (n = 5) was too small for formal statistical comparison, the pathology in various neocortical brain areas or in the hippocampus appear to be more severe in ALS-dementia than in ALS without cognitive dysfunction. There was a trend suggesting that some subcortical structures were more severely affected in ALS-dementia, and it is possible that studies of a larger cohort of ALS-dementia patients might demonstrate statistical significance for these findings. Also, the small number of patients without bulbar findings (n = 4) did not significantly differ from patients with bulbar dysfunction. Neuronal cytoplasmic inclusion pathology included skein-like inclusions, Lewy body–like inclusions, and smaller, dense granules detected by multiple anti-TDP-43 antibodies (Figure 2). As described, the nuclei of affected neurons were devoid of the normal endogenous TDP-43 staining.¹ This also occurred in neurons with dense, diffuse, or granular cytoplasmic staining in the absence of bulky inclusions, which may reflect incipient TDP-43 inclusions or preinclusions. Presence of TDP-43 neuronal intranuclear inclusions was rare. Glial TDP-43 pathology was more frequent in the white as compared with the gray matter. In fact, oligodendroglial and, less commonly, astrocytic TDP-43-positive inclusions were encountered. Also, TDP-43 immunoreactive dystrophic neurites were widespread (Figures 1B and 2I). Double-labeling immunofluorescence studies on nigral sections showed that the Lewy body–like inclusions were TDP-43-positive and ubiquitin-positive, but some neu ronal inclusions and almost all glial inclusions were positive only for TDP-43 (Figure 3). The cohort contained 4 cases with a family history of neurodegenerative disease, including motor neuron disease (without superoxide dismutase-1 mutations), parkinsonism, and dementia (in 2 cases). Of these, the first 2 cases were the only ones that showed neuronal intranuclear inclusions. Control participants were virtually devoid of pathological TDP-43.

Until recently, ALS has been commonly regarded as a neurodegenerative disorder primarily involving the pyramidal motor system, but there is growing evidence that degenerative changes can occur elsewhere in the CNS of ALS patients (eg, hippocampus and nonmotor neocortical areas).^6-9 Further, degeneration in additional brain areas such as the substantia nigra and basal ganglia that are sometimes coupled with clinical evidence of dysfunction in other systems, such as somatomotor, visceromotor, autonomic, limbic, sensory, and additional cortical systems, have been described in case reports and small patient cohorts.^8,10-17 Recently, immunohistochemical studies have shown TDP-43 aggregates in upper and lower motor neurons in familial ALS cases (owing to genetic abnormalities other than those caused by superoxide dismutase-1 mutations) and in the lower motor neurons in more than 100 sporadic ALS cases.^1,3,6 Pathological TDP-43 has also been detected in the dentate gyrus of the hippocampus and the neocortex in both familial and sporadic ALS,^1,3,6 and some subcortical lesion sites have also been reported.^3,18 The full extent of TDP-43 pathology throughout the CNS was unknown in ALS, but assumed to be present mainly in upper and lower motor neurons, in keeping with the widely held concept that the pyramidal motor system is selectively involved in ALS. Moreover, clinical manifestations aside from those indicative of pyramidal motor system involvement exclude the diagnosis of classic ALS, and indicate an ALS-Plus syndrome, according to the El Escorial criteria.⁴ Thus, the presence of widespread TDP-43 pathology throughout many CNS areas in a large series of 31 ALS cases suggests that TDP-43 pathology is far more extensive in this disease than previously reported. Indeed, all cases here showed gray and/or white matter TDP-43 pathology in multiple brain areas outside the pyramidal motor system. This widespread distribution indicates that bothphylogenetically young and old brain areas beyond the upper and lower motor neuron system undergo TPD-43-linked neurodegeneration (Figure 1). Our finding of widespread involvement of areas outside the pyramidal motor system is substantiated by several neuroimaging studies and the growing evidence of cognitive or behavioral impairments in ALS.^19,20

Since the acceptance of this paper, two papers^21,22 have reported genetic evidence of mutations in the TDP-43 gene (TARDBP) that are pathogenic for familial ALS in rare kindred as well as in rare cases of sporadic ALS, thereby supporting our view that ALS neurodegeneration is mechanistically linked to TDP-43 pathology.

In summary, the finding of widespread CNS deposition of pathological TDP-43 in the CNS of ALS patients implies that ALS is a multiple-system TDP-43 proteinopathy involving neurons, oligodendroglia, and astrocytes. Finally, ALS may be situated at 1 end of a broad clinicopathological spectrum of multisystem degenerations, including FTLD with ubiquitin-positive inclusions, with or without motor neuron disease, that may share similar disease mechanisms linked to pathological TDP-43.^1,23,24 Efforts to block or reverse TDP-43 pathology in ALS may lead to disease-modifying therapies for this spectrum of disorder.

Correspondence: John Q. Trojanowski, MD, PhD, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Hospital of the University of Pennsylvania, 3600 Spruce St, Maloney Bldg, 3rd Flr, Philadelphia, PA 19104-4283 (trojanow@mail.med.upenn.edu).

Accepted for Publication: November 15, 2007.

Author Contributions: Dr Trojanowski had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Geser, Brandmeir, Kwong, Lee, and Trojanowski. Acquisition of data: Geser,Brandmeir, Martinez-Lage, Elman, McCluskey, andTrojanowski. Analysis and interpretation of data: Geser, Brandmeir, McCluskey, Xie, Lee, and Trojanowski. Drafting of the manuscript: Geser, Brandmeir, and Trojanowski. Critical revision of the manuscript for important intellectual content: Geser, Brandmeir, Kwong, Martinez-Lage, Elman, McCluskey, Xie, Lee, and Trojanowski. Statistical analysis: Geser, Brandmeir, and Xie. Obtained funding: Lee and Trojanowski. Administrative, technical, and material support: Brandmeir, Martinez-Lage, Elman, McCluskey, Lee, and Trojanowski. Study supervision: Lee and Trojanowski.

Financial Disclosure: None reported.

Funding/Support: This work was funded by grants AG10124 and AG17586 from the National Institutes of Health.

Additional Contributions: Theresa Schuck, BA, Monica Getahun, BS, Natascha Gownaris, and John Robinson, BS, provided expert technical assistance, and Manuela Neumann, MD, provided critical comments on the manuscript. Furthermore, we thank our patients and families who made this research possible.