Frontotemporal Lobar Degeneration With Tdp43 Inclusions, Grn-Related

A number sign (#) is used with this entry because this form of frontotemporal lobar degeneration with TDP43 inclusions (FTLD-TDP) is caused by mutation in the GRN gene (138945) encoding progranulin.

Description

Clinically, FTLD-TDP is a type of frontotemporal dementia (see FTD; 600274) which shows variable phenotypic expression, but most commonly presents with social, behavioral, or language deterioration, rather than memory or motor deficits. Other variations of the phenotype have been referred to as 'dysphasic disinhibition dementia' and 'primary progressive aphasia' (PPA) (Huey et al., 2006; Mukherjee et al., 2006; Mesulam et al., 2007). Some patients may present with a clinical diagnosis of Alzheimer disease (AD; 104300) or Parkinson disease (PD; 168600), which are part of the phenotypic spectrum of this disorder (Brouwers et al., 2007).

Genetic Heterogeneity of FTLD-TDP

The specific presence of TDP43 (TARDBP; 605078)-positive inclusions on neuropathologic examination defines a genetically heterogeneous group of dementias known collectively as 'FTLD-TDP.' FTLD-TDP is a neuropathologic diagnosis; only about 20% of patients with this neuropathologic diagnosis have GRN mutations (review by Van Deerlin et al., 2010).

TDP43-positive inclusions also occur in ALS10 (612069), caused by mutation in the TARDBP gene (605078); IBMPFD (167320), caused by mutation in the VCP gene (601023); and FTDALS (105550), caused by mutation in the C9ORF72 gene (614260).

Mackenzie and Rademakers (2007) provided a detailed review of the molecular genetics of FTLD, with special emphasis on FTLDU. Cairns and Ghoshal (2010) reviewed the molecular pathology and genetic heterogeneity of FTLD, including FTLD-TDP, and also noted that FTLDU is now referred to as FTLD-TDP.

Clinical Features

In an Ohio family of Bavarian origin, Morris et al. (1984) described a distinct disease entity, which they termed 'hereditary dysphasic dementia,' in 10 of 16 members who lived past the age of 60. Over 3 generations, the disease was inherited as an autosomal dominant trait; male-to-male transmission was observed. The clinical characteristics included progressive cognitive deficits with memory loss and personality changes, severe dysphasic disturbances leading to mutism, and hyperphagia. Morris et al. (1984) suggested that the kindred reported by Kim et al. (1981) had the same disorder. This was a family of Italian extraction in which 4 of 10 members of a single generation developed dementia, dysphasia, and, in some cases, parkinsonian signs and bulimia.

Froelich et al. (1997) and Basun et al. (1997) reported a large Swedish family with rapidly progressive FTD inherited in an autosomal dominant pattern. The mean age at onset was 51 years, and mean disease duration was 3 years. Four patients were described in detail. Two patients presented with speech disturbances leading to a progressive, nonfluent aphasia, 1 patient had onset symptoms of leg apraxia with akinesia and muscular rigidity, and 1 patient developed reckless driving and personality changes. All developed loss of spontaneous speech, and 3 had emotional bluntness. Cerebral perfusion was decreased in the frontal areas in all patients. Postmortem examination showed frontal lobe degeneration with spongy changes and gliosis. Skoglund et al. (2009) reported follow-up of the Swedish family reported by Froelich et al. (1997) and Basun et al. (1997) and added 6 additional affected individuals. These additional patients presented with personality and behavioral changes (2), language impairment (2), and memory impairment (2). Although there was a large variation of the initial symptoms in this family, the common pattern of clinical features included rapid disease progression and ultimately nonfluent aphasia with loss of spontaneous speech patients. Limb ataxia and parkinsonism were uncommon symptoms, and there was no motor neuron disease, although dysphagia was seen in 3 patients. Neuropathologic examination revealed frontotemporal neurodegeneration with ubiquitin and TDP43 immunoreactive intraneuronal inclusions. Molecular analysis identified a frameshift mutation in the GRN gene (138945.0016) that resulted in functional haploinsufficiency.

Lendon et al. (1998) studied a kindred with the same manifestations as those in the kindred reported by Morris et al. (1984) and renamed the disorder hereditary dysphasic disinhibition dementia (HDDD). For both kindreds, the mean age of disease onset was approximately 60 years. The range of duration of disease was 5 to 11 years in the kindred reported by Lendon et al. (1998) and 4 to 22 years in the kindred reported by Morris et al. (1984). Early manifestations included gradual onset with progressive memory loss and other cognitive deficits. There were more abnormalities of language earlier in the course of disease, frequently as the initial symptom, than are typically seen in the dementia of Alzheimer disease (AD; 104300). These included hesitancy of speech, reduction in spontaneous output, diminished fluency, and dysnomia progressing to auditory and reading-comprehension deficits, and eventual mutism. Behavioral and personality changes were occasional in the kindred reported by Morris et al. (1984) and more notable in the kindred reported by Lendon et al. (1998), sometimes occurring as the initial symptoms. Disinhibition, hypersexuality, bulimia, inappropriate behavior, and unaccustomed excessive alcohol consumption was present in some patients in both kindreds. Also in both families, parkinsonian features frequently developed, usually at a later stage of the disease.

Kertesz et al. (2000) reported a family with a highly penetrant form of autosomal dominant frontotemporal dementia, or 'clinical Pick disease.' Age at onset ranged from 38 to 44 years, with a rapid progression. Affected members had similar clinical features characterized by social withdrawal, lack of motivation, inappropriate behavior, disinhibition, and perseveration. Strikingly similar features were hyperorality and stealing. Kertesz et al. (2000) noted the similarities to Kluver-Bucy syndrome. Neuropsychologic examination indicated inattention, word-finding difficulties, and disorganization of memories.

Rosso et al. (2001) reported a large Dutch family, previously reported by Heutink et al. (1997) with autosomal dominant frontotemporal dementia (FTD) linked to chromosome 17q21-q22 (lod score of 3.46), but without mutations in the tau gene (MAPT; 157140). Mean age at onset was 61 years, with loss of initiative and decreased spontaneous speech as the most predominant presenting symptoms. Other features included agitation and restlessness, language difficulties, and hyperorality.

Krefft et al. (2003) reported 3 sibs with classic primary progressive aphasia, defined as an isolated progressive language dysfunction. Onset of word-finding and naming difficulties occurred at ages 60, 61, and 65 years, respectively. All showed left frontotemporal atrophy. The eldest sib had mild right motor impairment, including hemiparesis and mild tremor, and died mute and bedridden 12 years after onset. A younger brother had behavioral changes, and a younger sister had significant parkinsonism, cortical release signs, and dementia. She died 4 years after onset, and postmortem neuropathologic examination showed marked cortical thinning with neuronal loss, gliosis, spongiosis, and ubiquitin-positive inclusions within cortical neurons. Other features included hippocampal sclerosis and Lewy bodies in the substantia nigra, which may have represented a concurrent pathologic process. Genetic analysis excluded common mutations in the MAPT gene.

Mesulam et al. (2007) reported 2 sisters with a typical presentation of primary progressive aphasia. The phenotype was an isolated aphasia beginning at ages 65 and 62 years, respectively, without other features. There was rapid progression of aphasia to complete mutism within 2 to 3 years. One sister developed a right-sided tremor, clumsiness, rigidity, and impaired motor function. Genetic analysis identified a heterozygous mutation in the GRN gene (R493X; 138945.0009) in both sisters. Gliebus et al. (2010) reported neuropathologic findings of 1 of the sisters with PPA reported by Mesulam et al. (2007). She died at the age of 67 years, 5 years after onset. She had significantly more neuronal intranuclear and cytoplasmic TDP43-positive inclusions and dystrophic neurites in the left inferior parietal lobule and superior temporal gyrus compared to the right, consistent with involvement of language areas of the brain. The findings illustrated a concordance between the aphasic phenotype in this patient and pathologic involvement of the language-related regions of the brain rather than memory-related regions.

In a retrospective study, Van Deerlin et al. (2007) compared the clinical features of 9 patients with FTLDU and GRN mutations to 19 patients with the same pathologic diagnosis of FTLDU but without GRN mutations. Family history was significantly more common in those with mutations. Although both patient groups had social and personality deficits, neuropsychologic testing showed that those with a GRN mutation had a significant recognition memory deficit, whereas those without a GRN mutation had a significant language deficit. Van Deerlin et al. (2007) concluded that patients with a GRN mutation differ clinically from those with the same pathologic diagnosis but no GRN mutation.

Rogalski et al. (2008) observed a significantly higher frequency of self-reported learning disabilities among 108 PPA probands (14.8%) and their first-degree relatives (29.6%) compared to 84 individuals with the behavioral variant of FTD (7.1%) and their relatives (14.3%), 154 individuals with AD (4.5%) and their relatives (10.4%), and 353 controls (1.4%) and their relatives (6.8%). PPA patients and PPA family members reported deficits specifically in the area of language, suggesting that individuals who develop PPA may have an antecedent selective vulnerability in the anatomic brain regions that underlie language.

Rohrer et al. (2008) reported a large British kindred with FTLDU associated with a mutation in the GRN gene (138945.0005). The average age at disease onset was 57.8 years. All patients had clinical and radiographic features of frontotemporal lobar degeneration with behavioral changes and language deficits. Nonfluent aphasia was present in 2 patients, and 3 became mute several years into the illness. Most also had features suggestive of parietal lobe involvement, including dyscalculia, visuoperceptual/visuospatial dysfunction, and limb apraxia. Brain imaging showed extension of the atrophy to the parietal lobe.

Moreno et al. (2009) reported the clinical features of 21 patients of Basque origin with FTD caused by the same heterozygous mutation in the GRN gene (138945.0019). The mean age at onset was 59.2 years (range, 42 to 71 years), and 4 of 21 patients died after a mean duration of 4.75 years. Overall, 14 (66.7%) had a diagnosis of behavioral variant FTD, 10 (47.6%) had a diagnosis of corticobasal degeneration, and 7 (33.3%) had a diagnosis of progressive nonfluent aphasia. None developed signs of motor neuron disease/ALS, but 8 with corticobasal degeneration had motor signs, including limb rigidity or apraxia and myoclonus. The most prominent behavioral symptoms were apathy, impulsivity, disinhibition, and bulimia, suggesting involvement of the medial frontal and orbitofrontal cortex. Dysgraphia, dyscalculia, apraxia, and hemineglect suggested parietal dysfunction.

Kelley et al. (2010) reported a large family with autosomal dominant inheritance of FTLD associated with a heterozygous mutation in the GRN gene (154delA; 138945.0017). Of 10 affected individuals, 6 presented with early amnestic symptoms resulting in initial clinical diagnoses of Alzheimer disease or amnestic mild cognitive impairment, and 3 with frontotemporal dementia; 1 had nonspecific dementia. Neuropathologic examination of 6 individuals showed FTLD with ubiquitin-positive neuronal cytoplasmic and intranuclear inclusions, even in those with an AD diagnosis. The mean age at onset was younger in the third generation (60.7 years) than in the second generation (75.8 years). Kelley et al. (2010) noted that the presentation in some individuals with GRN-related FTLD may include Alzheimer disease-like clinical features, particularly anterograde amnesia.

In an international collaborative study comparing clinical features of 97 unrelated patients with FTLD due to GRN mutations and 453 patients with FTLD who did not have GRN mutations, Chen-Plotkin et al. (2011) found clinical differences between the 2 groups. Those with GRN mutations had a younger age at disease onset (median age 58.0 vs 61.0 years), younger age at death (median age 65.5 vs 69.0 years), and less motor neuron disease (5.4% vs 26.3%). Clinical diagnoses of Parkinson disease, corticobasal syndrome, and progressive supranuclear palsy were more common in GRN mutation carriers (5.3% vs 1.3%), and more patients with GRN mutations presented with aphasia. Fifty different mutations were identified, with the most common being R493X (138945.0009), found in 18.6% of GRN cases. The A9D mutation (138945.0008) was found in 6.2% of cases and associated with an even younger age at onset (51.0 years) and more parkinsonian features compared to those with other GRN mutations.

Neuropathologic Findings

Morris et al. (1984) reported that complete neuropathologic examination of 4 of their patients showed asymmetric focal cerebral atrophy (characteristic of Pick disease), neuritic plaques (characteristic of Alzheimer disease), and depletion of neurons in the pigmented nuclei of the brainstem (characteristic of paralysis agitans). There was also cortical neuronal loss, nonspecific spongiform degeneration of the external layers of cerebral cortex, and reactive gliosis. Pick cells and Pick bodies were absent, and in 1 patient, Lewy bodies were present in nigral neurons. Transmissibility studies were negative. Morris et al. (1984) concluded that this is a distinct entity but 'may be best considered as part of a Pick-Alzheimer spectrum of cortical neuronal degenerations.'

Behrens et al. (2007) provided neuropathologic analysis of another member of the family reported by Morris et al. (1984). She had a disease course similar to that of other family members, with onset at age 62 years of personality changes and disinhibition, followed by nonfluent dysphasia and memory loss that progressed to mutism and total dependence, with death at age 84. There was severe generalized brain atrophy. Histopathology showed superficial microvacuolation, marked neuronal loss, gliosis, and ubiquitin-positive, tau-negative cytoplasmic and intranuclear neuronal inclusions in the frontal, temporal, and parietal cortices. There were also frequent neuritic plaques and neurofibrillary tangles in the parietal and occipital cortices. The case met neuropathologic criteria for both FTLD-U and Alzheimer disease. No Pick bodies were present.

Neuropathologic examination of affected members of a family by Kertesz et al. (2000) showed severe hippocampal atrophy, frontal lobe atrophy, loss of pigmented neurons in the substantia nigra, and ubiquitin-positive inclusions that were not immunoreactive to tau or alpha-synuclein (SNCA; 163890). The authors referred to these inclusions as ITSNU. No mutations in the MAPT gene were detected. Neuropathologic examination of affected members of a family by Rosso et al. (2001) also showed ubiquitin-positive inclusions that were not immunoreactive to tau or alpha-synuclein. There were no abnormalities in tau isoform distribution.

Four affected members of the large Canadian family reported by Mackenzie et al. (2006) had undergone postmortem examination. In all cases cerebral atrophy was moderate to severe and was largely restricted to the frontal lobe. There was often mild atrophy at the head of the caudate nucleus and some loss of pigmentation of the substantia nigra. Microscopic examination of the neocortex showed nonspecific chronic degenerative changes including neuronal loss and gliosis with an anterior-to-posterior anatomic gradient of severity. Apart from a small number of neurofibrillary tangles identified in 1 family member, no other pathology was identified with silver stains, or immunohistochemistry for tau, alpha-synuclein, or nonphosphorylated or phosphorylated neurofilament; specifically, there were no senile plaques, Pick bodies, Lewy bodies, glial inclusions, or achromatic neurons. In contrast, ubiquitin-immunoreactive neurites and neuronal cytoplasmic inclusions were present in the superficial laminae of the frontal and temporal neocortex in all cases. Discrete dense ubiquitin-immunoreactive neuronal intranuclear inclusions were identified in all 4 cases and had a similar anatomic distribution to the dense neuronal cytoplasmic inclusions, being most numerous in the frontal neocortex and striatum and less common in the dentate granule cells, globus pallidus, and thalamus. The number of neuronal intranuclear inclusions varied between the anatomic regions, and they were always much less numerous than the neuronal cytoplasmic inclusions. All neuronal intranuclear inclusions were reactive to SUMO1 in addition to ubiquitin, and a proportion was positive for PML, suggesting to Mackenzie et al. (2006) that these inclusions form in the nuclear body and providing a possible mechanism of neurodegeneration in tau-negative FTD linked to chromosome 17q21.

Neumann et al. (2006) identified TDP43 (605078) as the major disease protein in both ubiquitin-positive, tau-, and alpha-synuclein-negative frontotemporal lobar degeneration and amyotrophic lateral sclerosis (see 105400). Pathologic TDP43 is hyperphosphorylated, ubiquitinated, and cleaved to generate C-terminal fragments and was recovered only from affected central nervous system regions, including hippocampus, neocortex, and spinal cord. Neumann et al. (2006) concluded that TDP43 represents the common pathologic substrate linking these neurodegenerative disorders.

Mukherjee et al. (2006) described additional neuropathologic findings of 8 affected individuals from the family reported by Lendon et al. (1998). Atrophy varied from moderate to severe and was most pronounced in the frontal lobe, with lesser degrees of atrophy in the temporal and parietal lobes. All cases had severe cortical neuronal loss, status spongiosus, and reactive astrocytosis. The hippocampus was less atrophied and showed less neuronal loss than is usually seen in Alzheimer disease. Immunohistochemical analysis showed ubiquitin-positive, tau-negative, neuronal cytoplasmic inclusions; dystrophic neurites; and neuronal intranuclear inclusions in 7 of 8 cases. Further studies showed that progranulin was not a component of any of the pathologic inclusions.

Forman et al. (2006) performed a clinicopathologic assessment of 124 patients with either a clinical or pathologic diagnosis of frontotemporal dementia. Neuropathologic examination showed that 46% had a tauopathy, 29% had FTLD with ubiquitin inclusions, and 17% had findings consistent with Alzheimer disease. Patients with FTLD with ubiquitin inclusions were more likely to present with social and language dysfunction; tauopathies were more commonly associated with an extrapyramidal disorder; and AD was associated with greater deficits in memory and executive function.

Davion et al. (2007) found that 4 of 9 patients with a pathologic diagnosis of FTLDU had mutations in the GRN gene (138945.0009; 138945.0012; 138945.0013). Two presented with frontotemporal dementia and 2 with primary progressive aphasia. Neuropathologic examination of all 9 cases showed that those with the GRN mutations had more frequent ubiquitinated neuronal cytoplasmic and intranuclear inclusions in the frontal lobe, temporal lobe, and striatum than those without GRN mutations, who had more neurocytoplasmic inclusions in the dentate gyrus.

Grossman et al. (2007) noted that several subtypes of FTLDU had been characterized based on neuropathologic localization of TDP43-positive ubiquitin inclusions. In a retrospective review of 23 patients, the authors found notable clinical differences between the subtypes. Only 4 of the 23 patients had mutations in the GRN gene; this was a pathologic study. Patients with numerous TDP43-positive neuronal intracytoplasmic inclusions had shorter survival; patients with numerous TDP43-positive neurites had difficulty with object naming; and patients with TDP43-positive neuronal intranuclear inclusions had substantial executive deficits. Different anatomical distributions of ubiquitin pathologic features in FTLDU subgroups were consistent with their cognitive deficits.

Diagnosis

Finch et al. (2009) used ELISA analysis to measure plasma GRN levels in a consecutive series of 207 patients with FTLD, 70 control individuals, 72 early-onset probable Alzheimer disease patients, and 9 symptomatic and 18 asymptomatic relatives of GRN mutation carriers. All 8 FTLD patients with GRN loss-of-function mutations showed significantly reduced plasma GRN levels of about one-third of the levels found in non-GRN carriers and controls individuals (p less than 0.001). There was no overlap in the distribution of plasma GRN levels between the 8 GRN mutation carriers (range, 53-94 ng/ml) and 191 non-GRN mutation carriers (range, 115-386 ng/ml). Similar low levels of GRN were identified in asymptomatic GRN mutation carriers. ELISA analysis also identified 1 (1.4%) probable AD patient who carried a GRN mutation. The ELISA technique only identified full-length GRN, but not GRN fragments. The study demonstrated that ELISA analysis of plasma GRN can detect asymptomatic carriers of pathogenic GRN mutations.

Mapping

Froelich et al. (1997) mapped FTD in a large Swedish family to chromosome 17. Probable linkage to chromosome 17q12-q21 was found, with a maximum 2-point lod score of 2.76 at theta = 0.0 for marker D17S806, and a peak multipoint lod score of 2.86 for the same marker. Sequencing analysis excluded mutations in the MAPT gene.

By linkage analysis, Lendon et al. (1998) demonstrated linkage of the disorder in their kindred to chromosome 17q21-q22, with a maximum lod score of 3.68 at 0.0 recombination.

Rademakers et al. (2002) reported the results of a genomewide search in a 4-generation pedigree with autosomal dominant early-onset dementia (mean age of onset 64.9 years, range 53 to 79 years). In this family they excluded mutations in Alzheimer disease genes, mutations in the prion protein gene (PRNP; 176640), and exons 9 through 13 of the MAPT gene. Rademakers et al. (2002) obtained conclusive linkage with chromosome 17q21 markers with a maximum multipoint lod score of 5.51 at D17S951 and identified a candidate region of 4.8 cM between D17S1787 and D17S958 containing MAPT. Clinical and neuropathologic follow-up of the family showed that the phenotype most closely resembled frontotemporal dementia characterized by dense ubiquitin-positive neuronal inclusions that were tau-negative. Extensive mutation analysis of MAPT identified 38 sequence variants in exons, introns, untranslated regions, and the 5-prime regulatory sequence; however, none occurred within the disease haplotype.

Mackenzie et al. (2006) reported a large Canadian family with autosomal dominant FTD linked to chromosome 17q21 with a maximum multipoint lod score of 3.911 containing MAPT. They could not identify point mutations in MAPT by direct sequencing or any gross MAPT gene alterations using fluorescence in situ hybridization.

Molecular Genetics

Zhukareva et al. (2001) stated that no tau gene mutation had been detected in the family reported by Lendon et al. (1998) in which linkage to 17q21-q22 had been established. However, they identified a loss of tau protein by Western blot analysis of protein extracts from brain regions both with and without neuronal degeneration, and concluded that, functionally, this loss of tau protein may be equivalent to pathogenic mutations in the tau gene.

In a series of 98 genealogically unrelated Belgian patients with frontotemporal lobar degeneration (FTLD), van der Zee et al. (2006) identified an ancestral 8-cM MAPT-containing haplotype in 2 patients belonging to multiplex families DR2 and DR8, without demonstrable MAPT mutations, in which FTLD was conclusively linked to 17q21 (maximum summed lod score of 5.28 at D17S931). Interestingly, the same DR2/DR8 ancestral haplotype was observed in 5 additional familial FTLD patients, indicating a founder effect. In the FTLD series, the DR2/DR8 ancestral haplotype explained 7% (7 of 98) of FTLD and 17% (7 of 42) of familial FTLD and was 7 times more frequent than MAPT mutations (1 of 98, or 1%). Clinically, DR2/DR8 haplotype carriers presented with FTLD often characterized by language impairment, and in 1 carrier the neuropathologic diagnosis was FTLD with rare tau-negative ubiquitin-positive inclusions. Together, van der Zee et al. (2006) concluded that their results strongly suggested that the DR2/DR8 founder haplotype in 17q21 harbors a tau-negative FTLD-causing mutation that is a much more frequent cause of FTLD in Belgium than MAPT mutations.

Baker et al. (2006) identified a frameshift mutation in exon 1 of the progranulin gene (138945.0005) in family UBC17 reported by Mackenzie et al. (2006). Baker et al. (2006) then sequenced GRN in affected individuals from an additional 41 families with clinical and pathologic features consistent with tau-negative FTD. This analysis identified an additional 7 GRN mutations in 8 families, each predicted to cause premature termination of the coding sequence. The mutations included 4 nonsense mutations, 2 frameshift mutations, and a mutation at the 5-prime site of exon 8.

Cruts et al. (2006) independently found mutation in progranulin in a Belgian founder family reported by van der Zee et al. (2006): a mutation in the splice donor site of intron 0, indicating loss of mutant transcript by nuclear degradation (138945.0001). Cruts et al. (2006) also found a mutation of the initiating methionine and found that GRN haploinsufficiency leads to neurodegeneration because of reduced GRN-mediated neuronal survival. Furthermore, in a Belgian series of familial FTD patients, GRN mutations were 3.5 times more frequent than mutations in MAPT, underscoring a principal involvement of GRN in FTD pathogenesis.

Both Baker et al. (2006) and Cruts et al. (2006) found a premature termination mutation in the Dutch family reported by Rademakers et al. (2002) (138945.0002).

In affected members of the family with HDDD originally reported by Lendon et al. (1998), Mukherjee et al. (2006) identified a heterozygous mutation in the GRN gene (A9D; 138945.0008).

Huey et al. (2006) identified a nonsense mutation in the GRN gene (R493X; 138945.0009) in 3 unrelated patients with rapidly progressive frontotemporal dementia. All had predominantly behavioral symptoms, and 2 families showed mild parkinsonism. Brain imaging of 2 patients showed frontotemporal atrophy and hypometabolism with a right-sided predominance. In the patients with primary progressive aphasia reported by Krefft et al. (2003), Mesulam et al. (2007) identified a heterozygous R493X mutation in the GRN gene.

Le Ber et al. (2007) identified 9 novel null mutations in the GRN gene in 10 (4.8%) of 210 unrelated patients with frontotemporal dementia. The frequency was 12.8% (5 of 39) in familial cases and 3.2% (5 of 158) in sporadic cases. The phenotype was heterogeneous with age at onset ranging from 45 to 74 years and frequent occurrence of early apraxia (50%), visual hallucinations (30%), and parkinsonism (30%). No GRN mutations were identified in 43 patients with a dementia and motor neuron disease. Le Ber et al. (2007) stated that 31 GRN mutations had been identified in patients worldwide.

In affected members of the family originally reported by Morris et al. (1984), Mukherjee et al. (2008) identified a heterozygous mutation in the GRN gene (138945.0014). Affected members of another family with the disorder were found to carry the same mutation, and haplotype analysis indicated a founder effect. Western blot analysis showed a 50% reduction in GRN protein compared to controls, suggesting haploinsufficiency.

Borroni et al. (2008) identified a pathogenic mutation in the GRN gene (138945.0015) in 4 (1.64%) of 243 unrelated Italian patients with a clinical diagnosis of FTLD. Two female patients were diagnosed with the behavioral variant of frontotemporal dementia, and 2 males with progressive nonfluent aphasia. The estimated age at onset ranged from 53 to 64 years, and all showed evidence of hypoperfusion of the frontotemporal brain regions. Three of the 4 had a family history of the disorder. Considering all patients in the study with a well-known family history for dementia, the frequency of this mutation was 6% (4 of 84). Haplotype analysis indicated a founder effect.

Gijselinck et al. (2008) provided a detailed review of granulin mutations associated with frontotemporal lobar degeneration. They noted that 63 heterozygous loss-of-function mutations had been identified in 163 families worldwide, representing about 5 to 10% of FTLD.

Seelaar et al. (2008) found a family history consistent with autosomal dominant inheritance in 98 (27%) of 364 probands with frontotemporal dementia. Among the familial cases, mutations in the GRN and MAPT gene were identified in 6% and 11%, respectively. Those with GRN mutations had a higher mean age at onset (61.8 years) compared to those with MAPT mutations (52.4). Neuropathologic findings, when available, were consistent with genetic analysis.

In a population-based study of 59 patients with pathologically confirmed FTLDU and 433 controls, Rademakers et al. (2008) identified a C-to-T SNP (rs5848; 138945.0018) in the 3-prime untranslated region of the GRN gene that conferred increased risk for the development of FTLDU when present in the homozygous state. Functional studies showed that the minor T allele increased binding of MIR659, a translation suppressor, providing a novel mechanism for loss of GRN function.

Among 225 patients with a diagnosis of FTLD, Rohrer et al. (2009) found that 41.8% had some family history of the disorder, although only 10.2% had a clear autosomal dominant history. Those with the behavioral variant of the disorder were more likely to have a positive family history than those with the language syndromes. Mutations in the MAPT and GRN genes were found in 8.9% and 8.4% of the cohort, respectively.

Yu et al. (2010) reported the results of a large collaborative study of GRN mutations involving 8 academic centers. Twenty-four pathogenic GRN mutations, including 8 novel mutations, were found among 434 patients with various forms of cognitive neurodegenerative diseases. Approximately 55% of the patients with FTD for whom information was available had a family history of the disorder. Overall, the frequency of GRN mutations was 6.9% (30 of 434) of all FTD-spectrum cases. The frequency was 21.4% (9 of 42) in those with a pathologically confirmed diagnosis of FTLD-U; 16.0% (28 of 175) of FTD-spectrum cases with a family history; and 56.2% (9 of 16) of FTLD-U with a family history. The authors noted that GRN mutations were found only in FTD-spectrum cases and not in other related neurodegenerative diseases, such as Pick disease (172700) or progressive supranuclear palsy (PSNP1; 601104). In addition, GRN mutations were not found in patients with ALS (105400) or multiple system atrophy (MSA; 146500) in whom TDP43 deposits were a neuropathologic feature. Yu et al. (2010) concluded that haploinsufficiency of GRN is the predominant mechanism leading to FTD.

Genetic Modifiers

Van Deerlin et al. (2010) performed a genomewide association study of 515 individuals with FTLD-TDP and 2,509 controls from 45 clinical centers representing 11 countries. Eighty-nine (17.7%) of the patients had mutations in the GRN gene, and 23% had accompanying motor neuron disease. A significant association was found between FTLD-TDP and 12 SNPs in strong linkage disequilibrium (LD) within a 68-kb interval spanning the TMEM106B gene (613413) on chromosome 7p21.3. The most significant association was with a T-to-C transition (rs1990622) located 6.9-kb downstream of TMEM106B (odds ratio of 0.61; p = 1.08 x 10(-11)). The findings were replicated in a cohort of 89 patients, including 10 (13.5%) with GRN mutations. When stratified by GRN mutation status, rs1990622 still showed a significant association with FTLD-TDP, both in those with and without GRN mutations (p = 1.34 x 10(-9) and 6.90 x 10(-7), respectively). Further studies showed the T risk allele of rs1990622 was significantly correlated with increased TMEM106B protein and mRNA expression in lymphoblastoid cell lines. Brain tissue from 18 patients with FTLD-TDP showed increased TMEM106B expression compared to controls, and this expression correlated with presence of the T risk allele; these results were found in both GRN-positive and GRN-negative individuals. The results were compatible with a model in which mutations in GRN may act upstream of TMEM106B expression in increasing the risk for FTLD-TDP. Van Deerlin et al. (2010) concluded that a locus on chromosome 7p21.3, most likely reflecting variants affecting expression of the TMEM106B gene, represents a genetic risk factor for the development of FTLD-TDP, both in patients with and without GRN mutations.

Finch et al. (2011) found a significant association between SNPs in the TMEM106B gene and disease penetrance in patients with FTLD due to GRN mutations. In a group of 78 GRN mutation-positive patients, there was a highly significant decrease in the frequency of homozygous carriers of the minor alleles of 3 SNPs, with the strongest results for rs1990622 (CC genotype frequency of 2.6% in patients vs 19.1% in controls, p = 0.009). Only 2 patients were homozygous for these SNPs, and both showed later disease onset compared to other patients. There was also a significant association between the minor alleles of these SNPs and increased plasma GRN protein levels in controls and increased GRN mRNA levels in patients and controls. There was an inverse correlation between TMEM106B levels and GRN levels. Sequencing of the TMEM106B gene identified a coding variant (thr185-to-ser, T185S, rs3173615) that was in linkage disequilibrium with 2 of the other SNPs. Finch et al. (2011) noted that haploinsufficiency for GRN is associated with disease development, and postulated that variation in TMEM106B may alter GRN levels, thus influencing disease penetrance in mutation carriers.

In a study of 50 GRN mutation carriers from 4 families with FTLD, Cruchaga et al. (2011) found a significant association between the A risk allele of rs1990622 and earlier age at onset: those homozygous for the risk A allele had a median onset 13 years earlier than those heterozygous and homozygous for the minor G allele (p = 9.9 x 10(-7)). There was also an association between the risk allele and decreased GRN plasma levels in both mutation carriers and healthy older adults. However, unlike the findings of Van Deerlin et al. (2010), Cruchaga et al. (2011) found no association between the SNP and TMEM106B or GRN mRNA levels in frontal cortex from individuals without dementia, suggesting that the association of rs1990622 with GRN plasma levels is not driven by changes in gene expression. The T185S variant was in perfect linkage disequilibrium with rs1990622.

Gallagher et al. (2017) found an association between the A risk allele of rs1990622 and increased TMEM106B expression in several cells types, including lymphoblastoid cells lines, and in multiple brain regions of healthy human individuals. Increased expression of TMEM106B in HeLa cells and neurons resulted in enlarged lysosomes and increased cell death. A nearby noncoding variant, rs1990620, was also identified, and was demonstrated to recruit the chromatin organizing protein CTCF (604167) to influence long-range chromatin-looping interactions between multiple cis-regulatory elements, including the TMEM106B promoter. The findings suggested that a variant at the locus on 7p21 may influence gene expression, including CTCF-mediated gene regulation of TMEM106B, with a putative role in neurodegeneration.

Animal Model

Tsai et al. (2010) generated an FTLDU mouse model by transgenically overexpressing Tdp43 in forebrain. Transgenic mice exhibited impaired learning and memory, progressive motor dysfunction, and hippocampal atrophy. The impairments were accompanied by reduced levels of phosphorylated Erk (see MAPK1; 176948) and phosphorylated Creb (CREB1; 123810) and increased levels of gliosis in brains of transgenic mice. Cells with Tdp43-positive, ubiquitin-positive neuronal cytoplasmic inclusions (NCIs) and Tdp43-deleted nuclei appeared in transgenic mouse brains in an age-dependent manner. Tsai et al. (2010) concluded that increased levels of TDP43 protein in forebrain are sufficient to lead to formation TDP43-positive, ubiquitin-positive NCIs and neurodegeneration.