Parietal Lobe Deficits in Frontotemporal Lobar Degeneration Caused by a Mutation in the Progranulin Gene

Abstract

Objective

To describe the clinical, neuropsychological and radiological features of a family with a C31LfsX35 mutation in the progranulin gene (PGRN).

Design

Case series

Patients

A large British kindred (DRC255) with a PGRN mutation was assessed. Affected individuals presented with a mean age of 57.8 (54 to 67) and mean duration of disease of 6.1 years (2 to 11).

Results

All cases exhibited a clinical and radiological phenotype compatible with FTLD based on current consensus criteria. However, unlike sporadic FTLD, parietal deficits consisting of limb apraxia, dyscalculia, visuoperceptual and/or visuospatial impairment were a common feature, and brain imaging showed posterior extension of frontotemporal atrophy to involve the parietal lobes. Other common clinical features included language output impairment with either dynamic aphasia or non-fluent aphasia, and a behavioural syndrome dominated by apathy.

Conclusions

We propose that parietal features may be a prominent feature of PGRN mutations and that this may be due to disruption of fronto-parietal functional pathways.

INTRODUCTION

Frontotemporal lobar degeneration (FTLD) is a group of disorders characterized by focal atrophy of the frontal and temporal lobes. Three clinical syndromes are described by consensus criteria (1): frontotemporal dementia (FTD), progressive non-fluent aphasia (PNFA) and semantic dementia (SD). FTD is characterized by early personality change and progressive behavioural symptoms. PNFA presents with speech production difficulties. SD presents characteristically with anomia, poor single-word comprehension and fluent aphasia. These syndromes overlap clinically and radiologically with one another and with other neurodegenerative disorders including corticobasal degeneration (CBD), progressive supranuclear palsy (PSP) and motor neuron disease (MND) (2,3).

Approximately 30-50% of patients with FTLD have a family history of dementia with autosomal dominant inheritance (4). A proportion of these families have a mutation in the tau gene (MAPT) on chromosome 17q21.1 (www.molgen.ua.ac.be/FTDMutations). However, tau mutations collectively account for only around 4% of all patients with FTLD and 25% of cases with a positive family history (5,6). Recently, mutations in the progranulin gene (PGRN), located 1.7Mb away from MAPT at 17q21.32, have been identified in families with tau-negative, ubiquitin-positive inclusions on neuropathological examination (FTLD-U). Over 30 mutations have been detected to date (6,7,8,9), and together may account for around 5-10% of FTLD cases (6). Initial descriptions suggest that PGRN mutations are particularly associated with PNFA (7,8), however the phenotype continues to be defined.

METHODS

Here we report a large British family (DRC255) comprising ten affected individuals in three successive generations (fig 1) with a clinical FTLD phenotype and a genetically confirmed PGRN mutation in two affected individuals. Five of the family members were seen either in the Specialist Cognitive Disorders Clinic of the National Hospital for Neurology and Neurosurgery, London, UK or as part of a longitudinal study of patients ‘at-risk’ for developing FTLD. We describe clinical, neuropsychological, radiological, neuropathological and genetic findings in this family.

Figure 1.

Pedigree of the DRC255 family. Circles represent females and squares males. Diamonds are anonymized individuals. Filled symbols represent affected family members.

RESULTS

Clinical features

The average age at onset of disease in the family is 57.8 years (range 54 to 67) and mean length of clinical history from onset of symptoms to death is 6.1 years (range 2 to 11). All affected family members developed a clinical syndrome in keeping with FTLD.

Historical cases

Little information is available about family members from earlier generations. Case I.1 died in his 50’s with a diagnosis of presenile dementia. Case II.1 died aged 56 and was said to have both Pick’s disease and Parkinson’s disease on clinical grounds at the time of death. Case II.4 was noted to have word-finding difficulties aged 56. He deteriorated cognitively over the next two years with apraxia a prominent feature, dying aged 58. Case II.7 had a diagnosis of presenile dementia from the age of 54 and died aged 58. Case II.8 developed a marked personality change in her mid 50’s becoming initially socially withdrawn and apathetic. Behavioural problems progressed and she became mute prior to her death aged 66.

Detailed case descriptions

Detailed clinical, neuropsychological and brain imaging data are available for five affected individuals from the kindred and their case descriptions are presented below.

Case III.2

This right-handed man developed behavioural symptoms and personality change at the age of 65. He initially became socially withdrawn and apathetic. Around the same time his wife also noticed that he had lost empathy: he remained uncharacteristically indifferent after she had fractured her arm. He developed a sweet tooth, hypersomnia and his conversational speech diminished. Around a year into the illness he developed route-finding problems and became disinhibited, making sexual advances to strangers. Eighteen months after symptom onset he had difficulty putting clothes on correctly and was unable to perform simple calculations. He was first assessed two years into the illness and at this time Mini-Mental State Examination (MMSE) score was 23/30 (10). Spontaneous speech was reduced without evidence of aphasia. Neurological examination was normal apart from brisk facial reflexes and mild bilateral ideomotor limb apraxia. A clinical diagnosis of FTD was made. The patient continued to deteriorate and died aged 72.

Case III.3

This man (handedness uncertain) developed behavioural change aged 56. He became apathetic and hypersomnolent, stopped caring for himself and displayed aggression towards family members. Around the same time his speech decreased in quantity with increasing use of ‘stock phrases’. He was first assessed two years into the illness, at which time there was widespread cognitive impairment. Speech was reduced in quantity but without evidence of aphasia. Apart from a pout reflex, neurological examination was unremarkable. A clinical diagnosis of FTD was made. Behavioural and speech problems progressed and were accompanied by difficulties with episodic memory. Four years into the illness he was noted to be mute with generalized rigidity. He died aged 64.

Case III.4

This left-handed man developed behavioural disturbance and language impairment aged 56. He became increasingly apathetic and aggressive, and the quantity of propositional speech diminished. Over the next year, he developed inappropriate social behavior, sweet tooth and hyperphagia. Two years into the illness there were difficulties with episodic and topographical memory. The legibility of his writing deteriorated and he had increasing difficulty dressing and with calculation. He was first assessed two and half years into the illness when MMSE score was 10/30. He had little spontaneous speech and was echolalic with verbal perseverations, but there were no phonemic, grammatical or semantic errors. Neurological examination revealed severe bilateral ideomotor and ideational limb apraxia. Parkinsonian features were present with bradykinesia, rigidity and postural tremor of both upper limbs, slightly more marked on the left. There was a supranuclear gaze palsy affecting upgaze. A clinical diagnosis of FTD was made. The patient died aged 61.

Case III.9

This right-handed woman presented at the age of 58 with increasing difficulty using her hands over six months such that she became unable to dress herself, switch on the television or put on spectacles. She lost the ability to write, had problems with calculation and her speech output became effortful with a stammer and increasing difficulty finding words. Over the same period she became mildly apathetic and developed a sweet tooth. On examination MMSE score was 15/30 with evidence of anomia. Neurological examination revealed marked bilateral ideomotor and ideational limb apraxia and a predominantly left-sided asymmetrical extrapyramidal syndrome with tremor, rigidity and myoclonus. The tendon reflexes were symmetrically brisk with flexor plantar responses. Eye movements were abnormal with hypometric saccades and jerky pursuit. The clinical diagnosis was FTD with features of both CBD syndrome and non-fluent aphasia. The patient’s condition deteriorated over the next two years such that she became immobile, mute, unable to swallow and doubly incontinent. She died aged 63.

Patient III.11

This right-handed woman first developed problems with word-finding difficulty at the age of 67. On examination, six months after symptom onset, MMSE score was 28/30 and there was anomia. The cognitive and neurological examination was otherwise normal. When assessed a year later, anomia had progressed and she had developed phonemic paraphasias, agrammatism and difficulty repeating polysyllabic words. Her family reported that she had become apathetic and socially withdrawn. MMSE score had fallen to 22/30. There was mild bilateral ideomotor and ideational limb apraxia but the neurological examination was otherwise normal. The clinical diagnosis was PNFA.

Summary of clinical features

The clinical features in this series fall within the wide phenotypic spectrum described previously for sporadic and familial FTLD (1,11). Considering the affected individuals as a group (Table 1), the most consistent features are impaired language output, a behavioural syndrome with early and prominent apathy and limb apraxia. Language impairment consisted of non-fluent aphasia in some cases (III.9 and III.11) and reduced quantity of spontaneous speech in others (III.2, III.3 and III.4). Parkinsonian and other extrapyramidal features emerged in three of the cases. None of the patients had bulbar or limb features suggestive of motor neuron disease. Hypersomnia, an infrequent feature in most FTLD series (12,13) was prominent in two cases.

Table 1. Clinical features in affected individuals.

Case	III.2	III.3	III.4	III.9	III.11
Gender, handedness	M, R	M,NK	M, L	F, R	F, R
Age of onset	65	56	56	58	67
Length of history (onset to death)	7	8	5	5	N/A
First symptoms	Apathy Decreased speech Loss of empathy Sweet tooth Hypersomnia	Apathy Decreased speech Aggression Hypersomnia	Apathy Decreased speech Aggression Sweet tooth	Limb apraxia Dyscalculia Stuttering Anomia Apathy	Anomia Apathy
Behavioural features
Apathy	+	+	+	+	+
Loss of empathy	+	-	+	-	-
Disinhibition	+	-	+	-	-
Poor judgment	+	+	+	-	-
Aggression	+	+	+	-	-
Impaired insight	+	+	+	+	+
Sweet tooth	+	-	+	+	-
Hypersomnia	+	+	-	-	-
Language features
Decreased quantity of speech	+	+	+	+	+
Echolalia	-	-	+	-	-
Hesitant or stuttering speech	-	-	-	+	-
Anomia	-	-	-	+	+
Agrammatism	-	-	-	-	+
Phonemic paraphasias	-	-	-	-	+
Mutism (years to mutism)	NK	+ (4)	+ (4)	+ (2)	-
Other clinical findings
Limb apraxia	mild	NK	mild	severe	mild
Ideational	-	NK	+	+	+
Ideomotor	+	NK	+	+	+
Parkinsonism	-	+ (late)	+ (late)	+ (early)	-
Symmetry	-	R=L	L>R	L>R	-
Rigidity	-	+	+	+	-
Bradykinesia	-	-	+	-	-
Tremor	-	-	+	+	-
Pyramidal signs	-	-	-	+	-
Abnormal eye movements	-	-	+	+	-
Clinical diagnosis	FTD	FTD	FTD	CBD/PNFA	PNFA

NK =Not known

Neuropsychological findings

Four patients were assessed at their initial presentation (Table 2). Both episodic memory and executive function were impaired in cases III.2, III.4 and III.9. Case III.11 scored within the normal range for memory but showed deterioration in recognition memory for words from above the 50^th percentile at diagnosis to between the 10^th and 25^th percentile one year later, and also subsequently developed reduced verbal fluency. Naming was relatively preserved in cases III.2 and III.4 but impaired in cases III.9 and III.11 (14). These latter two cases also had impaired word repetition consistent with non-fluent aphasia. Simple calculation was impaired in three cases (III.2, III.4 and III.9) and perceptual and spatial skills were impaired in two cases (III.2 and III.9).

Table 2. Summary of neuropsychological findings at presentation.

Case	III.2	III.4	III.9	III.11
Duration since symptom onset (years)	2	2	1	1
General intellectual function
Verbal IQ	84	61***	72**	85
Performance IQ	72**	<54***	60***	108
Memory
Verbal memorya	32/50**	fail	21/25	38/50
Visual memoryb	29/50***	fail	14/25***	44/50
Language
Namingc	16/30	13/30*	8/30***	3/30***
Word comprehensiond	36/50	NT	NT	33/50
Word repetition	normal	normal	impaired	impaired
Other cognitive domains
Spellinge	14/30	NT	2/30***	17/30
Calculationf	1/24***	0/24***	0/24***	14/24
Visuospatial/perceptual skillsg	<5th percentile	25-50th percentile	<5th percentile	75-90th percentile
Limb praxis	mild impairment	severe impairment	severe impairment	mild impairment
Executive functionh, i	fail	fail	fail	normal

^aRecognition Memory Test for Words (out of 50) (35) [short version - out of 25]

^bRecognition Memory Test for Faces (out of 50) (35) [short version - out of 25]

^cGraded Naming Test (14)

^dWarrington Synonyms Test (out of 50) (36)

^eGraded Difficulty Spelling Test (37)

^fGraded Difficulty Arithmetic Test (out of 24) (38)

^gVisual Object and Spatial Perception (VOSP) battery (39)

^hWeigl test (40)

ⁱModified card sorting test (41)

^***<1^st percentile

^**=1^st to < 5^th percentile

^*=5^th to <10^th percentile; no stars ≥10^th percentile

NT = not tested

The neuropsychometric profile of these cases, taken together, is noteworthy for the relatively early involvement of parietal lobe (calculation, visuoperceptual and visuospatial) functions. In contrast to features such as executive dysfunction, anomia and memory impairment, which are observed in a broad spectrum of FTLD cases, including familial cases with mutations in the tau gene (15), parietal lobe features are unusual in FTLD. Coupled with the high frequency of limb apraxia, this pattern argues for a more posterior extension of disease associated with this PGRN mutation than would be regarded as typical in FTLD.

Brain imaging findings

Four patients (cases III.2, III.4, III.9 and III.11) underwent brain imaging. Findings are summarized in Table 3 and representative magnetic resonance (MR) images are presented in Figure 2. Brain imaging findings show considerable individual variation in this series but atrophy involving the frontal, temporal and parietal lobes was present in all cases. Cerebral atrophy was strikingly asymmetrical in cases III.2, III.4 (both predominantly right-sided) and III.11 (predominantly left-sided).

Table 3. Summary of brain imaging findings.

Case	Duration since symptom onset (years)	Modality	Key findings
III.2	2	MRI	Marked asymmetrical frontal, temporal and parietal lobe atrophy predominantly affecting the right side. In the frontal lobes there is greater atrophy affecting the inferior frontal regions than the frontal convexities, although the convexity sulci are wider bilaterally. There is asymmetrical temporal lobe atrophy, with the whole right temporal lobe affected more than the left. There is also widespread right-sided parietal atrophy.
III.4	1	CT	Asymmetrical generalized atrophy, affecting the right hemisphere more than the left.
III.9	1	MRI	Mild diffuse atrophy with increased prominence of the parietal sulci.
III.11	1	MRI	Asymmetrical frontal, temporal and parietal lobe atrophy predominantly affecting the left side. In the frontal lobes there is atrophy of the medial superior frontal and frontopolar regions, and involvement of the anterior cingulate gyrus. There is marked atrophy of the left temporal pole. The left superior temporal, middle and inferior temporal and fusiform gyri are significantly affected, with some atrophy of left amygdala and hippocampus. In the parietal lobes there is relatively selective atrophy of the left angular gyrus posteriorly.

Figure 2.

Representative brain images in affected individuals (III.2 and III.11)

Figs 2a,2b: Coronal T1 MR images showing asymmetrical right-sided fronto-temporo-parietal atrophy (III.2)

Figs 2c, 2d: Coronal T1 MR images showing asymmetrical left-sided fronto-temporo-parietal atrophy (III.11)

Neuropathological findings

Three members of the kindred (cases III.2, III.3 and III.4) have come to post-mortem examination with similar appearances seen in all cases. Macroscopically, all showed severe bilateral atrophy of the frontal and temporal lobes with moderate atrophy of the parietal lobes and relative sparing of the occipital lobes. Histological investigation showed superficial spongiosis, nerve cell loss and gliosis of the frontal, temporal and parietal cortices. There were numerous ubiquitin-positive neurites and neuronal cytoplasmic inclusions (NCI) in the superficial cortical laminae and striatum. Scattered ubiquitin-positive neuronal intranuclear inclusions (NII) were found, which were lentiform or round in shape. The dentate fascia granule cells also contained scattered NCI, which often had a granular appearance, as well as occasional NII. Figure 4 shows the findings in case III.2. All three cases showed features of type 1 FTLD-U according to the classification described by Mackenzie et al. (16, 17) which corresponds to previously reported cases with PGRN mutations (18). Of note, the neurites, NCI and NII also stained for TDP-43 which has recently been described to be a component of the ubiquitin-positive neurites and inclusions in FTLD-U (17).

Figure 4.

Histological features in case III.2. A = dentate fascia ubiquitin-positive inclusions; B = ubiquitin-positive neuronal cytoplasmic inclusions (NCI) and neurites in the frontal cortex; C & D = ubiquitin-positive neuronal intranuclear inclusions (NII) in the frontal cortex. The NII on C is of the lentiform type.

Genetic analyses

All 13 exons of the PGRN gene were sequenced in cases III.3 and III.11 in at least one direction. Analysis of the electropherogram traces revealed a c.90_91insCTGC mutation in the first coding exon on the reverse complement strand in both cases (figure 3 shows the electropherogram for III.3), which would be predicted to cause a frameshift and premature termination (C31LfsX35). This mutation was originally described in the UBC17 family (7,19). Analysis of four microsatellites near to PGRN in patient III.3 identified rare alleles linked to the UBC17 mutation (17), suggesting a common ancestry.

Figure 3.

Electropherogram traces from the first coding exon of PGRN in the reverse complement. (A) healthy population control DNA (B) patient III.3. Arrow indicates the presence of a c.90_91insCTGC mutation.

COMMENT

Here we describe the phenotype of a large family with autosomal dominant TDP-43-positive FTLD-U due to a PGRN mutation. Common clinical features include language output impairment leading to mutism, behavioural disturbance with early prominent apathy, and extrapyramidal features. In addition to these features, which are commonly associated with FTLD, we propose that parietal deficits (dyscalculia, visuospatial/perceptual dysfunction, and/or limb apraxia), which are infrequently described in FTLD, may also be a feature of the progranulin phenotype. Our case III.9 was diagnosed during life with a CBD syndrome in which parietal dysfunction is well recognized (20) and which is pathologically heterogeneous (21). Indeed, the CBD syndrome has been described previously in association with a PGRN mutation (9). However, it is notable that all of the cases here had evidence of parietal dysfunction. Neuropsychometry may expose parietal lobe deficits that might otherwise be overlooked.

We suggest that language and behavioural features may have greater value in differentiating these cases from other causes of FTLD when accompanied by parietal deficits. Furthermore, the specific features of the speech or behavioural syndrome may also have diagnostic potential. All five of the patients described in detail in family DRC255 had impairment of language output. Cases III.9 and III.11 had features consistent with PNFA (1). In contrast, cases III.2, III.3 and III.4 exhibited a decrease in quantity of spontaneous speech in the absence of semantic, grammatical or phonemic errors and with relatively intact naming and verbal comprehension. This would be in keeping with dynamic aphasia (22) rather than the PNFA or SD subtypes of FTLD (1). Our cases suggest that the common endpoint for both the PNFA and the dynamic aphasia presentations is mutism. Behavioral features were early and prominent in cases III.2, III.3 and III.4 and less prominent in the other two cases. However, all cases developed apathy as the earliest and most salient behavioral feature.

How does our family (DRC255) compare clinically with previously described families with PGRN mutations (1083: 23, DR2-DR8 Belgian founder family: 24, HDDD2: 25,26, HFTD3: 27,28, F53 and F337: 29, and UBC17 which appears to be the same family as DRC255: 17) and with other genetically mediated causes of FTLD, notably mutations in the tau gene? The most commonly described features in patients with PGRN mutations are language output impairment with features suggestive of either dynamic aphasia or non-fluent aphasia and a behavioural syndrome often characterized by apathy. Published data on the clinical phenomenology of patients with PGRN mutations are limited, however parietal lobe features have been described in many families (19,25, 27, 29). These clinical observations are further supported by histopathological evidence of significant parietal lobe involvement in association with PGRN mutations (19,25,26,29). Tau mutations are associated with wide phenotypic variation in published series (30,31). However, in detailed studies of patients with a tau 10+16 mutation (15), disinhibition rather than apathy dominated the behavioural syndrome and while ‘language deficits’ occurred in these patients, features of a non-fluent aphasia were not reported and language impairment was not a prominent feature. Of particular note is that parietal features are generally not described. Parkinsonism and other extrapyramidal features appear to occur commonly in association with both PGRN and tau mutations (19,29,30,31).

The cases described in the present series showed variability in the pattern of radiological atrophy, but it is notable that each of the cases with detailed MR imaging showed frontotemporal atrophy with associated involvement of the parietal lobes. Longitudinal structural imaging in one patient with a PGRN mutation revealed progressive asymmetrical fronto-temporo-parietal atrophy more marked on the right (32), while a recent study specifically comparing PGRN cases with ubiquitin-positive PGRN-negative cases showed that PGRN-positive cases had greater gray matter loss in the parietal lobes (as well as the frontal lobes) (33). These findings combined with the results of this study support the hypothesis that more posterior extension of atrophy to involve the parietal lobes may be a radiological marker for PGRN-associated disease, consistent with the clinical and neuropsychological profile. The involvement of the parietal lobe as well as more anterior areas suggests that PGRN mutations may disrupt functional pathways linking the frontal and parietal lobes, for which anatomical substrates exist in the human brain (34). Preferential disease spread within hemispheric pathways (rather than between hemispheres) would also be in keeping with the striking asymmetry seen early in the disease both in our own and other cases reported in the literature.

Conclusions

The mutation in the PGRN gene in this family produces a clinico-radiological phenotype that overlaps substantially with the spectrum of FTLD due to other sporadic and genetically mediated pathologies. However, certain clinical and radiological features common to a majority of affected individuals suggest that there may be a core phenotype of progranulin-associated disease. In particular, parietal lobe deficits are a salient feature of the progranulin phenotype. Future work will be directed toward testing these observations in other affected families, assessment of their specificity and predictive value, and elucidation of the mechanisms by which PGRN mutations produce their phenotypic effects.