Abstract
Background: Mutations in the progranulin gene (PGRN) were identified as the causal mechanism underlying frontotemporal lobar degeneration (FTLD). Most of these mutations are predicted to create null alleles leading to a 50% loss of progranulin transcript.
Methods: Patients underwent clinical and neurologic examination at the Memory Clinic of the IRCCS S. Giovanni di Dio-Fatebenefratelli, Brescia, Italy. We enrolled affected (n = 6) and unaffected at risk members (n = 73) of families carrying the FTLD associated progranulin Leu271LeufsX10 mutation; additionally, we included subjects affected by sporadic/familial FTLD (n = 65), controls (n = 75), and a family carrying the tau P301L mutation. The presence of mutations in PGRN and MAPT genes was investigated by direct sequencing of exonic and flanking intronic regions. Progranulin plasma and CSF levels were measured using ELISA.
Results: We demonstrated that progranulin protein is strongly reduced (up to 3.93-fold) both in plasma and CSF of affected and unaffected subjects carrying mutations in progranulin gene (PGRN Leu271LeufsX10 and Q341X). We established a plasma progranulin protein cutoff level of 74.4 ng/mL that identifies, with specificity and sensitivity of 100%, mutation carriers among unaffected subjects. In FTLD, values ≤110.9 ng/mL give a specificity of 92.8% and a sensitivity of 100% for PGRN mutations.
Conclusions: We propose the dosage of plasma progranulin as a useful tool for a quick and inexpensive large-scale screening of carriers of progranulin mutations and for monitoring future treatments that might boost the level of this protein.
GLOSSARY: FTD = frontotemporal dementia; FTLD = frontotemporal lobar degeneration; PPA = primary nonfluent aphasia.
Frontotemporal lobar degeneration (FTLD) is the third most common cause of neurodegenerative dementia after AD and dementia with Lewy bodies.1 The main clinical subtypes of FTLD are frontotemporal dementia (FTD), in which social and executive dysfunction predominate, semantic dementia, and progressive primary nonfluent aphasia (PPA) with prominent early language disturbances.2,3 Additionally, FTLD is closely related both clinically and pathologically to corticobasal degeneration and progressive supranuclear palsy. Up to 40% of FTLD cases have a positive family history of dementia, thus indicating an important genetic contribution to the etiology of this disease.4 Mutations in the progranulin gene (PGRN; MIM 138945) were identified as a causal mechanism underlying FTLD.5,6 Progranulin (PGRN) is a 593 amino acid secreted glycoprotein composed of 7.5 tandem repeats of highly conserved motifs of 12 cysteines, which can be proteolytically cleaved to form a family of 6-kDa peptides called granulins7; PGRN is a growth factor involved in the regulation of multiple processes including tumorigenesis, wound repair, development, and inflammation.8,9 To date, 47 pathogenic PGRN mutations have been identified in 111 families or individuals (AD&FTD Mutation database: http://molgen.ua.ac.be/FTDmutations). Most of these mutations are predicted to result in mutant transcript that is rapidly degraded: thus, these mutations create null alleles leading to a 50% loss of progranulin mRNA.5,6,10–13 This suggests that FTLD results from progranulin haploinsufficiency rather than the accumulation of mutant protein. We previously described, in Italian pedigrees, a mutation in PGRN gene (Leu271LeufsX10) associated with highly variable clinical phenotypes.14 The aim of this study is to investigate the impact of PGRN mutations on plasma and CSF levels of PGRN in affected and unaffected carriers.
METHODS
Participants.
Patients underwent clinical and neurologic examination at the Memory Clinic of the IRCCS Centro San Giovanni di Dio-Fatebenefratelli, Brescia, Italy. Clinical diagnosis of patients was made according to international guidelines2,15: patient diagnosis was derived by a multidisciplinary team of neurologists, neuropsychologists, psychiatrists, and nurses, who performed extensive behavioral, neuropsychological, and neuroimaging assessments. The family history was determined by collection of the Family History Questionnaire.16 Patients affected by familial dementia had two or more first-degree relatives who had a neurologic disorder, fitting the clinical criteria for familial FTLD. Starting from nuclear families, three large pedigrees were drawn by contacting as many as possible affected and unaffected members from families carrying the progranulin Leu271LeufsX10 mutation14 (affected n = 6; unaffected n = 73) and subjects belonging to a family carrying the tau P301L mutation17(affected n = 1; unaffected n = 2). Additionally, we included in the present study subjects affected by sporadic/familial FLTD (positive family history n = 23; negative family history n = 42) and healthy controls (n = 75). Demographic and clinical features of participants are reported in the table. Biologic samples were collected after obtaining informed consensus, as approved by the local ethical committee (CEIOC, Brescia, Italy; 46/2004). DNA, CSF, and plasma were isolated according to standard procedures. Plasma was available for all enrolled subjects; CSF was obtained from a subgroup of 5 controls, 17 patients with sporadic/familial FLTD, and 5 patients with FTLD carrying the progranulin Leu271LeufsX10 mutation.
Table Demographic and clinical features of subjects enrolled for plasma analysis
Genetic and biochemical analyses.
PGRN and MAPT (MIM 157140) genes were analyzed by direct sequencing of exonic and flanking intronic regions as previously described.14,17 Validation of new mutations was performed screening 120 aged control individuals sequencing the relevant exons. Plasma and CSF levels of PGRN were measured, in duplicate, using an ELISA kit (Human Progranulin ELISA Kit, AdipoGen Inc., Seoul, Korea). All tests for each sample were performed with ELISA kit of the same lot by technicians who were unaware of whether the sample belonged to cases or to controls. Data were analyzed with use of the statistical software SPSS 13.0 (SPSS Inc., Chicago, IL).
RESULTS
Mutation analysis.
We previously described Italian pedigrees carrying the progranulin Leu271LeufsX10 mutation: for the present biochemical study, six patients with the mutation were enrolled. Sequencing analysis of 73 unaffected subjects belonging to these pedigrees revealed the PGRN Leu271LeufsX10 mutation in 22 unaffected relatives (unaffected mutated subjects n = 22; unaffected non mutated subjects n = 51). This mutation, causing a frameshift at codon 271, introduces a premature termination codon after a read through of 10 residues: the resultant mutant protein is predicted to be 280 amino acids in length, instead of the 593 residues of the wild type progranulin. In two patients belonging to a fourth pedigree, we identified a novel heterozygous point mutation in exon 9 (c.1021 C>T), introducing a premature termination codon (CAG>TAG) at residue 341 (Q341X). The resultant mutant protein is predicted to be 340 amino acids in length. The Q341X mutation was absent in 120 aged cognitively intact controls (mean age = 75.68 ± 5.25; MMSE = 28.36 ± 1.80). We also included a previously described pedigree carrying the tau P301L mutation17: patients with FTLD were screened for mutations in MAPT and PGRN genes.
Biochemical study.
Levels of progranulin protein were different among the study groups (p < 0.001, Kruskal Wallis test) (figure 1A). We found that PGRN is strongly reduced in both affected and nonaffected carriers of PGRN Leu271LeufsX10 mutation: considering the group of unaffected family relatives, we observed a 3.93-fold decrease of plasma progranulin levels in those subjects carrying the genetic defect (p < 0.001) (figure 1A). The ROC analysis fixed the best result at the PGRN cutoff level of 74.4 ng/mL: values ≤74.4 give a specificity and a sensitivity of 100% (figure 1B). These data were further confirmed in affected patients with FTLD: affected carriers of PGRN Leu271LeufsX10 mutation showed a 3.21-fold decrease of progranulin with respect to PGRN (−) patients with FTLD (p < 0.001) (figure 1A). Interestingly, the striking relationship between progranulin mutation and reduction of genetic product was also observed in patients carrying the new PGRN Q341X mutation, where the peripheral plasma levels of progranulin were 3.62-fold decreased with respect to PGRN (−) patients with FTLD (p < 0.01). The ROC analysis fixed the best result at the PGRN cutoff level of 110.9 ng/mL: values ≤110.9 give a specificity of 92.8% and a sensitivity of 100% (figure 1C). Tau P301L mutation did not influence progranulin protein plasma levels (unaffected non mutated relative: 172.2 ng/mL; unaffected tau P301L relative: 214.4 ng/mL). We tested, in all controls (unrelated control subjects + unaffected family members belonging to PGRN Leu271LeufsX10 pedigrees, n = 126), a possible correlation between age and progranulin levels: we found a correlation of PGRN with age (r = 0.227, p = 0.011, Pearson). Analyzing the distribution of progranulin in this group stratified by age, there was a trend for increasing levels of progranulin with age (mean values ± SD: age ≤ 50, 155.0 ± 40.6, n = 29; age 50–66, 189.9 ± 58.6, n = 68; age > 66, 199.4 ± 65.7, n = 29, p = 0.01, Kruskal Wallis test). Levels of PGRN were also tested in CSF (figure 2A). PGRN (+) patients with FTLD showed a 2.35-fold decrease of progranulin with respect to PGRN (−) patients with FTLD (p < 0.001). Protein levels in PGRN (−) patients with FTLD and controls were comparable. The ROC analysis fixed the best result at the PGRN cutoff level of 5.18 ng/mL: values ≤ 5.18 give a specificity and a sensitivity of 100 (figure 2B).
Figure 1 Progranulin protein plasma dosage
(A) Plasma progranulin protein levels (ng/mL) in the study groups: unaffected nonmutated relatives PGRN (−) (n = 51); unaffected PGRN (+) Leu271LeufsX10 relatives (n = 22); affected unrelated PGRN (−) patients with FTLD (n = 63); affected PGRN (+) Leu271LeufsX10 patients with FTLD (n = 6); affected PGRN (+) Q341X patients with FTLD (n = 2); control subjects (n = 75). *Plasma progranulin levels are decreased in unaffected subjects carrying the genetic defect (mean values ± SD: unaffected nonmutated relatives 167.1 ± 51.8 ng/mL; unaffected PGRN [+] Leu271LeufsX10 relatives 42.5 ± 11.7 ng/mL, p < 0.001 analysis of variance [ANOVA], Sidak post hoc test). **Plasma progranulin levels are decreased in affected carriers of PGRN Leu271LeufsX10 mutation with respect to PGRN (−) patients with FTLD (mean values ± SD: affected PGRN [−] patients 197.9 ± 67.6 ng/mL; affected PGRN [+] Leu271LeufsX10 patients 61.6 ± 25.9; p < 0.001). ***Plasma progranulin levels are decreased in affected carriers of PGRN Q341X mutation with respect to PGRN (−) patients with FTLD (mean values ± SD: affected PGRN [+] Q341X patients 54.6 ± 17.5 ng/mL, p < 0.01, ANOVA, Sidak post hoc test). (B) Scatterplot of the plasma progranulin values in unaffected subjects: a cutoff level of 74.4 ng/mL separates PGRN (+) subjects (filled circles) from PGRN (−) relatives (empty circles) with a specificity and a sensitivity of 100%. (C) Scatterplot of the plasma progranulin values in affected patients with FTLD and control subjects: a cutoff level of 110.9 ng/mL separates PGRN (+) patients (filled circles: Leu271LeufsX10 mutation carriers; filled triangles: Q341X mutation carriers) from PGRN (−) patients (empty circles) and control subjects (empty triangles) with a specificity of 92.8% and a sensitivity of 100%.
Figure 2 Progranulin protein CSF dosage
(A) CSF progranulin protein levels (ng/mL) in affected unrelated PGRN (−) patients with FTLD (n = 17); affected PGRN (+) Leu271LeufsX10 patients with FTLD (n = 5); control subjects (n = 5). *PGRN (+) patients with FTLD showed a decrease of CSF progranulin with respect to PGRN (−) patients with FTLD (mean values ± SD: affected PGRN [−] patients 8.7 ± 2.1 ng/mL; affected PGRN [+] patients 3.7 ± 0.6 ng/mL, p < 0.001 analysis of variance [ANOVA], Sidak post hoc test). Protein levels in PGRN (−) patients with FTLD and controls were comparable (mean values ± SD: controls, 8.2 ± 2.7 ng/mL, p = 0.94 ANOVA, Sidak post hoc test). (B) Scatterplot of the CSF progranulin values in affected patients with FTLD and control subjects: a cutoff level of 5.18 ng/mL separates PGRN (+) patients (filled circles: Leu271LeufsX10 mutation carriers) from PGRN (−) patients (empty circles) and control subjects (empty triangles) with a specificity and a sensitivity of 100%.
DISCUSSION
With the discovery of the PGRN gene, the heterogeneous world of FTLD has taken a giant step forward. Recent work13,18 proposed that mutations other than null mutation produce a pathogenic loss of progranulin. This suggests that, in PGRN (+) cases, FTD results from PGRN haploinsufficiency rather than the accumulation of mutant protein. The PGRN Leu271LeufsX10 and Q341X mutations introduce premature termination codons that fulfill the general requirements for the induction of the nonsense-mediated decay.19 We demonstrated, in a large group of affected and unaffected subjects, that progranulin protein is strongly reduced both in plasma and CSF of carriers of genetic defect. This biologic phenotype is tightly associated with mutations in PGRN: PGRN (−) FTLD cases, including tau P301L carriers, were comparable to controls. Thus, in our clinical series, PGRN cannot be considered a biologic marker for PGRN (−) FTLD. The functions and mechanism of PGRN in the CNS are largely speculative and based on more established evidence from the periphery. PGRN expression in activated microglial cells suggests a possible role of this trophic factor in neuroprotection and inflammatory responses associated with neurodegeneration.9 PGRN seems to be modulated during normal aging process since, in controls, we observed an age-related increase of its protein. Unaffected relatives carrying PGRN mutations have small quantity of circulating protein (25% with respect to controls): thus, during lifetime, they partially lack mechanisms related to the PGRN physiologic role including neuroprotection. Of note, the PGRN detectable both in the central and peripheral compartments is less than the 50% expected by haploinsufficiency mechanism. A possible explanation could be that mutant mRNA degradation partially affects also wild type mRNA, thus amplifying the null allele effect. Alternatively, we cannot exclude an unbalanced progranulin metabolism. Herein, we propose the dosage of plasma progranulin as a useful tool for a quick and cheap large-scale screening of carriers of genetic PGRN defects. The strong phenotypic variability—ranging from PPA to amyotrophic lateral sclerosis20—associated with PGRN mutations often makes it difficult to trace inheritance for a specific neurodegenerative disease within families that consequently escape genetic studies. PGRN dosage might capture, at very high specificity and sensitivity (and low costs), also those cases, thus better defining the phenotypical features of “progranulopathies.” Eventually, the level of progranulin in plasma could be a useful marker both for early identification of at risk asymptomatic subjects and for monitoring future treatments that might boost the level of this protein.
ACKNOWLEDGMENT
The authors thank the families for their collaboration.
Footnotes
-
e-Pub ahead of print on September 3, 2008, at www.neurology.org.
Supported by grants RC2007, 2005/agreement number PS-NEURO ex56/05/4, and 2004/agreement number 533/F/B 1, Italian Ministry of Health.
Disclosure: The authors report no disclosures.
neurobiologia{at}fatebenefratelli.it
Received February 25, 2008. Accepted in final form May 27, 2008.
REFERENCES
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.