Abstract

Frontotemporal lobar degeneration (FTLD) and amyotrophic lateral sclerosis (ALS) are overlapping neurodegenerative disorders. Mutations in the growth factor progranulin (PGRN) gene cause FTLD, sometimes in conjunction with ALS; such mutations are also observed in some ALS patients. Most PGRN mutations underlying FTLD are null mutations that result in reduced PGRN levels. We investigated PGRN expression in human ALS and in mouse models of motor neuron degeneration. Progranulin plasma or CSF levels in newly diagnosed ALS patients did not differ from those in healthy or disease controls (PGRN mutation-negative FTLD and Alzheimer disease patients). In the mutant SOD1G93A mouse model of ALS, spinal cord PGRN levels were normal in presymptomatic animals but increased during the degenerative process. This increase in PGRN correlated with enhanced expression of PGRN in microglia. In CSF, PGRN levels were normal in presymptomatic and early symptomatic animals, but with disease progression, a raise in PGRN was detectable. These data indicate that upregulation of PGRN is a marker of the microglial response that occurs with progression in motor neuron diseases.

Introduction

Progranulin (PGRN) is a multifunctional growth factor that contains a signal peptide and 7.5 granulin domains that are highly conserved throughout evolution (1–3). The protein is widely expressed; it is secreted and is then proteolytically cleaved into granulins (GRN A–F and paragranulin) by elastase in the extracellular matrix (4, 5). Progranulin has been implicated in development, tumor growth, inflammation, and wound healing (6, 7).

Although no PGRN receptor has been identified to date, PGRN stimulates the mitogen-activated protein kinase and the phosphatidylinositol 3-kinase pathways (8). The full-length precursor and some of the GRN peptides act as autocrine growth factors for different cell lines (9), but some of the GRN peptides inhibit proliferation in other cell lines (3). Therefore, the PGRN gene encodes a family of proteins (i.e. the full-length precursor, 7 GRN peptides, paragranulin, and several intermediates), which likely exert different actions on different targets and receptors (6).

Although PGRN has been studied most extensively in tumor growth, inflammation, and wound healing, dominant mutations in the PGRN gene were recently identified as a cause of familial frontotemporal lobar degeneration (FTLD), (10, 11) sometimes in conjunction with amyotrophic lateral sclerosis (ALS) (12, 13), a neurodegenerative disorder linked to FTLD (14–17). Most disease-causing mutations are loss-of-function mutations that result in reduced PGRN messenger RNA (mRNA) and/or protein levels in all tissues studied to date (10, 11, 18–23). Therefore, haploinsufficiency with shortage of PGRN is thought to be the disease-causing mechanism. In line with this hypothesis, we recently observed a survival-promoting effect of PGRN on cultured cortical and motor neurons (19).

Missense mutations, some of unproven pathogenic nature, have also been observed in patients with FTLD (24–26), Alzheimer disease (AD) (27), and in some patients with ALS (28–30). In addition, PGRN has been found to be a genetic modifier of the course of sporadic ALS (29).

Apart from reductions in PGRN observed in PGRN mutations carriers, upregulation of PGRN seems to be part of the inflammatory response after acute (31–33) or chronic insults (10, 34–40) in the CNS.

Whether changes in PGRN expression levels also occur in ALS is currently unknown. In this study, the expression of PGRN was studied in ALS patients and in mouse models of motor neuron degeneration.

Materials and Methods

Patient Samples

All human plasma and cerebrospinal fluid (CSF) samples were obtained with informed consent and approved by the local ethical committee. Patient samples were collected at the neuromuscular and memory clinic of the University Hospital Leuven (Belgium) and of the Radboud University Nijmegen Medical Center (the Netherlands). The diagnoses of ALS, FTLD, or AD were made based on published criteria (41–43). For most patients, there was no postmortem study to confirm the diagnosis. Patient demographic data for the CSF study are given in Table 1. Patient data for the study of plasma PGRN levels are shown in Table 2. Patients were sampled during the diagnostic early phase of the disease. For most patients, the lumbar puncture was planned at the first visit. Cerebrospinal fluid control samples were taken after informed consent from patients who underwent a lumbar puncture for diagnostic reasons and who later were found to have normal CSF results on routine testing. Healthy spouses were used as controls for plasma samples. Platelet-poor plasma samples were obtained as described (44). Cerebrospinal fluid samples were obtained by lumbar puncture, collected in polypropylene tubes, centrifuged, and stored at −80°C until use. Progranulin mutations were excluded by sequencing all exons, exonintron boundaries, and the 5′ UTR, as previously described (11).

TABLE 1

Reduced CSF Progranulin Levels Are Restricted to Patients With PGRN Null Mutations

Patient Group Patients, n Age, Mean ± SD Men, n (%) Diagnostic Delay, Mean ± SD, mo Disease Duration at LP, Mean ± SD, mo PGRN Level, CSF ± SEM, ng/mL p (vs Control) 
Control 31 53.5 ± 18.4 13 (42) NA NA 6.4 ± 0.6  
ALS 29 60.2 ± 12.4 17 (59) 12.2 ± 9.0 13.9 ± 8.9 7.0 ± 0.5 0.99 
FTLD no PGRN mutation 24 61.1 ± 12.2 17 (71) 36.2 ± 22.1 35.7 ± 22.1 6.9 ± 0.6 0.98 
FTLD PGRN mutation — — — — 2.2 ± 1.0 0.04 
FTLD-ALS — — — — 5.5 ± 0.7 0.71 
AD 34 63.4 ± 8.5 16 (47) 29.5 ± 22.1 30.0 ± 22.8 6.0 ± 0.5 0.69 
Patient Group Patients, n Age, Mean ± SD Men, n (%) Diagnostic Delay, Mean ± SD, mo Disease Duration at LP, Mean ± SD, mo PGRN Level, CSF ± SEM, ng/mL p (vs Control) 
Control 31 53.5 ± 18.4 13 (42) NA NA 6.4 ± 0.6  
ALS 29 60.2 ± 12.4 17 (59) 12.2 ± 9.0 13.9 ± 8.9 7.0 ± 0.5 0.99 
FTLD no PGRN mutation 24 61.1 ± 12.2 17 (71) 36.2 ± 22.1 35.7 ± 22.1 6.9 ± 0.6 0.98 
FTLD PGRN mutation — — — — 2.2 ± 1.0 0.04 
FTLD-ALS — — — — 5.5 ± 0.7 0.71 
AD 34 63.4 ± 8.5 16 (47) 29.5 ± 22.1 30.0 ± 22.8 6.0 ± 0.5 0.69 

Progranulin (PGRN) levels were measured in CSF from controls and patients with amyotrophic lateral sclerosis (ALS), frontotemporal lobar degeneration (FTLD) without PGRN mutations, FTLD-ALS patients, and Alzheimer disease (AD) patients. For comparison, data on samples from FTLD patients with PGRN null mutation (19) were added. Analysis of variance with post hoc comparison with control was used to calculate significance.

LP, lumbar puncture; n, number; NA, not applicable.

TABLE 2

Demographic Data of Amyotrophic Lateral Sclerosis Patients and Controls Used in Plasma Progranulin Measurements

 ALS Patients Controls 
91 56 
Age, mean ± SD, yr 62 ± 13 58 ± 13 
Men, n (%) 52 (57) 20 (36) 
Diagnostic delay, mean ± SD, mo 10.1 ± 8.0 NA 
Disease duration at sampling, mean ± SD, mo 13.7 ± 7.5 NA 
 ALS Patients Controls 
91 56 
Age, mean ± SD, yr 62 ± 13 58 ± 13 
Men, n (%) 52 (57) 20 (36) 
Diagnostic delay, mean ± SD, mo 10.1 ± 8.0 NA 
Disease duration at sampling, mean ± SD, mo 13.7 ± 7.5 NA 

Progranulin levels were measured in plasma samples from controls and ALS patients.

ALS, amyotrophic lateral sclerosis; NA, not applicable.

Human Progranulin Enzyme-Linked Immunosorbent Assay

Progranulin levels were measured by enzyme-linked immunosorbent assay (ELISA), as previously described (19, 22).

In brief, 96-well plates were coated with monoclonal anti-human PGRN antibody (R&D Systems, Minneapolis, MN), blocked with 1% BSA, before addition of samples and standard, that is recombinant human PGRN (R&D Systems). Bound PGRN was detected using biotinylated polyclonal anti-human PGRN antibody (R&D Systems), ABC complex and orthophenylenediamine. Absorbance was measured at 490 nm on a VictorX3 plate reader (Perkin Elmer, Waltham, MA). The combination of antibodies used in this sandwich ELISA recognizes only the full-length PGRN (19). Progranulin was detectable in plasma and CSF, with average values comparable to other reports (20, 21).

Animal Models With Motor Neuron Degeneration

All experiments on rodents were approved by the local ethical committee of the K.U. Leuven, Belgium. Mutant SOD1G93A mice [B6SJL-TgN(SOD1-G93A)1Gur/J; stock number 002726] were obtained from Jackson Laboratory (Bar Harbor, ME). Homozygous transgenic mice with a deletion in the hypoxia-response element in the vascular endothelial growth factor (VEGF) gene (VEGFδ/δ) (45), and transgenic mice overexpressing human mutant tauP301L driven by the mouse Thy1 gene promoter (46) were previously generated. Heterozygous mutant SOD1G93A mice display an aggressive form of motor neuron degeneration (47) with onset of signs at around 80 days and a progressive disease course with death at around 140 to 150 days. Homozygous VEGFδ/δ mice develop a slowly progressive motor neuron degeneration phenotype that usually starts at around 5 months, and survival is usually beyond 1 year (45). In mutant TauP301L mice, onset of disease is around 8 to 9 months. In homozygous mice, disease is characterized by an aggressive motor phenotype with motor neuron loss in spinal cord and extensive brainstem tau pathology that results in premature death before the age of 12 months (46, 48).

Isolation of Microglia From Spinal Cord

Microglial cells were isolated from adult spinal cord, as previously described (49). Briefly, mice were transcardially perfused with ice-cold PBS, and spinal cords were dissected. The cords were mechanically homogenized and filtered through a 70-μm cell strainer to obtain a single cell suspension. Cells were then loaded on a Percoll gradient (40%–80%) (Percoll; GE Healthcare, Uppsala, Sweden). After centrifugation, mononuclear cells from the interface were isolated and stained with a phycoerythrin-coupled antibody against the myeloid cell marker CD11b (Becton Dickinson, Franklin Lake, NJ). CD11b+ cells were then gated and sorted using a flow cytometer FACS ARIA (Becton Dickinson).

RNA Extraction and Complementary DNA Preparation

RNA from total spinal cord lysates or freshly isolated spinal microglia was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA) and further purified using the RNeasy cleanup protocol with on-column DNase digestion (Qiagen, Valencia, CA). One microgram of RNA was transcribed to complementary DNA using the QuantiTect reverse transcription kit (Qiagen), according to the manufacturer's instructions.

Quantitative Polymerase Chain Reaction

For real-time polymerase chain reaction (PCR) of mouse progranulin, the forward primer was 5′ TTG GGT ACT GGA GTG AAG TGG AT 3′, the reverse primer was 5′ GCC CAG ACG TGT CAT TTC CT 3′, and the probe was FAM-CCT AAC TGA GCT CCC-MGB. Thermal cycling was performed on a 7300 Sequence Detection System (Applied Biosystems, Carlsbad, CA). Each reaction was performed in duplicate. All samples were normalized to the level of β-actin RNA.

Immunohistochemistry

All mice were anesthetized using 10% pentobarbital sodium in PBS. After transcardiac perfusion with ice-cold PBS followed by fixation with 4% paraformaldehyde, spinal cords were dissected and postfixed for 2 hours in 4% para-formaldehyde. Free-floating spinal cord sections (20 μm) were cut on a cryostat (Leica, Wetzlar, Germany) and incubated in PBS with 0.1% Triton X-100 (PBST). Sections were blocked in PBST with 10% normal donkey serum (Sigma, St Louis, MO) for 1 hour. The following primary antibodies were used: sheep polyclonal anti-mouse progranulin (R&D Systems), mouse monoclonal anti-SMI-32 (neurofilament heavy antibody; Sternberger Monoclonals, Baltimore, MD), mouse anti-NeuN (Millipore, Billerica, MA), and rat anti-CD11b (Serotec, Martinsried, Germany). Antibodies were incubated overnight on a shaker at 4°C. After 3 washing steps, sections were incubated with the corresponding secondary antibodies (Alexa Fluor 555 or Alexa Fluor 488; Invitrogen). After 3 washing steps, sections were mounted using Vectashield (Vector Laboratories, Burlingame, CA) (with 4′,6-diamidino-2-phenylindole) and analyzed under a fluorescence microscope (DMIRB; Leica).

Human spinal cord samples were obtained after informed consent from patients who died of clinically confirmed ALS, according to the revised El Escorial criteria (41), or controls (n = 3 per group). The 3 ALS patients studied (2 men and 1 woman) were 49, 47, and 74 years old at the time of death, respectively. The 2 male patients had a spinal form of the disease with onset in lower limbs and the female patient experienced bulbar ALS. They all died of respiratory insufficiency. The postmortem delay was 20, 34, and 25 hours for the 3 samples, respectively. The 3 controls (2 men and 1 woman) were 48, 64, and 60 years old, respectively. The causes of death were cardiac decompensation, liver failure, and respiratory decompensation due to pneumonia, respectively. The postmortem delay for the 3 control spinal cords was 17, 8, and 20 hours, respectively.

Part of the spinal cord was fixed in formalin and embedded in paraffin, and part of the spinal cord was freshly frozen. Cryosections (10 μm) were stained with a goat polyclonal human PGRN antibody (R&D Systems) and a mouse monoclonal CD68 antibody (KP1; Biogenex, San Ramon, CA). Alexa Fluor 555 and Alexa Fluor 488 were used to visualize CD68 (a microglial cell/macrophage marker) and progranulin; 4′,6-diamidino-2-phenylindole was used to visualize the nuclei. In some experiments, serial sections of the cords were alternatively stained with hematoxylin and eosin or with a bio-tinylated anti-human PGRN antibody (R&D Systems). Then, the avidin-biotin complex detection system (Vector Laboratories, Burlingame, CA) and 3,3′-diaminobenzidine (Sigma) as chromogen were used to develop the stain.

Pictures were taken using a DXM 1200 camera (Nikon Instruments, Melville, NY) and analyzed using NIS Elements software (Nikon Instruments). Detailed pictures were taken on a confocal Zeiss 200M microscope (Zeiss, Munich, Germany).

Measurement of PGRN Protein Levels by ELISA in Mice

Progranulin protein levels in CSF and spinal cord lysates from mice were measured using a mouse progranulin ELISA kit (AdipoGen, Seoul, Korea). Cerebrospinal fluid samples were obtained by suboccipital puncture in mice anesthetized using intraperitoneal pentobarbital sodium. Spinal cord samples were lysed in RIPA buffer (50 mmol/L Tris, 150 mmol/L NaCl, 0.1% SDS, 1% Triton X-100, 0.5% sodium deoxycholate, and a Complete protease inhibitor cocktail tablet; Roche, Basel, Switzerland); PGRN content was measured according to the manufacturer's manual. The protein concentration was measured using the BCA protein assay (Pierce, Rockford, IL), and values were normalized to the protein content in the sample.

Statistics

Data are shown as mean ± SEM. Student t-tests or Mann-Whitney U test were used to calculate significance. When more than 2 groups were compared, a one-way analysis of variance with Fisher least significant difference post hoc test was used.

Results

Progranulin Protein in Human CSF and Plasma

Samples of ALS, FTLD, AD, and control CSF were obtained in the diagnostic phase, when the patients were first presented. Mutations in the PGRN gene were excluded. Patient characteristics are shown in Table 1. The measured levels did not depend on age or sex of the patients or on the duration of sample storage before analysis (data not shown). No differences were noted between ALS patients and controls (Table 1). Patients with the combination of ALS and FTLD, with FTLD alone, or with AD had normal CSF PGRN levels (Table 1). No correlation between the disease duration at the time of lumbar puncture and PGRN levels was observed.

The PGRN plasma levels in ALS patients (n = 91) were not different from those in controls (n = 56, p = 0.4). The mean was 261 ± 25 and 231 ± 24 ng/mL in ALS patients and controls, respectively. In 1 patient without a family history of ALS, an R110Q mutation was found. The plasma level in this patient was within the reference range (200 ng/mL). Of the 91 patients, 7 patients carried the GA instead of GG genotype at rs9897526, which was previously shown to reduce the age of onset of ALS by 7.7 years and to reduce survival (29). The mean PGRN plasma levels were slightly lower in GA carriers (266 ± 27 for GG and 202 ± 36 ng/mL for GA, p = 0.5), but these differences were not significant. Altogether, these results suggest that reductions in PGRN levels are restricted to patients carrying PGRN loss-of-function mutations.

Spinal PGRN Levels Increase With Motor Neuron Degeneration

Progranulin mRNA levels were measured by quantitative PCR in spinal cord lysates from presymptomatic (50 days) and late symptomatic (130 days) SOD1G93A mice. We also included young (50 days) and aged (1 year) nontransgenic animals. There was a slight increase in PGRN mRNA levels with aging in nontransgenic animals, but significance was not reached (Fig. 1A). Presymptomatic animals had normal PGRN mRNA levels, whereas in late symptomatic mutant SOD1G93A mice, there was a distinct upregulation of PGRN mRNA levels (Fig. 1A), that is, a 261% increase in PGRN transcripts compared with presymptomatic animals.

FIGURE 1

Progranulin (PGRN) mRNA and protein levels increase with motor neuron degeneration in mutant SOD1G93A mice. (A) Spinal cord PGRN mRNA levels measured by quantitative PCR and normalized to presymptomatic stage and β-actin levels in nontransgenic (NTG) and mutant SOD1G93A mice (analysis of variance, F value = 6.5; overall, p = 0.002). (B) Spinal cord PGRN protein levels measured by ELISA normalized to protein content (analysis of variance, F value = 4.9; overall, p = 0.002).

FIGURE 1

Progranulin (PGRN) mRNA and protein levels increase with motor neuron degeneration in mutant SOD1G93A mice. (A) Spinal cord PGRN mRNA levels measured by quantitative PCR and normalized to presymptomatic stage and β-actin levels in nontransgenic (NTG) and mutant SOD1G93A mice (analysis of variance, F value = 6.5; overall, p = 0.002). (B) Spinal cord PGRN protein levels measured by ELISA normalized to protein content (analysis of variance, F value = 4.9; overall, p = 0.002).

Full-length PGRN protein levels were measured by ELISA in spinal cord lysates from mutant SOD1G93A mice and controls. Similar PGRN levels were observed in young and old nontransgenic animals and presymptomatic mutant SOD1G93A mice (Fig. 1B). In contrast, PGRN levels were significantly higher in the diseased mutant SOD1G93A spinal cord (Fig. 1B). Compared with presymptomatic animals, there was a 33% increase in PGRN protein levels. In comparison to changes in mRNA levels, the increase in full-length PGRN protein level was smaller, possibly because a considerable amount of PGRN is secreted and cleaved into smaller fragments.

Microglial Upregulation of PGRN

In the healthy mouse spinal cord, PGRN expression was observed in neurons in the ventral and dorsal horn, as shown by NeuN double staining (Figure, Supplemental Digital Content 1, ). Large caliber SMI-32 (neurofilament heavy)+ motor neurons were also PGRN+ (Figs. 2A–C). In presymptomatic mutant SOD1G93A mice, (before the massive expansion of the CD11b+ microglial population), large motor neurons were also PGRN+; resting CD11b+ microglia did not express strong PGRN staining (Figs. 2D–F). However, in symptomatic stages, numerous intensely PGRN+ microglia were observed in the ventral horn of the spinal cord (Figs. 2G–L), suggesting that the microglial response is responsible for the upregulation of PGRN. In the diseased spinal cord, PGRN did not only colocalize with the microglial marker CD11b (Figure, Supplemental Digital Content 2, parts A–C, ), but also with CD11c, a dendritic cell marker expressed by mature microglia (Figure, Supplemental Digital Content 2, parts D–F, ). The surviving motor neurons in spinal cords of symptomatic mice remained PGRN+ (Figs. 2M–O). Progranulin immunoreactivity was not seen in reactive astrocytes or in NG2+ glial precursor cells (Figure, Supplemental Digital Content 3, ).

FIGURE 2

Proganulin (PGRN) expression in the mouse spinal cord. (A-C) PGRN+ motor neurons in the ventral horn of the spinal cord. Progranulin staining (A), SMI32 staining (B), and overlay (C) of ventral part of spinal cord from nontransgenic mouse; inset shows detail of motor neuron. (D-L) Proganulin staining of the ventral part of the spinal cord in presymptomatic, early, and late symptomatic mutant SOD1G93A mice (in red, D, G, J), CD11b staining (in green, E, H, K), and overlay (F, I, L); 2 motor neurons are marked by white arrows in D. (D-L) There is a gradual increase in the number of PGRN+ microglia in the spinal cord of mutant SOD1G93A mice. (M-O) At the end stage, surviving cells with motor neuron morphology (CD11b-negative, white arrows) remained PGRN+ in mutant SOD1G93A mice. Scale bar = 50 μm.

FIGURE 2

Proganulin (PGRN) expression in the mouse spinal cord. (A-C) PGRN+ motor neurons in the ventral horn of the spinal cord. Progranulin staining (A), SMI32 staining (B), and overlay (C) of ventral part of spinal cord from nontransgenic mouse; inset shows detail of motor neuron. (D-L) Proganulin staining of the ventral part of the spinal cord in presymptomatic, early, and late symptomatic mutant SOD1G93A mice (in red, D, G, J), CD11b staining (in green, E, H, K), and overlay (F, I, L); 2 motor neurons are marked by white arrows in D. (D-L) There is a gradual increase in the number of PGRN+ microglia in the spinal cord of mutant SOD1G93A mice. (M-O) At the end stage, surviving cells with motor neuron morphology (CD11b-negative, white arrows) remained PGRN+ in mutant SOD1G93A mice. Scale bar = 50 μm.

To confirm the microglial upregulation of PGRN with disease progression, we performed quantitative PCR on freshly isolated spinal cord microglial cells from presymptomatic and late symptomatic mutant SOD1G93A mice. Microglial cells isolated from diseased mice had a much higher PGRN expression (Fig. 3).

FIGURE 3

Microglial upregulation of progranulin (PRGN). (A) Microglial cells were isolated from adult spinal cord of SOD1G93A mice by loading the cell suspension on a 40% to 80% Percoll gradient. Viable mononuclear cells were gated (P1) and cells immunoreactive for CD11b-phycoerythrin (PE) were collected for further analysis (49). (B) Progranulin mRNA expression measured by quantitative PCR in freshly isolated spinal microglia from presymptomatic and symptomatic mutant SOD1G93A mice (*, p < 0.05).

FIGURE 3

Microglial upregulation of progranulin (PRGN). (A) Microglial cells were isolated from adult spinal cord of SOD1G93A mice by loading the cell suspension on a 40% to 80% Percoll gradient. Viable mononuclear cells were gated (P1) and cells immunoreactive for CD11b-phycoerythrin (PE) were collected for further analysis (49). (B) Progranulin mRNA expression measured by quantitative PCR in freshly isolated spinal microglia from presymptomatic and symptomatic mutant SOD1G93A mice (*, p < 0.05).

We also analyzed 2 less frequently studied animal models in which motor neuron degeneration occurs, i.e. homozygous transgenic mice with a deletion in the hypoxia-response element in VEGFδ/δ (45), and homozygous trans-genic mice overexpressing human mutant tauP301L driven by the mouse Thy1 gene promoter (46). When these mice developed signs of motor neuron degeneration, a similar increase in PGRN levels was observed in the spinal cord at the mRNA and protein levels (Fig. 4). Immunostaining confirmed that the motor neuron loss was also characterized by PGRN staining in microglia (results not shown).

FIGURE 4

Progranulin (PGRN) mRNA and protein levels increase with motor neuron degeneration in VEGFδ/δ and mutant tauP301L mice. (A) Spinal cord PGRN mRNA levels measured by quantitative PCR and normalized to β-actin levels in presymptomatic and symptomatic VEGFδ/δ and mutant TauP301L mice. (B) Spinal cord PGRN protein levels measured by ELISA normalized to protein content.

FIGURE 4

Progranulin (PGRN) mRNA and protein levels increase with motor neuron degeneration in VEGFδ/δ and mutant tauP301L mice. (A) Spinal cord PGRN mRNA levels measured by quantitative PCR and normalized to β-actin levels in presymptomatic and symptomatic VEGFδ/δ and mutant TauP301L mice. (B) Spinal cord PGRN protein levels measured by ELISA normalized to protein content.

In control human spinal cords, PGRN immunoreactivity was seen in several cell types, including large motor neurons in the ventral horn (Fig. 5A). Compared with control spinal cord without microgliosis (Figs. 5B, C), the spinal cord of ALS patients showed a pronounced microglial response, consisting of CD68+ microglia. These microglia were intensely PGRN+ (Figs. 5D–F), as described in a recent study (50). Surviving cells with neuronal morphology in the ventral horn of the spinal cord remained PGRN+ (Figs. 5G–I).

FIGURE 5

Motor neurons and reactive microglia express progranulin (PGRN) in the human spinal cord. (A–I) Progranulin staining (red, A, D, I) and CD68 staining (green, B, E, I) of control (A–C) and amyotrophic lateral sclerosis (ALS) (D–F, I) spinal cords. Surviving cells with motor neuron morphology at a higher magnification in the ventral horn of ALS spinal cord on H&E staining remained positive for PGRN (G and H). Fluorescent staining of a CD68-negative PGRN+ cell with motor neuron morphology, surrounded by 2 microglial cells (white arrows) is shown in panel I. Scale bar = 50 μm.

FIGURE 5

Motor neurons and reactive microglia express progranulin (PGRN) in the human spinal cord. (A–I) Progranulin staining (red, A, D, I) and CD68 staining (green, B, E, I) of control (A–C) and amyotrophic lateral sclerosis (ALS) (D–F, I) spinal cords. Surviving cells with motor neuron morphology at a higher magnification in the ventral horn of ALS spinal cord on H&E staining remained positive for PGRN (G and H). Fluorescent staining of a CD68-negative PGRN+ cell with motor neuron morphology, surrounded by 2 microglial cells (white arrows) is shown in panel I. Scale bar = 50 μm.

Because PGRN is a secreted protein, we also assessed PGRN levels in the CSF of mutant SOD1G93A mice. Small volumes of CSF (~15 μL) were obtained by suboccipital puncture in anesthetized mice, and the samples were analyzed by ELISA. In presymptomatic mutant SOD1G93A mice, PGRN levels were equal to nontransgenic controls. In late symptomatic mice, there were elevated PGRN CSF levels (Fig. 6). Interestingly, early symptomatic mice (around the age of 80 days, when muscle denervation is already present [51, 52] and when the microglial response has already been initiated [53]) did not yet display the elevated PGRN levels in CSF. These data are in accordance with the human findings in CSF and suggest that microglial PGRN is secreted and accumulates in CSF with disease progression.

FIGURE 6

Increased progranulin (PGRN) in CSF with disease progression in mutant SOD1G93A mice. Cerebrospinal fluid PGRN levels measured by ELISA in nontransgenic and mutant SOD1G93A mice (analysis of variance, F value = 15.8; overall, p = 0.000001; ***, p < 0.0001 compared with other groups).

FIGURE 6

Increased progranulin (PGRN) in CSF with disease progression in mutant SOD1G93A mice. Cerebrospinal fluid PGRN levels measured by ELISA in nontransgenic and mutant SOD1G93A mice (analysis of variance, F value = 15.8; overall, p = 0.000001; ***, p < 0.0001 compared with other groups).

Discussion

Interest in the role of progranulin in the CNS was raised by the discovery of disease-causing mutations in the pro-granulin gene in patients with FTLD (10, 11). Most mutations result in reduced PGRN levels, and shortage of PGRN is thought to underlie neuronal death. In addition, missense mutations were also observed in some sporadic FTLD patients (25, 26), in AD patients (27), and in some patients with ALS (28–30). Progranulin has been identified as a genetic modifier of the phenotype of motor neuron degeneration (29), but mutations segregating with the disease have not been observed in classic ALS families to date. How the sequence variants found in individual patients contribute to the disease pathogenesis is unclear at present.

We found that PGRN levels are normal in plasma and CSF from newly diagnosed ALS patients, but expression increases with disease progression in the spinal cord of trans-genic animal models experiencing motor neuron degeneration, mainly due to an elevated expression in microglia. This increase was detectable in CSF, suggesting that PGRN may become a disease state marker of motor neuron degeneration.

It has been shown by several groups that FTLD patients with null mutations have reduced PGRN protein levels in their blood and CSF (18–22). Given the potential pathogenic role of PGRN in ALS, we measured PGRN levels in plasma and CSF from ALS patients. The normal PGRN values in ALS patients, as well as in FTLD patients and AD patients without PGRN mutations, suggest that shortage of PGRN is restricted to neurodegeneration caused by PGRN null mutations. Interestingly, the SNP rs9897526 associated with more severe disease in ALS patients (29) resulted in slightly lower PGRN plasma levels. Significance was not reached because of the small sample size of patients with the at-risk genotype in this study. Larger studies are required to study the effect of the rs9897526 SNP on PGRN protein levels.

In some patients with FTLD or AD and missense mutations of unknown pathogenic nature, PGRN blood levels are slightly reduced, whereas other mutations do not affect the PGRN levels (18, 21, 22). Here we report normal PGRN plasma levels in an ALS patient with an R110Q mutation. Recently, another missense mutation (C521Y) with proven segregation in a family with progressive nonfluent aphasia was found to leave PGRN plasma levels untouched (54). How missense mutations with normal protein expression can cause disease is currently unknown. Shortage of PGRN cannot be excluded at present because changes in protein processing or function may still result in loss of function.

In mouse models of motor neuron degeneration, we observed upregulation of PGRN in the spinal cord, mainly because of the increased expression in microglial cells. Strongly PGRN+ microglia were also observed in the ventral horn of the spinal cord in ALS patients. Upregulation of PGRN transcripts after acute or chronic damage to the CNS has been observed in previous mRNA expression studies. Progranulin mRNA was upregulated in mouse brain lysates after Sindbis infection (32), in microglial cells isolated from Creutzfeldt-Jakob disease inoculated mouse brains (36), in mouse brain from a transgenic model of mucopolysacchari-dosis I and IIIB (37), in amyloid plaques (39), and in human lumbar spinal cord from ALS patients (38).

The strong expression of PGRN in microglia has been noted not only in ALS (50) but also around neuritic plaques in AD patients (10, 39). Interestingly, also in patients with FTLD caused by loss-of-function mutations in the PGRN gene, intense PGRN staining was observed in activated microglial cells (34, 35). Microglial upregulation occurs despite reduced blood, CSF, and brain levels of PGRN in these patients (18–22), suggesting that the healthy allele can compensate for the diseased allele carrying a null mutation in activated microglial cells. Upregulation of progranulin in microglial cells seems to be a common response to injury in the CNS. Progranulin produced by microglia and released in the extracellular milieu may be one of the factors derived from micro-glia that can protect neurons from degeneration and may be important in microglial proliferation. Further research is required to unravel the functions PGRN in microglia and to clarify the effects of secreted PGRN by microglial cells on the microenvironment.

In the adult CNS, PGRN expression is limited to some neuronal populations, such as cortical and hippocampal pyramidal neurons and Purkinje cells (55). Here, we confirm that motor neurons in the spinal cord also express PGRN (56). Although downregulation of PGRN in spinal motor neurons has been observed after axotomy (57), motor neurons in the diseased spinal cord in mouse models of motor neuron degeneration or ALS patients remained PGRN+. The functions of PGRN in motor neurons require further elucidation. Despite advances in the search for biomarkers in ALS, markers that are sensitive to presymptomatic diagnosis or to the disease progression of motor neuron degeneration are still lacking (58). In view of microglial upregulation of PGRN (i.e. a secreted protein associated with disease progression), we evaluated the potential of PGRN as a diagnostic bio-marker. No differences in PGRN serum or CSF levels were found between ALS patients or controls, indicating that PGRN ELISA on CSF is not a useful diagnostic biomarker for ALS. Normal CSF levels of PGRN were found in early symptomatic mutant SOD1G93A mice, but there was a distinct increase in CSF PGRN with disease progression. The CSF samples from ALS patients in our study obtained during the early diagnostic phase of the disease were not different from controls. In view of the variable disease course of ALS patients, however, serial CSF sampling in ALS patients with disease progression into later disease stages would be necessary to confirm a disease state-dependent upregulation of PGRN in ALS. Cerebrospinal fluid PGRN levels may reflect the degree of microgliosis and thus might be an independent disease state marker. Owing to improved symptomatic care of ALS patients, the survival of patients in advanced stages of the disease is improving, without affecting the decline in functional measures in the earlier stages of the disease (59). Therefore, markers that reflect the biologic disease state in more advanced stages would be of additional value as a surrogate marker of disease severity in clinical studies.

ACKNOWLEDGMENTS

The authors thank the Cell Imaging Core K.U. Leuven for use of a confocal microscope.

REFERENCES

1
Bateman
A
Belcourt
D
Bennett
H
et al
.
Granulins, a novel class of pep-tide from leukocytes
.
Biochem Biophys Res Commun
 
1990
;
173
:
1161
68
2
Bhandari
V
Palfree
RG
Bateman
A
.
Isolation and sequence of the granulin precursor cDNA from human bone marrow reveals tandem cysteine-rich granulin domains
.
Proc Natl Acad Sci U S A
 
1992
;
89
:
1715
19
3
Shoyab
M
McDonald
VL
Byles
C
et al
.
Epithelins 1 and 2: Isolation and characterization of two cysteine-rich growth-modulating proteins
.
Proc Natl Acad Sci U S A
 
1990
;
87
:
7912
16
4
Zhu
J
Nathan
C
Jin
W
et al
.
Conversion of proepithelin to epithelins: Roles of SLPI and elastase in host defense and wound repair
.
Cell
 
2002
;
111
:
867
78
5
He
Z
Ong
CH
Halper
J
et al
.
Progranulin is a mediator of the wound response
.
Nat Med
 
2003
;
9
:
225
29
6
He
Z
Bateman
A
.
Progranulin (granulin-epithelin precursor, PC-cell-derived growth factor, acrogranin) mediates tissue repair and tumori-genesis
.
J Mol Med
 
2003
;
81
:
600
12
7
Ahmed
Z
Mackenzie
IR
Hutton
ML
et al
.
Progranulin in frontotemporal lobar degeneration and neuroinflammation
.
J Neuroinflammation
 
2007
;
4
:
7
8
Zanocco-Marani
T
Bateman
A
Romano
G
et al
.
Biological activities and signaling pathways of the granulin/epithelin precursor
.
Cancer Res
 
1999
;
59
:
5331
40
9
Zhou
J
Gao
G
Crabb
JW
et al
.
Purification of an autocrine growth factor homologous with mouse epithelin precursor from a highly tumorigenic cell line
.
J Biol Chem
 
1993
;
268
:
10863
69
10
Baker
M
Mackenzie
IR
Pickering-Brown
SM
et al
.
Mutations in progranulin cause tau-negative frontotemporal dementia linked to chromosome 17
.
Nature
 
2006
;
442
:
916
19
11
Cruts
M
Gijselinck
I
van
der
Zee
J
et al
.
Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to chromosome 17q21
.
Nature
 
2006
;
442
:
920
24
12
Benussi
L
Ghidoni
R
Pegoiani
E
et al
.
Progranulin Leu271LeufsX10 is one of the most common FTLD and CBS associated mutations worldwide
.
Neurobiol Dis
 
2009
;
33
:
379
85
13
Lopez de Munain
A
Alzualde
A
Gorostidi
A
et al
.
Mutations in pro-granulin gene: Clinical, pathological, and ribonucleic acid expression findings
.
Biol Psychiatry
 
2008
;
63
:
946
52
14
Neary
D
Snowden
JS
Mann
DM
.
Cognitive change in motor neurone disease/amyotrophic lateral sclerosis (MND/ALS)
.
J Neurol Sci
 
2000
;
180
:
15
20
15
Lomen-Hoerth
C
.
Characterization of amyotrophic lateral sclerosis and frontotemporal dementia
.
Dement Geriatr Cogn Disord
 
2004
;
17
:
337
41
16
Lomen-Hoerth
C
Anderson
T
Miller
B
.
The overlap of amyotrophic lateral sclerosis and frontotemporal dementia
.
Neurology
 
2002
;
59
:
1077
79
17
Abrahams
S
Goldstein
LH
Simmons
A
et al
.
Word retrieval in amyo-trophic lateral sclerosis: A functional magnetic resonance imaging study
.
Brain
 
2004
;
127
:
1507
17
18
Mukherjee
O
Wang
J
Gitcho
M
et al
.
Molecular characterization of novel progranulin (GRN) mutations in frontotemporal dementia
.
Hum Mutat
 
2008
;
29
:
512
21
19
Van Damme
P
Van Hoecke
A
Lambrechts
D
et al
.
Progranulin functions as a neurotrophic factor to regulate neurite outgrowth and enhance neuronal survival
.
J Cell Biol
 
2008
;
181
:
37
41
20
Ghidoni
R
Benussi
L
Glionna
M
et al
.
Low plasma progranulin levels predict progranulin mutations in frontotemporal lobar degeneration
.
Neurology
 
2008
;
71
:
1235
39
21
Finch
N
Baker
M
Crook
R
et al
.
Plasma progranulin levels predict pro-granulin mutation status in frontotemporal dementia patients and asymptomatic family members
.
Brain
 
2009
;
132
:
583
91
22
Sleegers
K
Brouwers
N
Van Damme
P
et al
.
Serum biomarker for progranulin-associated frontotemporal lobar degeneration
.
Ann Neurol
 
2009
;
65
:
603
9
23
Carecchio
M
Fenoglio
C
De Riz
M
et al
.
Progranulin plasma levels as potential biomarker for the identification of GRN deletion carriers. A case with atypical onset as clinical amnestic Mild Cognitive Impairment converted to Alzheimer's disease
.
J Neurol Sci
 
2009
;
287
:
291
93
24
Cruchaga
C
Fernandez-Seara
MA
Seijo-Martinez
M
et al
.
Cortical atrophy and language network reorganization associated with a novel progranulin mutation
.
Cereb Cortex
 
2009
;
19
:
1751
60
25
van der Zee
J
Le Ber
I
Maurer-Stroh
S
et al
.
Mutations other than null mutations producing a pathogenic loss of progranulin in frontotemporal dementia
.
Hum Mutat
 
2007
;
28
:
416
26
Spina
S
Murrell
JR
Huey
ED
et al
.
Clinicopathologic features of frontotemporal dementia with progranulin sequence variation
.
Neurology
 
2007
;
68
:
820
27
27
Brouwers
N
Sleegers
K
Engelborghs
S
et al
.
Genetic variability in progranulin contributes to risk for clinically diagnosed Alzheimer disease
.
Neurology
 
2008
;
71
:
656
64
28
Schymick
JC
Yang
Y
Andersen
PM
et al
.
Progranulin mutations and amyotrophic lateral sclerosis or amyotrophic lateral sclerosi-frontotemporal dementia phenotypes
.
J Neurol Neurosurg Psychiatry
 
2007
;
78
:
754
56
29
Sleegers
K
Brouwers
N
Maurer-Stroh
S
et al
.
Progranulin genetic variability contributes to amyotrophic lateral sclerosis
.
Neurology
 
2008
;
71
:
253
59
30
Del Bo
R
Corti
S
Santoro
D
et al
.
No major progranulin genetic variability contribution to disease etiopathogenesis in an ALS Italian cohort
.
Neurobiol Aging
  July 24, 2009 [Epub ahead of print]
31
Matzilevich
DA
Rall
JM
Moore
AN
et al
.
High-density microarray analysis of hippocampal gene expression following experimental brain injury
.
J Neurosci Res
 
2002
;
67
:
646
63
32
Johnston
C
Jiang
W
Chu
T
et al
.
Identification of genes involved in the host response to neurovirulent alphavirus infection
.
J Virol
 
2001
;
75
:
10431
45
33
Naphade
SB
Kigerl
KA
Jakeman
LB
et al
.
Progranulin expression is upregulated after spinal contusion in mice
.
Acta Neuropathol
 
2009
;
119
:
123
33
34
Mackenzie
IR
Baker
M
Pickering-Brown
S
et al
.
The neuropathology of frontotemporal lobar degeneration caused by mutations in the pro-granulin gene
.
Brain
 
2006
;
129
:
3081
90
35
Mukherjee
O
Pastor
P
Cairns
NJ
et al
.
HDDD2 is a familial fronto-temporal lobar degeneration with ubiquitin-positive, tau-negative inclusions caused by a missense mutation in the signal peptide of progranulin
.
Ann Neurol
 
2006
;
60
:
314
22
36
Baker
CA
Manuelidis
L
.
Unique inflammatory RNA profiles of micro-glia in Creutzfeldt-Jakob disease
.
Proc Natl Acad Sci U S A
 
2003
;
100
:
675
79
37
Ohmi
K
Greenberg
DS
Rajavel
KS
et al
.
Activated microglia in cortex of mouse models of mucopolysaccharidoses I and IIIB
.
Proc Natl Acad Sci U S A
 
2003
;
100
:
1902
7
38
Malaspina
A
Kaushik
N
de Belleroche
J
.
Differential expression of 14 genes in amyotrophic lateral sclerosis spinal cord detected using gridded cDNA arrays
.
J Neurochem
 
2001
;
77
:
132
45
39
Pereson
S
Wils
H
Kleinberger
G
et al
.
Progranulin expression correlates with dense-core amyloid plaque burden in Alzheimer disease mouse models
.
J Pathol
 
2009
;
219
:
173
81
40
Chen-Plotkin
AS
Xiao
J
Geser
F
et al
.
Brain progranulin expression in GRN-associated frontotemporal lobar degeneration
.
Acta Neuropathol
 
2010
;
119
:
111
22
41
Brooks
BR
Miller
RG
Swash
M
et al
.
El Escorial revisited: Revised criteria for the diagnosis of amyotrophic lateral sclerosis
.
Amyotroph Lateral Scler Other Motor Neuron Disord
 
2000
;
1
:
293
99
42
Neary
D
Snowden
JS
Gustafson
L
et al
.
Frontotemporal lobar degeneration: A consensus on clinical diagnostic criteria
.
Neurology
 
1998
;
51
:
1546
54
43
Dubois
B
Feldman
HH
Jacova
C
et al
.
Research criteria for the diagnosis of Alzheimer's disease: Revising the NINCDS-ADRDA criteria
.
Lancet Neurol
 
2007
;
6
:
734
46
44
Wynendaele
W
Derua
R
Hoylaerts
MF
et al
.
Vascular endothelial growth factor measured in platelet poor plasma allows optimal separation between cancer patients and volunteers: A key to study an angiogenic marker in vivo?
Ann Oncol
 
1999
;
10
:
965
71
45
Oosthuyse
B
Moons
L
Storkebaum
E
et al
.
Deletion of the hypoxia-response element in the vascular endothelial growth factor promoter causes motor neuron degeneration
.
Nat Genet
 
2001
;
28
:
131
38
46
Terwel
D
Lasrado
R
Snauwaert
J
et al
.
Changed conformation of mutant Tau-P301L underlies the moribund tauopathy, absent in progressive, nonlethal axonopathy of Tau-4R/2N transgenic mice
.
J Biol Chem
 
2005
;
280
:
3963
73
47
Gurney
ME
Pu
H
Chiu
AY
et al
.
Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation
.
Science
 
1994
;
264
:
1772
75
48
Terwel
D
Muyllaert
D
Dewachter
I
et al
.
Amyloid activates GSK-3beta to aggravate neuronal tauopathy in bigenic mice
.
Am J Pathol
 
2008
;
172
:
786
98
49
Gowing
G
Philips
T
Van Wijmeersch
B
et al
.
Ablation of proliferating microglia does not affect motor neuron degeneration in amyotrophic lateral sclerosis caused by mutant superoxide dismutase
.
J Neurosci
 
2008
;
28
:
10234
44
50
Irwin
D
Lippa
CF
Rosso
A
.
Progranulin (PGRN) expression in ALS: An immunohistochemical study
.
J Neurol Sci
 
2009
;
276
:
9
13
51
Kennel
PF
Finiels
F
Revah
F
et al
.
Neuromuscular function impairment is not caused by motor neurone loss in FALS mice: An electromyo-graphic study
.
Neuroreport
 
1996
;
7
:
1427
31
52
Fischer
LR
Culver
DG
Tennant
P
et al
.
Amyotrophic lateral sclerosis is a distal axonopathy: Evidence in mice and man
.
Exp Neurol
 
2004
;
185
:
232
40
53
Alexianu
ME
Kozovska
M
Appel
SH
.
Immune reactivity in a mouse model of familial ALS correlates with disease progression
.
Neurology
 
2001
;
57
:
1282
89
54
Wang
J
Van Damme
P
Cruchaga
C
et al
.
Pathogenic cysteine mutations affect progranulin function and production of mature granulins
.
J Neu-rochem
 
2010
;
112
:
1305
15
55
Daniel
R
He
Z
Carmichael
KP
et al
.
Cellular localization of gene expression for progranulin
.
J Histochem Cytochem
 
2000
;
48
:
999
1009
56
Ryan
CL
Baranowski
DC
Chitramuthu
BP
et al
.
Progranulin is expressed within motor neurons and promotes neuronal cell survival
.
BMC Neurosci
 
2009
;
10
:
130
57
Moisse
K
Volkening
K
Leystra-Lantz
C
et al
.
Divergent patterns of cytosolic TDP-43 and neuronal progranulin expression following axot-omy: Implications for TDP-43 in the physiological response to neuronal injury
.
Brain Res
 
2009
;
1249
:
202
11
58
Turner
MR
Kiernan
MC
Leigh
PN
et al
.
Biomarkers in amyotrophic lateral sclerosis
.
Lancet Neurol
 
2009
;
8
:
94
109
59
Qureshi
M
Schoenfeld
DA
Paliwal
Y
et al
.
The natural history of ALS is changing: Improved survival
.
Amyotroph Lateral Scler
 
2009
;
10
:
324
31

Supporting Information

Author notes

Philips and Muynck contributed equally to this work.
This work was supported by grants from the Fund for Scientific Research Flanders (FWO-V 1.5.081.08), the Interuniversity Attraction Poles-program P6/43 of the Belgian Federal Science Policy Office. P.V.D. is supported by the FWO-V and by the Stichting voor Alzheimer Onderzoek-Belgium (SAO-FRMA no. 08615). P.V.D. holds a clinical investigatorship of the FWO-V and K.S. holds a postdoctoral fellowship of the FWO-V. T.P., L.D.M., and J.D. hold a PhD grant of the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT- Vlaanderen).
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.jneuropath.com).